BITZ0

A futures bet is the least liquid thing on a sportsbook. You commit money in September and find out in February, with no realistic way to change your mind in between.
That illiquid stake is the defining feature, and it is worth understanding before the appeal of a long price takes over.
That Window Has Already Closed
Worth saying plainly up front, since the honest version is more useful than the alternative.
The 2026 regular season began on 9 September, which means pre-Week One prices are gone. Every futures market has already absorbed a round of results and repriced accordingly.
What follows is therefore two things at once: a record of what that window offered, so you know what to look for next August, and a guide to the in-season futures market that replaced it. Both are live questions, since futures trade all the way to February.
Four Futures Markets, Ranked by Tractability
They are not equally difficult, and the order below runs from hardest to most approachable.
Outright Winner
The outright winner market is the longest-dated on the board and the one with the widest prices.
Appeal is obvious: a longshot at a big number, held through a season. The cost is equally obvious once stated. Your stake is committed for five months, and a season-ending injury to one player ends the bet with no refund mechanism of any kind.
Thirty-two teams start and one wins. That is the market in a sentence, and it explains why the prices look generous.
Conference and Division Winners
Shorter-dated, smaller fields, fewer variables.
A division winner market prices four teams against each other instead of 32, which makes it considerably more tractable for anyone with a genuine read on a specific division. Conference winner sits between the two in difficulty.
These also resolve earlier than the outright, so the capital comes back sooner if the bet loses.
Season Win Totals
The most analytically approachable futures market on the board, and the one that gets least attention.
A win total prices one team against a number instead of against a field. Will this team win more or fewer than nine and a half games? That is a question about schedule strength, roster continuity and injury luck for a single organisation, which is a far smaller problem than modelling an entire conference.
The over and under are usually priced close to even money, so the margin is thinner than on the longer-shot markets, and the outcome depends on a season's accumulation, not a single result.
Award Markets
MVP, Offensive Player of the Year and similar. The widest margins of the four.
These concentrate on a handful of plausible candidates, which sounds tractable and is not, because award voting incorporates narrative and team success alongside individual production. They are also priced by books that know the market is driven by sentiment more than analysis.
Interesting to hold, expensive to buy.
The Liquidity Problem Nobody Mentions
Here is the practical constraint that separates futures from every other bet you place.
A match bet resolves in three hours. A futures bet resolves in five months, and during that time your stake is unavailable. Most books do not offer cash out on futures markets, so there is generally no exit: the position runs to settlement whether your view has changed or not.
That means a futures stake should be sized as money you are content to have locked up until February. Not money you might want back in November. The bet is not liquid, and treating it as though it were is how people end up depositing again to cover positions they cannot unwind.
Three Drivers of Price Movement
Three drivers, and only one of them is results.
Results are the obvious one. A team winning changes its price, and the price compression accelerates as the field narrows.
Injury news moves futures markets more sharply than it moves single-game lines, because a quarterback lost in October affects every remaining game, not one. A futures price can move substantially on a Wednesday injury report with no football played.
Schedule strength matters as the season progresses. A team sitting at 6-2 with an easy remaining slate is priced differently from a 6-2 team facing four contenders, and that distinction is invisible in the record alone.
The same pattern appears in football generally, where early-season odds move for reasons that have nothing to do with the table.
Placing Them
Futures need a board that carries them properly, which not every crypto sportsbook does.
Dexsport publishes over 100 markets on major matches, and its event-tiered limits rise for major competitions, which matters for futures because they are typically capped lower than match markets.
Being non-custodial, settled funds return to a wallet you hold, and its $1 minimum makes it practical to take several small futures positions instead of one large one.
That last point is the sensible structure for this market: spreading a modest total across three or four positions is a better shape than concentrating on one longshot, given that most futures bets lose. Outright and long-shot markets work the same way in other sports.
What Is Still Open
The pre-season window is closed, and the in-season one is open.
Win totals are still live and still the most tractable market on the board
Division winners reprice weekly as records separate
The outright market compresses sharply once the playoff field is set in January
A position taken now is a position taken before the largest single repricing event of the cycle.
Confirm what is legal where you live, keep stakes within a set budget, and play only if you are of legal age, since KYC or AML checks may apply.
Responsible gambling has a specific angle here: a futures position running for months keeps a betting account live through a whole season, and a stake you cannot withdraw is a stake still in play.


Disclaimer: The information here is provided for general purposes only and is not legal, tax, investment, or financial advice, and nothing here is a betting tip or prediction. Market availability, cash-out policies and limits vary by operator and change, so read the current rules before placing a futures bet. Betting carries risk, and rules vary by country, so check the law where you live. Please gamble responsibly, within your means, and only if you are of legal age.

Monday Night Football is the only NFL game in its window. Everything else has finished, every other result is settled, and one fixture carries the entire evening.
That isolation changes the market. It also means that for most of the world, the game kicks off in the middle of the night.
When It Actually Starts Where You Are
A standard 8:15pm Eastern kickoff, converted.
Region
Local kickoff
What that means
US Eastern
8:15pm Monday
Prime time
US Pacific
5:15pm Monday
Early evening
United Kingdom
1:15am Tuesday
Middle of the night
Central Europe
2:15am Tuesday
Middle of the night
Gulf states
4:15am Tuesday
Pre-dawn
India
5:45am Tuesday
Early morning
Australia Eastern
11:15am Tuesday
Late morning
One caveat on those figures: North America and Europe change their clocks on different dates in autumn, so the offset between them shifts for part of the season. A conversion that held in October may be an hour out in November.
Australia is the outlier in the other direction. A Monday night game in the United States is a Tuesday lunchtime game there, which is the most civilised viewing slot on the entire NFL calendar.
A Lone Fixture Prices Differently
The scheduling is only half the story. The market behaves unusually too.
Because Monday night stands alone, it attracts a concentration of money that a Sunday afternoon game never sees. Every bettor with an opinion has one game to express it on, and every book knows it.
Competition produces a sharply priced main market. The spread and total on Monday night are among the most heavily traded lines of the week, which pushes the margin down and makes the number close to fair. Competition does the work.
There is a second effect worth knowing. Monday night is the most common accumulator final leg built across the weekend, so a large volume of open tickets settle on that one game.
That is why books promote the fixture so heavily and why cash-out activity spikes during it: thousands of players are watching a multi-leg bet come down to one result.
The Small-Hours Problem
For anyone outside North America, the practical difficulty is not the market. It is the hour.
In-play betting on Monday Night Football from Europe means being awake and making decisions between one and five in the morning. Those are worse decisions than the ones you would make at seven in the evening, and no amount of discipline fully compensates for it.
The honest recommendation is therefore structural, not motivational: take your pre-match position before you go to bed. A pre-match bet placed on Monday evening in your own time zone is the same bet, at a comparable price, made when you are capable of thinking about it.
If you do want live exposure, decide the exit in advance. Cash Out on eligible bets lets you close a position at a stated price, and setting a level you would accept before kickoff is considerably easier than judging one at 3am.
Three Platform Requirements at This Hour
Three practical requirements for a fixture at this hour.
Mobile access without friction comes first. Dexsport runs entirely in the browser on iOS and Android with no app to download, which for a game you may be following from bed is more useful than it sounds: no install, no update prompt, no storage.
Second is Cash Out on eligible bets, so a position can be closed at a known price instead of held to a conclusion you may sleep through.
Third is a board deep enough to matter, since a single fixture is the whole of Monday's card. Dexsport publishes over 100 markets on major matches, and its event-tiered limits rise for major competitions.
Its casino and sportsbook run from one balance, which for anyone genuinely awake at 3am is either a convenience or a warning depending on temperament.
Market depth varies more than platform marketing suggests, and coverage and market count differ substantially between books on the same fixture.
Betting a Game You Will Not Watch Live
Plenty of people outside North America bet Monday night and watch the highlights on Tuesday. That is a reasonable approach, and it changes what you should bet.
Pre-match spreads and totals settle without you
Live markets do not work if you are asleep
Accumulator legs settle overnight either way
Cash Out is unavailable to a bettor who is unconscious when the price moves
So the sensible board for a sleeping bettor is the pre-match one, sized as a position you are content to leave alone. The alternative is setting an alarm for a game that finishes around 4am, which is a decision to make with clear eyes, not at kickoff.
Confirm what is legal where you live, keep stakes within a set budget, and play only if you are of legal age, since KYC or AML checks may apply.
Responsible gambling deserves specific attention on this fixture, because late-night betting on a lone game after a weekend of results is the situation most likely to produce a recovery bet.


Disclaimer: The information here is provided for general purposes only and is not legal, tax, investment, or financial advice, and nothing here is a betting tip or prediction. Kickoff times are typical and subject to change by the league, and clock changes vary by country, so confirm local timings before betting. Betting carries risk, and rules vary by country, so check the law where you live. Please gamble responsibly, within your means, and only if you are of legal age.

Football is scored in threes and sevens, and that single fact decides where the value sits on every spread you will ever bet.
If you already know how a spread and a total work, this is the layer underneath: which numbers matter, why half a point costs more in some places than others, and what the standard price is actually charging you.
Five Things the Key Numbers Tell You
NFL key numbers work like this, and once you know them the board stops looking like a list of arbitrary figures.
Three is the most common point spread margin in the sport. Roughly 15% of NFL games end with exactly three points separating the teams. One field goal. Nothing else on the board comes close, and it is nearly twice as frequent as the next number.
Seven is second, at around 9%. A touchdown with the extra point. After that the frequency drops away through 10, 6 and 14, all of them combinations of the same two scoring events.
The half point is called the hook, and it removes the push. A spread of -3 can land exactly on the number and return your stake. A spread of -2.5 cannot. That half point is the difference between a refund and a win on roughly one game in seven.
The hook is worth far more at some numbers than others. Moving a favourite from -3 to -2.5, or an underdog from +3 to +3.5, is genuinely valuable because it crosses the most common margin in football. Moving from -8 to -8.5 crosses nothing, and is worth almost nothing.
The standard price already charges you. Most spreads are posted at -110 on both sides, meaning you risk 110 to win 100. That -110 pricing carries the book's built-in margin, applied whether the line is good or bad, and it is calculable from the prices themselves.
Seven Is Slowly Getting Weaker
A nuance worth knowing, because it is a genuine trend, not a theory.
The NFL moved the extra point attempt back to the 15-yard line in 2015. Conversion stopped being a formality and teams started going for two more often.
From 2006 to 2014 teams kicked the extra point after 95.2% of touchdowns. From 2015 to 2023 that fell to 90.7%, with an all-time low of 88.3% in 2021. Fewer kicked extra points means fewer margins landing exactly on seven.
Three is unaffected, since a field goal is still three points and always will be. But the gradual erosion of seven means the hook around it is worth marginally less than it was a decade ago, and the hook around three is worth exactly what it always was.
Totals Have Key Numbers Too
Less discussed and built from the same arithmetic.
Combined scores cluster on numbers that combine touchdowns and field goals cleanly. 41, 43, 37, 44, 51, 33 and 47 come up disproportionately often, for the same reason margins cluster on 3 and 7.
The practical consequence mirrors the spread. A total moving from 43 to 43.5 crosses a common combined score and matters. A total moving from 45 to 45.5 crosses nothing much.
Weather is the other variable on totals specifically, and January football in an open stadium is a different proposition from September football indoors.
Two Minutes of Shopping Pays Here
Everything above points in one direction.
Two books rarely post identical NFL lines. One at -3 and another at -2.5 on the same game are offering materially different bets, and the difference is concentrated at exactly the numbers where games actually finish. Half a point elsewhere on the board is noise. Half a point across three is not.
Line shopping pays here more than in most sports. Two sportsbooks show different odds on the same match for structural reasons, and in the NFL those differences land precisely where the outcomes cluster.
What to Look For at a Crypto Sportsbook
Four things, and the first is the one people skip.
Whether the book posts the hook at all, since a platform offering only whole numbers on key spreads is pushing more of your bets than one that does not
Whether alternate lines are available, letting you buy or sell the half point at an adjusted price instead of accepting the posted number
How deep the board goes, since first-half and team-total markets have their own key numbers
What the limits look like on the fixtures you actually bet, because a headline maximum usually describes a marquee game
Dexsport publishes over 100 markets on major matches, which is the depth level where alternate spreads and first-half lines become available instead of just the headline number.
Its event-tiered limits rise for major competitions, and a $1 minimum makes it practical to take a position on several games at small size.
If you want the definitions instead of the pricing mechanics, the spread, moneyline and totals explainer covers that ground.
The Short Version
Three and seven are where NFL games finish, at roughly 15% and 9%
Half a point across those numbers is worth paying for
Half a point anywhere else usually is not
Everything else on a spread board is detail around that fact.
Confirm what is legal where you live, keep stakes within a set budget, and play only if you are of legal age, since KYC or AML checks may apply.
Responsible gambling applies to the standard -110 price as much as anything: a built-in margin on every bet means the total you stake matters more than which side of any single line you took.


Disclaimer: The information here is provided for general purposes only and is not legal, tax, investment, or financial advice, and nothing here is a betting tip or prediction. Frequency figures are historical and do not predict individual games. Odds, lines and market availability vary by operator and change constantly. Betting carries risk, and rules vary by country, so check the law where you live. Please gamble responsibly, within your means, and only if you are of legal age.

Bitmine owns 4.9% of the total ETH coin supply of 122.0 million
Bitmine is 98% of the way to the 'Alchemy of 5%' in just 15 months
ETH is the best performing macro asset in Q3 of 2026 to date, outperforming the S&P 500 by 5,866bp
Tom Lee to deliver the keynote at KBW on September 30, 2026
Bitmine was added to the Russell 1000 Large-cap index on June 26, 2026
Bitmine's Series A Preferred Stock is trading on the NYSE under the symbol BMNP
Bitmine has 5,067,309 staked ETH, representing $12.7 billion at $2,513 per ETH. MAVAN (Made in America VAlidator Network) is a premier Ethereum staking destination for BMNR and institutional investors
Bitmine owns $98 million of Eightco (NASDAQ: ORBS), now one of the only publicly listed equities in the world to provide investors indirect exposure to OpenAI
Bitmine Crypto + Total Cash Holdings & Marketable Securities + "Moonshots" total $15.8 billion, including 5.96 million ETH tokens, total cash & marketable securities of $549 million, and other crypto holdings
Bitmine remains supported by a premier group of institutional investors including ARK's Cathie Wood, MOZAYYX, Founders Fund, Bill Miller III, Pantera, Kraken, DCG, Galaxy Digital and personal investor Thomas "Tom" Lee to support Bitmine's goal of acquiring 5% of ETH
NORWALK, Conn., Sept. 14, 2026 /PRNewswire/ -- (NYSE: BMNR) Bitmine Immersion Technologies, Inc. ("Bitmine" or the "Company") a Bitcoin and Ethereum Network company with a focus on the accumulation of crypto for long term investment, today announced Bitmine crypto + total cash & marketable securities + "moonshots" holdings totaling $15.8 billion.
As of September 13, 2026 at 5:30pm ET, the Company's crypto holdings are comprised of 5,956,378 ETH at $2,513 per ETH (per Coinbase NASDAQ: COIN), 212 Bitcoin (BTC), $180 million stake in Beast Industries, $98 million stake in Eightco Holdings (NASDAQ: ORBS) ("moonshots") and total cash & marketable securities of $549 million. Bitmine's ETH holdings are 4.9% of the ETH supply (of 122.0 million ETH).
"The price ratio of ETH to BTC moved to the highest level since Jan 30th of this year and established a new uptrend, breaking the trendline in place since the COVID-19 highs. In our view, this reflects Ethereum's strengthened position as the settlement rails for Wall Street tokenization and the increased realization that Ethereum may play a critical central role in managing agentic-AI. What is impressive is this strength in the absolute price of ETH and ETH/BTC taking place while global equities remain rangebound due to war risks and rising yields." stated Thomas "Tom" Lee, Chairman of Bitmine.
Tom DeMark, founder of DeMark Analytics and a capital markets advisor to Bitmine is expecting ETH to make a sharp upward move in coming weeks. According to Tom DeMark, "In August, ETH moved sideways without a downside break and the 12-day metric expired, which implies a renewal of the upside move. We believe this further supports the continuation of the prior uptrend. We expect late August's sharp one-day rally was a likely preview of the pending advance."
"As we enter the final month of calendar Q3 2026, ETH is the best performing macro asset during the quarter, outperforming the S&P 500 by 5,866bp through last Friday. In fact, the top 3 performing assets since June 30th are ETH, SOL and BTC," stated Lee. "We believe this sets the stage for institutions to add to their crypto holdings given the substantial outperformance of crypto versus other macro assets in calendar Q3 so far."
"We believe there are multiple positive catalysts as we head into the final months of 2026," stated Lee. "These include the upcoming CLARITY Act vote scheduled in mid-September. Additionally, Korean investors have again started buying crypto and rotating away from AI stocks. The 4-year cycle is bottoming within the next few weeks in our view. And this sets the stage for what we expect to be sizable institutional participation in buying crypto in the final months of 2026, especially given the tailwinds of tokenization and Agentic-AI."
Tom Lee will also deliver the keynote at Korea Blockchain Week 2026 on September 30 at 11:20 a.m. at Walkerhill Hotels & Resorts in Seoul. The 25-minute keynote is part of Korea Blockchain Week, one of Asia's leading blockchain and digital asset conferences. Additional information is available on the Korea Blockchain Week website.
"This ETH/BTC ratio has moved up during crypto bull cycles, driven by increasing use of Ethereum relative to Bitcoin. These prior cycles were fueled by ICOs (2017-2018), NFTs (2020-2021), and stablecoins (2025). In this upcoming crypto cycle, we see the ETH/BTC ratio rising, driven by Wall Street tokenizing on the blockchain and by agentic-AI using blockchains," continued Lee.
"Over the past week, we acquired 27,180 ETH. Bitmine's track record of consistent buying of crypto is unmatched by any public company in the world. Bitmine has bought ETH each and every week since the inception of the ETH Treasury Strategy on June 30, 2025," stated Lee.
On July 16, 2026, Bitmine released the latest Chairman's Message (link here) for July 2026. The title of the Message is "ETH is the cure for the Uncanny Valley of Wealth."
Earlier in 2026, Bitmine launched MAVAN (the Made in America VAlidator Network), the institutional-grade staking platform. While MAVAN was originally developed to support Bitmine's own Ethereum treasury, MAVAN has expanded to serve institutional investors, custodians, and ecosystem partners seeking best-in-class staking infrastructure. A portion of Bitmine's ETH is already staked on the MAVAN platform.
As of September 7, 2026, Bitmine total staked ETH stands at 5,067,309 ($12.7 billion at $2,513 per ETH). "Bitmine has staked more ETH than other entities in the world. At scale (when Bitmine's ETH is fully staked by MAVAN and its staking partners), the projected ETH staking reward would be $392 million on an annualized basis (using 2.62% 7-day BMNR yield)," stated Lee.
"Annualized staking revenues are now projected at $334 million. And this 5.1 million ETH is 85% of the 5.96 million ETH held by Bitmine. Bitmine's own staking operations generated a 7-day yield of 2.62% (annualized)," continued Lee.
Bitmine is one of the most widely traded stocks in the US. According to data from Fundstrat, the stock has traded average daily dollar volume of $924 million (4-day average, as of September 11, 2026), ranking #98 in the US, behind Intuit Inc. (rank #97) and ahead of Verizon Communications (rank #99) among 5,704 US-listed stocks (statista.com and Fundstrat research).
Bitmine's crypto holdings reign as the #1 Ethereum treasury and #2 global treasury, behind Strategy Inc. (NASDAQ: MSTR), which reportedly owns 845,080 BTC valued at approximately $71 billion. Bitmine remains the largest ETH treasury in the world.
Bitmine management believes the GENIUS Act and the Securities and Exchange Commission's (SEC) Project Crypto are as transformational to financial services in 2026 as the US action on August 15, 1971, which ended the Bretton Woods system and took the U.S. dollar off the gold standard 55 years ago. This 1971 event was the catalyst for the modernization of Wall Street, creating the iconic Wall Street titans and financial and payment rails of today. These proved to be better investments than gold.
The Chairman's message can be found here:
https://www.Bitminetech.io/chairmans-message
The Fiscal Full Year 2025 Earnings presentation and corporate presentation can be found here: https://Bitminetech.io/investor-relations/
To stay informed, please sign up at: https://Bitminetech.io/contact-us/
About Bitmine
Bitmine Immersion Technologies, Inc. (NYSE: BMNR), together with its subsidiaries ("Bitmine" or the "Company"), is a blockchain technology infrastructure company operating across institutional digital asset staking and validation services, bitcoin mining, and strategic digital asset management. As the world's leading Ethereum Treasury company, it implements an innovative digital asset strategy for institutional investors and public market participants. The Company provides institutional-grade staking and validation infrastructure—through which it earns staking rewards and validation income—alongside bitcoin mining activities. Bitmine holds digital assets strategically, generating yield on those holdings to support liquidity and capital formation. Since 2025, the Company has expanded its blockchain infrastructure capabilities, including developing and deploying MAVAN, its institutional staking and validation platform. The Company's activities further include investments in early-stage blockchain opportunities ("moonshot" investments) and ancillary mining, hosting, and consulting services.
For additional details, follow on X:
https://x.com/bitmnr
https://x.com/fundstrat
Forward Looking Statements
This press release contains statements that constitute "forward-looking statements" within the meaning of the Private Securities Litigation Reform Act of 1995, as amended. Forward-looking statements include all statements that are not purely historical and can generally be identified by terms such as "expects," "projects," "intends," "plans," "believes," "anticipates," "estimates," "forecasts," "targets," "goals," "may," "will," "would," "could," "should," "view," "see," or similar expressions, or the negative of such terms, or other comparable terminology. This press release specifically contains forward-looking statements regarding, among other things: (i) the Company's goal of acquiring 5% of the total ETH supply (the "Alchemy of 5%" initiative) and statements that the Company is 98% of the way to achieving this goal in 15 months; (ii) the Company's digital asset accumulation and treasury strategy, including statements regarding continued weekly ETH acquisitions since the inception of the ETH Treasury Strategy on June 30, 2025 and the Company's status as the largest ETH treasury in the world; (iii) the Company's staking operations, including projected annualized ETH staking rewards of approximately $392 million at scale (assuming Bitmine's ETH is fully staked by MAVAN and its staking partners using 2.62% 7-day BMNR yield), currently projected annualized staking revenues of approximately $334 million, and the 7-day yield of 2.62% (annualized); (iv) MAVAN's expansion to serve institutional investors, custodians, and ecosystem partners seeking best-in-class staking infrastructure, and its intended position as a premier Ethereum staking destination for BMNR and institutional investors; (v) expectations regarding future ETH price performance and market movements, including Tom DeMark's expectation that ETH will make a sharp upward move in coming weeks based on technical analysis and the belief that the August sideways movement implies a renewal of the upside move, and that a previous one-day rally was a likely preview of the pending advance; (vi) statements regarding ETH's performance as the best performing macro asset in Q3 2026 to date, outperforming the S&P 500 by 5,866bp, and that this sets the stage for institutions to add to their crypto holdings; (vii) management's belief that multiple positive catalysts exist heading into the final months of 2026, including the upcoming CLARITY Act vote scheduled for mid-September 2026, renewed buying by Korean investors and rotation away from AI stocks, the view that the four-year crypto cycle is bottoming within the next few weeks, and the expectation of sizable institutional participation in buying crypto in the final months of 2026, especially given the tailwinds of tokenization and agentic-AI; (viii) statements and expectations regarding the ETH/BTC ratio, including that the ratio has moved to the highest level since January 30th of this year and established a new uptrend, and that the ratio will rise in the upcoming crypto cycle driven by Wall Street tokenizing on the blockchain and by agentic-AI using blockchains, similar to prior cycles fueled by ICOs (2017-2018), NFTs (2020-2021), and stablecoins (2025); (ix) statements regarding Ethereum's strengthened position as the settlement rails for Wall Street tokenization and the increased realization that Ethereum may play a critical central role in managing agentic-AI; (x) management's belief that the GENIUS Act and SEC Project Crypto are as transformational to financial services in 2026 as the end of the Bretton Woods system in 1971 and that investments resulting therefrom will prove better than gold; (xi) statements regarding the Company's investments, including that its investment in Eightco Holdings (NASDAQ: ORBS) provides investors indirect exposure to OpenAI and its $180 million stake in Beast Industries; and (xii) statements regarding the value of the Company's crypto, cash, marketable securities, and "moonshot" holdings, including aggregate holdings of $15.8 billion and ETH holdings representing 4.9% of the total ETH supply.
These forward-looking statements involve substantial risks and uncertainties that could cause actual results to differ materially from those expressed or implied. Factors that could cause or contribute to such differences include, but are not limited to: the extreme volatility and unpredictability of digital asset prices, including ETH and Bitcoin, and the speculative nature of digital asset investments; the risk that historical ETH price movements, technical analysis indicators, ETH/BTC ratio trends, and relative performance versus other macro assets will not recur or are not indicative of future performance; the Company's reliance on third-party pricing sources (including Coinbase) and reported market values in calculating the value of its crypto, cash, marketable securities, and "moonshot" holdings, and the risk that such values fluctuate materially after the date and time referenced in this release; changes in market conditions affecting the trading price and trading volume of the Company's common stock and Series A Preferred Stock, and the risk that the Company's inclusion in the Russell 1000 index does not produce anticipated benefits; the Company's ability to successfully execute its digital asset acquisition strategy, continue its record of weekly ETH acquisitions, and achieve its ETH accumulation targets, including the "Alchemy of 5%" goal; the Company's ability to finance its business operations, Ethereum treasury operations, and MAVAN expansion; operational, security, and technological risks associated with the Company's staking and validation operations, including network failures, slashing events, cybersecurity breaches, and protocol changes; the risk that actual staking participation, yields, rewards, and revenues differ materially from the projected amounts described in this release, which are based on a 7-day yield and assume ETH is fully staked at scale; competition in the digital asset treasury, staking, and mining industries; the Company's dependence on key personnel, including executive leadership and advisors such as Tom DeMark; regulatory developments affecting digital assets, blockchain technology, and staking activities in the United States and globally, including the timing and outcome of the scheduled CLARITY Act vote and the ultimate enactment, implementation, and interpretation of the GENIUS Act and other pending legislation and regulatory initiatives; actions by the SEC, CFTC, and other regulatory bodies affecting digital assets and related businesses; risks related to the Company's investments in early-stage blockchain opportunities ("moonshot" investments), including the investments in Eightco Holdings (including the nature and extent of any indirect exposure to OpenAI) and Beast Industries; macroeconomic factors, including inflation, interest rates, Federal Reserve monetary policy, labor market conditions, war risks, rising yields, and general economic conditions affecting investor sentiment toward digital assets, including the behavior of Korean and other international investors; the accuracy of technical analysis predictions and management's expectations regarding ETH price movements, the ETH/BTC ratio, Ethereum's role in Wall Street tokenization and agentic-AI, and the impact of tokenization and agentic-AI applications on Ethereum; the unpredictability of cryptocurrency market cycles and the accuracy of expectations regarding future crypto cycles, including whether the four-year cycle bottoms as anticipated and whether institutional participation materializes; changes to the Ethereum protocol, including staking mechanics, validator requirements, and reward structures; the performance of third-party service providers, exchanges, custodians, and staking partners; risks related to the concentration of the Company's assets in digital currencies, particularly Ethereum; and the other risk factors described in the Company's filings with the SEC.
The forward-looking statements contained in this press release are based on information available to management as of the date of this release and reflect management's current expectations, estimates, forecasts, projections, views, and beliefs concerning future events and circumstances. Actual results may vary materially from those expressed or implied by forward-looking statements based on a number of factors, including those described above and in the Risk Factors section of the Company's Annual Report on Form 10-K for the fiscal year ended September 30, 2025 filed with the SEC on November 21, 2025, the Company's Quarterly Reports on Form 10-Q, and the Company's other filings with the SEC, as amended or updated from time to time. Copies of these filings are available on the SEC's website at www.sec.gov and on the Company's website at https://Bitminetech.io/investor-relations/. The Company cautions readers not to place undue reliance on any forward-looking statements, which speak only as of the date on which they are made. Bitmine expressly disclaims any obligation or undertaking to update, revise, or supplement any forward-looking statements to reflect any change in its expectations or any change in events, conditions, or circumstances on which any such statements are based, except as required by applicable law or regulation.
Disclaimer: This is a sponsored press release and is for informational purposes only. It does not reflect the views of Bitzo, nor is it intended to be used as legal, tax, investment, or financial advice.

Bitcoin was trading near $77,572 when CryptoQuant’s 365-day moving-average threshold stood around $81,700, a gap of roughly 5%. The arithmetic makes the level appear unusually clean: a relatively modest advance would put price above a long-term trend line that CryptoQuant treats as confirmation of a new bull market.
But a market regime is not ordinarily settled by a single print above one indicator. The more useful reading is that $81,700 is a test with historical relevance, not a switch that can independently establish a durable advance. Above it sits a cluster of cost bases, ETF break-even estimates and liquidation levels. Beneath the price recovery sits an annual loss and a recovery in capital inflows that remains materially smaller than those seen in earlier expansion phases.
That does not make the 365-day average irrelevant. It defines a concrete level at which the bearish-to-bullish argument becomes stronger. It does mean that calling a new bull market on a decisive close alone would ask one technical threshold to do more work than the surrounding evidence supports.
The 365-day moving average turns $81,700 into a test, not a switch
CryptoQuant’s framework, as reported by The Block, treats a decisive close above Bitcoin’s 365-day moving average as “confirmation” of a new bull market. Contemporaneous estimates placed the average between approximately $81,700 and $83,100. At approximately $77,572, Bitcoin had not yet met CryptoQuant’s stated condition.
The $81,700 level is not a precise binary market switch. The Currency Analytics characterized it as part of a broader resistance context, while Glassnode reported additional overhead resistance and demand below levels seen in prior bull-market expansions. A reclaim of the moving average alone would therefore not establish a durable trend.
Glassnode made a similar distinction in May when it described the True Market Mean near $78,300 as a historical dividing line between bear and bull conditions. Glassnode explicitly said reclaiming it was necessary rather than sufficient, and that consolidation could require weeks to months.
Cost bases, ETF break-evens and liquidations concentrate resistance above $81,700
The case against a binary interpretation becomes clearer immediately above the moving average. The Currency Analytics characterized $81,700 as part of a resistance zone extending toward roughly $88,700, rather than a precise line separating two market states. The distance between the lower and upper bounds is material: a close at the bottom of that range would not mean the market had traversed the full supply area.
Glassnode identified a narrower but similarly consequential overhead band around $83,000 to $86,000. Its September assessment tied that range to long-term-holder cost basis, liquidation levels and U.S. spot ETF break-even estimates. Those are separate market references, but their overlap makes the area more consequential than a standalone chart level.
Cost basis matters because it locates prices at which holders acquired coins. ETF break-even estimates add another reference point for market participants whose positions moved underwater during weakness. Liquidation levels, meanwhile, can intensify price moves when leveraged positions are forced to close. None of these measures guarantees that sellers will emerge at a particular price, and Glassnode’s analysis does not make that claim. Their concentration instead explains why a move through $81,700 could encounter a more complicated market structure above it.
The practical implication is not that Bitcoin must clear $88,700 in one uninterrupted move to make a bullish case. Markets can consolidate, retest and advance in stages. Rather, it is that the relevant question after a 365-day-average reclaim would shift quickly: can Bitcoin absorb the $83,000–$86,000 band and remain supported while doing so?
That is a tougher standard than a first close above $81,700, but it is also closer to what traders usually mean by a durable trend change. The signal becomes more persuasive when the market demonstrates acceptance above a level where several distinct sources of overhead pressure converge.
The rebound has outpaced confirmation from annual performance and capital inflows
Glassnode’s figures establish the scale of the rebound: Bitcoin gained 23% across 21 sessions. The same Week 36 report placed Bitcoin down 10% year to date, leaving the broader annual decline only partly repaired.
On capital flows, Glassnode reported in May that realized-cap net inflows had recovered to about $2.8 billion per month. Prior bull-market expansions saw more than $10 billion in monthly inflows, according to the firm.
The comparison does not make $10 billion a required near-term threshold. It does, however, limit what can be inferred from a technical reclaim: the demand backdrop does not yet clearly resemble a mature bull-market expansion. Taken together, the short-run gain, the year-to-date loss and the partial inflow recovery point to a transition that may still be incomplete.
What the threshold can still establish after a sustained reclaim
The $81,700 level retains value because it offers a disciplined condition rather than an open-ended narrative. Bitcoin was below it at the time of CryptoQuant’s cited assessment, and a decisive close above the 365-day average would be a clear improvement in the technical picture. It would also place price near Glassnode’s earlier True Market Mean reference, which was around $78,300 in May and had been framed as a historical bear-to-bull dividing line.
What follows matters more than the label attached to the first close. Sustained trading above the average would show that the reclaimed level is being accepted rather than merely tested. Progress through the $83,000–$86,000 resistance band identified by Glassnode would address a separate cluster of long-term-holder cost bases, liquidation levels and ETF break-even estimates. A broader move toward the approximately $88,700 upper end of the resistance zone would further reduce the case that the market had only cleared the first obstacle.
Those are not interchangeable tests. The moving average measures a long-duration price trend. The overhead band maps areas where market structure may create friction. Realized-cap inflows offer a view of whether capital is returning with anything like the force seen in prior expansions. Their combination is more informative than any one of them in isolation.
That is also why the threshold should not be discarded merely because it is insufficient on its own. A clear rule can prevent a short-term rally from being mistaken for a confirmed regime change. The error lies in converting a useful first condition into a complete diagnosis.
For now, the evidence leaves Bitcoin with a defined technical hurdle and a wider market challenge above it. The 365-day moving average near $81,700 can establish that a recovery has crossed an important line. Whether it develops into a durable bull market depends on consolidation through the $83,000–$86,000 band, the wider resistance zone reaching toward $88,700, and demand that remains, on Glassnode’s historical comparison, well below prior expansion-phase inflows.
Disclaimer: This article is provided for informational purposes only. It is not offered or intended to be used as legal, tax, investment, financial, or other advice.

Symbiosis said it recovered approximately 15 BTC and moved the funds into a team-controlled multisignature wallet after an attacker exploited a vulnerability in its Bitcoin Bridge at about 04:28 UTC on September 11, 2026. The protocol suspended BTC-related routes following the incident, according to its official statement.
Independent on-chain monitoring cited by Lookonchain estimated roughly $336,000 in WBTC losses linked to abnormal syBTC minting; Symbiosis said its final loss calculation was still under way.
Bitcoin Bridge exploit triggers BTC route suspension
Symbiosis said it halted BTC-related routes after identifying its Bitcoin Bridge as the affected product. The company did not provide a technical account of the vulnerability or identify the attacker.
The operational impact was described as limited to that bridge: according to Lookonchain, EVM, TRON and TON routes, Octopools and other components remained operational and unaffected.
For users, the suspension covers BTC-related activity rather than every network and product supported by the cross-chain protocol. The available disclosures did not say when the halted BTC routes would resume.
15 BTC recovery sits alongside an unconfirmed loss estimate
Lookonchain attributed the reported loss estimate to independent monitoring that linked the incident to abnormal syBTC minting through Symbiosis BridgeV2. It put WBTC losses at approximately $336,000.
Separately, Symbiosis reported recovering roughly 15 BTC and placing it in a team-controlled multisignature wallet. The protocol has not disclosed further details about the custody setup.
The two figures should not be read as a completed accounting: Symbiosis said it was still calculating the final losses. The released information did not provide a breakdown of affected positions or recipients, and the recovery does not itself establish a final loss figure or completed restitution plan.
Symbiosis sets September 13 deadline for attacker bounty
Symbiosis has offered the attacker a 20% white-hat bounty in exchange for returning the stolen funds. The offer remains open through September 13, 2026, the protocol said in its September 11 statement.
After that deadline, Symbiosis said the same reward would instead be offered for information leading to the recovery of the funds. The announcement frames the proposal as a time-limited path for the attacker to return assets before the reward is redirected toward informants.
For now, the bounty shows only that a recovery effort remains outstanding. The protocol has not said whether it received a response to the offer or disclosed how or when a return would occur, so there is no evidence here that additional assets have been recovered.
Compensation framework remains under development for liquidity providers
The outstanding question for users is how the incident will be resolved for liquidity providers whose funds may have been exposed. PANews, in a September 12 report carried by KuCoin, said Symbiosis planned to contact affected liquidity providers and was developing a compensation framework.
Neither a final loss amount nor the terms of that framework had been confirmed at the time of the report. There is also no disclosed schedule for outreach, compensation calculations or distributions.
For now, Symbiosis has provided three concrete markers: BTC-related routes were suspended after the exploit, about 15 BTC was placed in a team-controlled multisignature wallet, and the attacker has until September 13 to accept the 20% return-for-bounty proposal. The eventual accounting and the compensation process remain pending.
Disclaimer: This article is provided for informational purposes only. It is not offered or intended to be used as legal, tax, investment, financial, or other advice.

Markets priced an 87% probability of a quarter-point Federal Reserve rate hike at the September 2026 FOMC meeting after the August CPI release, up from 72% a day earlier. The repricing followed a 0.4% rise in U.S. CPI in August and a 3.4% increase over the 12 months through August, according to PriceVia.
The jump is notable because it puts market pricing closer to J.P. Morgan Wealth Management's revised forecast for a September quarter-point increase while challenging Goldman Sachs' expectation that the Fed would hold rates. Market-implied odds are a measure of investor pricing, not a decision by policymakers.
Data Snapshot
MetricCurrentPreviousChangePeriodAs ofSourceFed September rate-hike probability87%72%up from 72% a day earlierSeptember 2026 FOMC meeting2026-09-11PriceViaMonthly CPI0.4%——August 20262026-09-11PriceViaAnnual CPI3.4%——12 months through August 20262026-09-11PriceViaCore monthly CPI0.3%——August 20262026-09-11PriceVia10-year Treasury yield4.9915%——Intraday high2026-09-11PriceViaJ.P. Morgan September policy call25-basis-point rate hikeno rate changes in 2026changed outlookSeptember 2026 FOMC meeting2026-08-05J.P. Morgan Wealth Management
August CPI and hike odds
August headline CPI rose 0.4% month over month and core prices increased 0.3%, while markets raised the implied probability of a quarter-point hike at the September 2026 FOMC meeting from 72% to 87%, PriceVia reported.
The 87% measure reflects investor pricing for that specific policy action, not the inflation readings or the votes of individual Fed officials; the one-day increase indicates that traders reassessed the likely policy response after the release.
J.P. Morgan's September call
J.P. Morgan Wealth Management strategists had already changed their outlook on August 5, calling for a 25-basis-point rate hike at the September 2026 meeting. Their earlier base case had been for no rate changes in 2026, according to J.P. Morgan Wealth Management.
That earlier shift means the firm was positioned for a September increase before the August CPI release pushed market-implied probabilities higher. The subsequent pricing move strengthened the alignment between its policy call and market expectations, without confirming the eventual outcome.
Goldman's hold call
Goldman Sachs chief economist Jan Hatzius was still expecting the Federal Reserve to hold rates in September as of August 31. His view was conditional: a hike would require upside surprises in August CPI and PPI, InvestingLive reported.
The CPI release was followed by the rise in implied hike odds to 87%, but the supplied data do not state whether Goldman Sachs changed its forecast after the report or provide August PPI results. The Fed's next concrete decision window is its September 15–16, 2026 meeting, as listed on the Federal Reserve's calendar.
In bond markets, the 10-year Treasury yield reached an intraday high of 4.9915% on September 11, with PriceVia describing the yield as near 5%. That reading accompanied the inflation-driven repricing ahead of the FOMC meeting.
Disclaimer: This article is provided for informational purposes only. It is not offered or intended to be used as legal, tax, investment, financial, or other advice.

Sportsbook margin works differently from a casino house edge, which is published in every game's information panel. A sportsbook publishes nothing, and the equivalent number is sitting in plain view on the odds themselves.
It takes about ten seconds to calculate, and almost nobody does it.
Working It Out
Five steps, using figures already on your screen.
Convert each price to an implied probability. Divide one by the decimal odds. A price of 1.91 becomes 1 ÷ 1.91 = 52.36%.
Do the same for every outcome in the market. A two-way market priced at 1.91 and 1.91 gives 52.36% twice.
Add them together. 52.36% + 52.36% = 104.71%.
Read the excess over 100%. That 4.71% is the overround, and it is the book's margin on that market. A fair market with no margin would sum to exactly 100%.
Convert to cost per unit staked if you want the precise figure. Divide the overround by the total: 4.71 ÷ 104.71 = roughly 4.5% of everything you stake into that market.
That is the whole calculation. No tools, no data feed, just the prices in front of you.
A Three-Way Example
Football match markets have three outcomes, so the sum runs across all of them.
Take a match priced at 2.40 for the home side, 3.40 for the draw, and 3.10 for the away side.
1 ÷ 2.40 = 41.67%
1 ÷ 3.40 = 29.41%
1 ÷ 3.10 = 32.26%
Total: 103.34%, so an overround of 3.34%, costing roughly 3.23% per unit staked.
That is a competitive market-result price. Hold the figure in mind, because the next section is where it stops being competitive.
Margin Varies Enormously Within One Book
This is the part that matters more than comparing operators, and it is invisible unless you check.
A casino's house edge is a property of the game. A sportsbook's margin is a property of the market, so market pricing shifts across one board, and the same book will run very different margins across its own board on the same match.
Mainstream match-result markets on major competitions carry the thinnest margins, since that is where competition between books is fiercest
Props, exotics, niche competitions and heavy combinations carry considerably more, since fewer people price-check them and the book carries more uncertainty
So a bettor who sticks to headline markets pays a fraction of what a bettor filling a slip with props pays, at the identical sportsbook, on the identical match. That is a decision entirely within your control and it is worth more than any promotion.
Platforms offering over 100 markets on a single match give you the widest possible spread of margins to choose between. Dexsport publishes that depth on major events, which means both the cheap markets and the expensive ones sit in the same list.
Line Shopping Follows Directly
The practical consequence follows directly.
Two books pricing the same event produce different implied probabilities, and the difference is real money on every bet you place. Taking 2.40 where another book offers 2.30 is not a rounding difference, it is roughly 4% more return on a winning selection, permanently, on every bet where that difference exists.
Margin is the reason that difference exists at all. A book running 3% overround can price more generously than one running 7%, and neither is being charitable, they are running different business models.
Two honest limits on this:
A low-margin book is not a winning proposition. It is a less costly one. The margin is positive at every book on every market, which is the entire point of the business.
Comparison takes seconds and applies per market, not per book, since the same operator runs thin margins in one place and wide ones in another.
What On-Chain Settlement Does and Does Not Cover
Worth being precise, since this is a sportsbook article and the distinction gets muddled.
Off-chain pricing is how Dexsport sets its odds, like every book operating at scale, because live pricing needs continuous data feeds, trader judgement, and latency no public chain can supply economically. Its settlement is then written to a public on-chain desk, so a resolved market leaves a timestamped record.
That record shows what was paid. It shows nothing about whether the price you accepted carried a 3% margin or a 9% one, because the price was set somewhere the chain never saw. On-chain recording covers a specific and narrow claim, and margin sits outside it entirely.
Which is why the calculation above is worth learning. It is the only way to see the number, and nobody is going to publish it for you.
Ten Seconds, Before Every Bet
Divide one by each price to get implied probability
Add every outcome together
Anything over 100% is what that market costs you
Do it on a market-result price and a prop on the same match, and the difference will change which markets you use.
Confirm what is legal where you live, keep stakes within a set budget, and play only if you are of legal age, since KYC or AML checks may apply.
Responsible gambling connects directly: margin is a certain cost applied to every stake, so a larger number of bets means a larger total paid regardless of how any individual one lands.


Disclaimer: The information here is provided for general purposes only and is not legal, tax, investment, or financial advice, and nothing here is a betting tip or prediction. Odds used are illustrative examples. Prices, margins and market availability vary by operator and change constantly. Betting carries risk, and rules vary by country, so check the law where you live. Please gamble responsibly, within your means, and only if you are of legal age.

A void bet is the version nobody expects: your selection won and the stake came back. Or the match stopped at 70 minutes and half your slip settled while the rest disappeared. Neither is an error, and both are covered by rules you agreed to.
Here are the five situations that undo a bet, and the one that is genuinely contentious.
Five Ways a Bet Comes Undone
The first four are procedural. The fifth is the one worth arguing about.
A Postponed Fixture
The near-universal convention is a 48-hour window. If the event is played within 48 hours of its original scheduled start, bets stand and settle on the result. Past that window, they are voided and stakes are returned.
One detail catches people: a fixture rescheduled before the original kick-off is usually not treated as a postponement at all, so bets simply carry to the new date. The clock only matters once the advertised start has passed.
Sport-specific windows vary. Track and field events commonly run to seven days, and several US sports are settled on a same-day or same-scheduling-week basis instead of the standard 48 hours.
An Abandoned Match
Where play has started and stopped, the rule splits your slip in two.
Markets already unconditionally determined settle as normal. Everything else voids.
The clearest illustration comes from the house rules themselves: a baseball match abandoned at 5-4 after five innings settles an Over/Under 8.5 runs market in full, because nine runs were already on the board and no further play could change the answer.
A match-winner bet on the same game voids, because it could have.
So an abandonment is not all-or-nothing. Part of your slip is decided, and part is refunded, which is why an abandoned match can produce a confusing mix of settlements.
A Non-Runner or a Retirement
A competitor who does not take part voids the market they were in. That applies to horse racing withdrawals, player props where the player did not appear, and similar.
Tennis is the notable exception. Markets generally stay open until officials declare a winner, and the 48-hour rule does not apply. On a mid-match retirement, markets already determined settle, so specific set and game results pay out, while the match-winner market voids.
A Push on the Line
Where a handicap or total is set at a whole number, and the result lands exactly on it, the bet voids and the stake returns.
A total of 2 goals with the match finishing 1-1 is neither over nor under. Half-point lines exist precisely to remove this outcome, which is why most books quote 2.5 instead.
A Palpable Error
This is the contested one, and it is the only category where a bet you have already won can be unwound.
Every major sportsbook reserves the right to void a wager accepted at an obviously incorrect price, under a clause usually named palpable error or obvious error.
The typical case is a pricing mistake far outside the wider market, such as an extra digit, or a live price that had not updated after a decisive event.
It is not intended to cover ordinary odds movement, and it does mean acceptance is not final. Your bet was confirmed, the stake left your balance, and the operator retains a route to unwind it.
Worth knowing: Cash Out values are bound by the same clause at many books, so an exit price taken in error is equally reversible.
Multiples Get Recalculated
One mechanical point that surprises people with accumulators.
A voided leg does not void the whole ticket. The multiple is recalculated to the next lowest number of selections, so a five-fold with one void leg becomes a four-fold at the combined odds of the remaining four.
That is usually fair and it does change the bet you thought you had, since the return drops in line with the shortened price.
This Is the Operator Applying Its Terms
Worth stating clearly, because these situations feel like mistreatment and mostly are not.
Every rule above sits in published house rules, and an operator applying them is doing what the terms describe. The exception is palpable error, where reasonable people disagree about how obvious an error has to be, and where the clause gives the operator a judgement call you cannot appeal on equal footing.
The useful response is therefore preparation instead of complaint. Read the abandonment window for the sport you bet on, since it varies, and check whether your book uses 48 hours, a same-day rule or something longer. Licensing determines what recourse exists if you do end up disputing one.
Where a Settlement Record Earns Its Keep
This is the specific scenario where an on-chain record does concrete work, which is worth naming since the feature is usually described in general terms.
A settlement dispute is an argument about what happened and what was paid. Dexsport writes settlement to a public on-chain desk, so a resolved market leaves a timestamped entry independent of the account screen. Both sides can point at the same record instead of at a support ticket and a screenshot.
Two honest boundaries. It records settlement, not the decision behind it, so it shows a market was voided without showing whether voiding was correct. And it covers settled play and not the terms layer, where how a platform handles funds and its licence tier decide what happens next.
On Dexsport that means retrieving one settlement early on a small bet, so you know where the record lives before you need it. Most people learn the process during a dispute, which is the worst possible time.
Before You Bet
Find the abandonment window for your sport, since 48 hours is common and not universal
Check whether the market uses whole numbers, because a whole-number line can push
Accept that palpable error makes acceptance provisional, at every book, not just one
Confirm what is legal where you live, keep stakes within a set budget, and play only if you are of legal age, since KYC or AML checks may apply.
Responsible gambling has an incidental connection: a voided bet returns a stake to a balance that is still in a betting account, and re-staking it immediately is the reflex worth resisting.


Disclaimer: The information here is provided for general purposes only and is not legal, tax, investment, or financial advice, and nothing here is a betting tip or prediction. Settlement rules, windows and void conditions vary by operator and sport and change over time, so read the current house rules of the platform you use. Betting carries risk, and rules vary by country, so check the law where you live. Please gamble responsibly, within your means, and only if you are of legal age.

Operator failure happens. Not often, and not usually dramatically, but a brand that was taking deposits last year can be a dead domain this year with balances still inside it.
That is not a hypothetical risk in 2026. It is a live consequence of a regulatory transition happening right now.
The Reason This Is Live Now
The Curaçao reform replaced the old master-and-sublicence structure with direct licensing under the Curaçao Gaming Authority. The register published in July 2026 carried roughly 660 entries with issue and expiry dates, including licences the regulator had revoked.
That transition created a specific problem. Grandfathered sublicence holders who never completed the move have been dropping off the register in batches, and each removal creates orphaned brands, clusters of white-label sites still trading with a seal that no longer verifies against anything.
So there are live sites today displaying licensing that stopped being real, with player balances sitting inside them. The question of whether an operator will still be here next year has stopped being abstract.
Four Signals That Predict Durability
None is conclusive. Together they are the strongest available read.
Signal
What it predicts
Length of trading history
Survival through at least one market cycle and one regulatory change
A named operating entity
That a dispute has a legal person to be against
Licence continuity
That the operator completed the reform instead of lapsing
Custody model
Whether a failure reaches your balance at all
The fourth row is different in kind from the first three. History, naming and licensing are evidence about an operator. Custody is a structural property that changes what an operator's failure can cost you, and licensing determines recourse only where there is a balance left to recover.
1. Cloudbet
Trading since 2013 with its operating company named on a Curaçao licence, which is the strongest combination available on the first three signals.
Thirteen years covers multiple market cycles, the collapse of several competitors, and a full regulatory transition. A named entity means a dispute has a defendant. Reformed Curaçao licensing adds a complaint channel above that.
Balances are custodial, so the fourth signal does not apply, and its orientation toward higher limits suits larger players more than casual ones.
2. Stake
A large operator with market-specific licences in several jurisdictions alongside its offshore position, and a scale that makes quiet disappearance implausible.
Size is itself a durability signal here: a business of that footprint has too much to lose from a failure to pay, and holds licences that would be forfeited by one. Balances are custodial, with withdrawal minimums per asset.
3. BC.Game
Years of continuous trading under Curaçao licensing, now operating within the reformed framework with named beneficial owners on record.
Completing the transition is the meaningful part. An operator that made it onto the new register did the work the lapsed brands did not, which is a filter in itself.
4. Dexsport
Dexsport is newer than the three above, so it scores lower on trading history, and there is no honest way around that.
Its answer is structural instead of historical. Because the platform is non-custodial, settled play returns to a wallet you control, and there is no operator-held balance accumulating between sessions. Whatever happens to the operator, there is nothing of yours sitting inside it to lose.
That is a genuine response to the risk this article is about, and it deserves a boundary. It covers settled funds.
It does not cover money committed to an open bet, and it says nothing about disputes over terms, where its Anjouan licence sits lighter than reformed Curaçao and well below Malta. Its licence is verifiable, since the Anjouan register accepts a domain search.
5. Vave
A conventional operator with multi-coin funding and a shorter public record than the platforms above.
Its licensing position takes more work to pin down, which on the second and third signals is itself informative. Custodial balances.
6. Mega Dice
Telegram-first access across around 50 providers, with the thinnest published documentation of the six.
Where an operator publishes least about its own licensing and terms, you have the least to go on when assessing whether it will still be trading in a year, and documentation quality clusters with everything else.
Signs That Come Before a Closure
Operators rarely vanish without warning. The warning is just quiet.
Withdrawal processing times lengthen first, often with plausible explanations attached. Support response times slip. Promotions get more aggressive as the operator works harder to attract deposits.
The licence entry stops resolving, or the register shows an expiry that has passed. And on platforms with visible reserves, treasury movement sometimes appears on-chain before any public complaint does.
None of those individually means a closure is coming. Several at once, particularly lengthening withdrawals alongside a licence that no longer verifies, is the pattern worth acting on immediately.
Recovery, Realistically
The honest answer, because it is short.
An unsecured balance at a failed offshore operator is usually unrecoverable. There is generally no deposit protection, no segregated fund requirement enforced by a light-touch regulator, and no insolvency process a player can practically join from another jurisdiction.
Which makes the only effective protection preventative:
Withdraw when you stop playing, instead of leaving a balance idle
Weigh licensing before depositing, since recourse is decided then
Understand the exposure: a balance in an account depends on that operator continuing to exist
Confirm what is legal where you live, keep stakes within a set budget, and play only if you are of legal age, since KYC or AML checks may apply. Responsible gambling overlaps here usefully: clearing a balance out at the end of a session protects it from the operator and from you at the same time.


Disclaimer: The information here is provided for general purposes only and is not legal, tax, investment, or financial advice. Trading histories, licensing positions and operator circumstances change, so verify current details with the relevant register directly. Nothing here is a prediction about any operator's future solvency or conduct. Betting carries risk, and rules vary by country, so check the law where you live. Please gamble responsibly, within your means, and only if you are of legal age.

The same slot can return 96% at one casino and 94% at another. Identical artwork, identical mechanics, identical name in the lobby.
RTP versions exist because the number is a setting, not a property, and checking which one you are playing takes about ten seconds.
Five Things to Do About It
The situation is fixed. What you do about it is not.
Understand that operator selection decides it. Studios certify several builds of one title, all certified builds of one title at different returns and ship them as a menu. The casino selects which runs across its whole player base, so the figure you see is a commercial decision made by the platform and not a characteristic of the game.
Know how wide the spread goes. Starburst has shipped at 90.10%, 92.13%, 94.10% and 96.09%. Aviator is configurable at 97%, 96% or 94%. Between the highest and lowest of those ranges the house edge more than doubles, with nothing visible on screen to distinguish them.Find the figure in the information panel. Certification requires it to be published, so it is always there. Usually under an information or rules icon, sometimes at the foot of a paytable.
Use demo mode so the check costs nothing. Opening a title in demo reaches the same panel without staking anything, which turns a commercial decision you cannot influence into one you can at least see.
Compare the same title across two platforms. This is the step nobody takes and the only one that tells you whether the version in front of you is generous or mean relative to what exists. Ten seconds on a second site settles it.
Six Platforms and Who Sets the Figure
Ranked on a single axis: whether the platform sets its own return figures or selects them from a supplier's menu. The first two build their own games. The other four choose from what studios ship them.
1. Stake
Runs its own originals at 99%, which means it sets the figure directly instead of selecting from a supplier's menu.
That is the strongest position on this specific question. A platform building its own games controls the edge and uses the low number as a competitive feature, so the floor across its originals is published and stable.
Its licensed catalogue is subject to the same selection process as everyone else's. Balances are custodial.
2. BC.Game
Also builds originals at 99% alongside a wide licensed library, under a long Curacao trading record.
Same structural advantage as Stake: a house-built layer where the platform sets the price, plus a licensed layer where it selects from menus. Its scale means panel figures are easy to locate across both.
3. Cloudbet
Trading since 2013 with its company name on the licence, and a conventional licensed catalogue.
No original suite, so every figure is selected from a supplier's range. Its orientation toward higher limits means the selections tend toward established titles where published figures are widely known and comparable.
4. Dexsport
Dexsport licenses its entire catalogue across roughly twenty slot studios and runs no in-house originals.
That consequence is worth stating plainly: every return figure in its lobby is a studio build the platform selected, with no 99% house-built floor sitting alongside them. That is the same position as any licensing-only operator and it is the honest description.
What it does offer is demo mode across much of the library, which makes step four free and step five easy. Open the title, read the panel, then check the same title elsewhere. Non-custodial, with an Anjouan licence lighter than Curacao or Malta.
5. Vave
Third-party coverage with multi-coin funding and no originals suite.
Every title inherits a studio-set figure selected by the operator, placing the whole lobby in the 95.6% to 98.2% band typical of licensed content with nothing below it and nothing above.
6. Mega Dice
Around 50 providers feed the catalogue through a Telegram-first product.
A large supplier count means a large number of selection decisions, and its documentation is thinner than the platforms above, so the panel check matters more here, not less.
What the Ranking Actually Reflects
Worth being explicit about the axis, since it is narrow.
Platforms building their own games rank higher here because they set a floor instead of selecting one. That says nothing about catalogue quality, licensing strength, withdrawal handling or anything else, and a licensing-only platform with a strong licence may suit you better on every other measure.
The point is only that on this question, in-house originals change who is making the decision, and the published figure is where it becomes visible.
Ten Seconds, Every New Title
Open the panel before the first spin
Note the figure, and whether a volatility rating sits beside it
Check the same title elsewhere if it looks low, before continuing
Certification means the number is always published, so the only thing standing between you and it is the habit of looking, and verifying what a game actually does is a separate check again.
Confirm what is legal where you live, keep stakes within a set budget, and play only if you are of legal age, since KYC or AML checks may apply. Responsible gambling matters regardless of which build you found, since a 97% game applied to enough rounds still returns less than it takes.



Disclaimer: The information here is provided for general purposes only and is not legal, tax, investment, or financial advice. Return configurations vary by provider, version and operator and change over time, so consult each game's published information before playing. Betting carries risk, and rules vary by country, so check the law where you live. Please gamble responsibly, within your means, and only if you are of legal age.

Japan’s all-commodities Producer Price Index rose 7.6% year on year in August 2026, according to preliminary data from the Bank of Japan. The reading was slightly below the revised 7.7% increase recorded previously, but it stood alongside a much faster 24.8% annual rise in all-commodities import prices on a yen basis.
The contrast puts imported costs at the sharper end of the August price data. The Bank of Japan’s Corporate Goods Price Index covers goods traded in Japan’s corporate sector and includes producer, export and import price indices.
Data Snapshot
MetricCurrentPreviousChangePeriodAs ofSourceProducer Price Index — All commodities, yearly change7.6%r 7.7%—August 20262026-09-10Bank of JapanProducer Price Index — All commodities, monthly change-0.2%r 0.4%—August 20262026-09-10Bank of JapanImport Price Index — All commodities, yen basis, yearly change24.8%r 29.3%—August 20262026-09-10Bank of JapanImport Price Index — All commodities, contract currency basis, yearly change16.7%r 17.8%—August 20262026-09-10Bank of Japan
Producer prices rise 7.6% annually but fall 0.2% from July
The 7.6% year-on-year gain in the Producer Price Index measures the change from August 2025, while the index fell 0.2% from the preceding month. The monthly result followed a revised 0.4% increase previously.
Those readings describe different comparisons and should not be treated as interchangeable: the annual figure shows the level against the same month a year earlier, whereas the monthly figure tracks the move from July. Both figures were released as preliminary August 2026 data in the Bank’s monthly CGPI report on September 10, 2026.
The annual producer-price increase therefore remained substantial even as the all-commodities index declined on the month. The Bank’s release uses CY2020 as the CGPI index base year.
Yen-basis import prices outpace domestic producer-price inflation
All-commodities import prices on a yen basis increased 24.8% from a year earlier in August, compared with the 7.6% gain in the all-commodities Producer Price Index. The import-price reading eased from a revised 29.3% annual increase previously, but remained far above the domestic producer-price measure.
Within the CGPI framework, the Producer Price Index measures producer prices and the Import Price Index measures import prices. Because their scopes differ, the gap shows only that the Bank’s August release recorded a steeper annual change for the Import Price Index; it does not establish that every domestic producer faced the same cost increase.
For wholesale-price watchers, the 24.8% yen-basis rise is the standout comparison in the report. It shows that imported goods prices, when expressed in yen, were rising materially faster than the broader producer-price index in August.
Contract-currency imports rose 16.7%, below yen-basis gain
On a contract-currency basis, all-commodities import prices rose 16.7% year on year in August, down from a revised 17.8% increase previously. That was lower than the 24.8% increase on a yen basis.
The two import readings use different bases: the contract-currency measure tracks prices in the currency used in the transaction, while the yen-basis measure expresses import prices in yen. Their difference should therefore not be read as a comparison of two identical price measures.
Both annual import-price measures slowed from their revised prior readings in August. Even so, the yen-basis index’s 24.8% increase remained well above the 16.7% contract-currency result and the 7.6% annual rise in producer prices.
Disclaimer: This article is provided for informational purposes only. It is not offered or intended to be used as legal, tax, investment, financial, or other advice.

The title of this article contains a mistake, and it is the same mistake most coverage makes. Those are not three independent companies.
Corporate ownership is the issue: Evolution owns Ezugi. The genuine split is Evolution against Playtech, with Ezugi operating as a specialist arm inside the larger group. That changes what a lobby carrying all three actually offers.
Three Profiles, Two Groups
Three profiles, and the third is the one people misread.
Evolution built the category and still dominates it. Dream Catcher launched the live game show genre in 2018, and the catalogue since includes Crazy Time, Monopoly Live, Mega Ball and Cash or Crash, the last at a published 99.59% return that makes it the least costly game show available anywhere. Its Lightning series applies multipliers to classic tables, and its production values are the industry benchmark.
Playtech is the only independent challenger at scale. It produces The Greatest Cards Show at 97.30%, which undercuts most Evolution wheels, alongside Family Feud and the Quantum series. The pattern is consistent: Playtech competes on price and format variety where Evolution competes on production and brand licensing.
Ezugi fills a different need from its parent. Now inside Evolution, it focuses on regionally localised tables with native-language dealers and variants built for specific markets. A lobby carrying Ezugi is not duplicating Evolution content, it is reaching players Evolution's flagship studios were not designed for.
So the useful question is whether a platform carries both corporate groups, since that is what determines whether you can compare a 97.30% Playtech show against a 96.08% Evolution wheel instead of taking one house's pricing.
1. Dexsport
Dexsport draws live content from all three: Evolution, Playtech and Ezugi.
That covers both independent groups, which is the meaningful threshold. In practice it means the Evolution game show catalogue, Playtech's alternatives at different price points, and Ezugi's localised tables in one place.
The range shows up most clearly in its roulette section, which runs to more than 50 tables spanning European, American and French variants plus dedicated VIP tables at higher limits. Three studios is what put French tables within reach at all, since a single-provider lobby inherits only that provider's range.
2. Stake
The widest live section of the four, with multiple studios and branded tables produced in partnership with providers.
Those branded variants are genuinely distinctive, since they are built for one operator and not licensed off the shelf. Scale means both corporate groups are represented alongside additional suppliers.
Balances are custodial and held between sessions.
3. BC.Game
A substantial live catalogue built over a long Curacao trading record, covering the major studios.
Its in-house originals sit alongside the licensed tables, which adds a third design source independent of both Evolution and Playtech, though the originals are RNG and not live.
4. Cloudbet
Trading since 2013 with its company named on the licence, and a live section oriented toward higher table limits.
Coverage focuses on the headline studios and established formats, with fewer novelty variants than the platforms above. For a player who wants a high-limit seat at a conventional table, that focus is a feature.
Studio Coverage Decides Your Options
Two consequences, and both are practical.
Variant availability follows the roster. French roulette, localised tables, specific game show formats and the price points attached to each all come from whichever studios a platform licensed. A lobby with one provider offers one company's decisions about what a live section should contain.
Price range follows it too. Cash or Crash at 99.59% and Mega Ball at 95.40% both come from Evolution, and The Greatest Cards Show at 97.30% comes from Playtech. Reaching all three means you can choose on cost instead of taking what is in front of you.
That is the argument for checking the studio list before the table count, since who supplies a lobby sets the boundary on everything inside it.
A Check Before You Sit
Open the live section and look for provider filters, which most platforms expose.
One corporate group only means one house's catalogue and one house's pricing
Both groups present means you can compare a Playtech show against an Evolution wheel on cost
Once you have chosen, running one balance across casino and sportsbook makes moving between sections straightforward.
Confirm what is legal where you live, keep stakes within a set budget, and play only if you are of legal age, since KYC or AML checks may apply.
Responsible gambling deserves a note at live tables specifically, since a human host and a continuous stream are designed to keep a session running past the point a solitary interface would let it end.


Disclaimer: The information here is provided for general purposes only and is not legal, tax, investment, or financial advice. Studio rosters, ownership structures, return figures and table availability vary by operator and change over time, so confirm current details before playing. Betting carries risk, and rules vary by country, so check the law where you live. Please gamble responsibly, within your means, and only if you are of legal age.

On September 10, 2026, Avalanche announced that Arya.ag and Finternet are bringing infrastructure for Indian agricultural finance onchain alongside four to five major lenders.
The initiative centres on warehouse collateral and lending records, with crops in Arya.ag's warehouses valued at about $2 billion. It is not a public venue for trading digitized grain.
According to CoinDesk, Arya.ag is developing a dedicated Avalanche-based Layer 1 for grain deposits, electronic negotiable warehouse receipts and loan-status information, allowing lenders to verify collateral through a shared ledger.
Arya.ag’s Avalanche Layer 1
According to CoinDesk, the dedicated chain is being built to record grain deposits, electronic negotiable warehouse receipts, known as e-NWRs, and loan-status data. The stated objective is to allow lenders to verify collateral through a shared ledger.
That focus matters because the records connect several parts of a warehouse-backed lending process. A grain deposit establishes what is held in storage; an e-NWR is the financing document tied to the warehouse-held goods; and loan-status information shows the state of associated borrowing. Bringing those data points into one system is distinct from placing a commodity itself into a freely traded crypto market.
Avalanche’s September 10 announcement named Arya.ag and Finternet as participants and said four to five major lenders were involved. Neither the announcement nor the reporting supplied in this account identifies those lenders or sets out their individual roles.
What the $2 billion represents
The headline figure describes crops held in Arya.ag warehouses across India, valued at approximately $2 billion, CoinDesk reported. It should not be read as a disclosed amount of assets already migrated to the Avalanche-based network or as the value of grain available for token trading.
Arya.ag separately facilitates roughly $1.3 billion in agricultural loans each year, according to the report. Arya Dhan, the company’s direct-lending operation, issues about $230 million. Those are related measures of financing activity, but they are not interchangeable with the estimated value of crops in storage.
The distinction provides a clearer view of the scale being discussed. Warehouse-held crop collateral is one measure; loans facilitated across a year are another; and loans directly issued by Arya Dhan are a narrower subset. Public reporting has not specified how much of any of those categories has been recorded onchain so far.
Physical grain and e-NWRs remain central
The project concerns digital records supporting warehouse-backed lending, not public trading in tokenized grain. Metaverse Post reported that the physical crops remain in warehouses while India’s legally recognized e-NWRs continue to serve as the underlying financing documents.
In practical terms, the proposed chain would coordinate information around collateral rather than replace the stored crop or the warehouse-receipt framework. The grain remains physical collateral, and the e-NWR remains the document on which the financing arrangement is based.
That boundary also narrows what can be inferred from the use of the word “onchain.” The disclosed purpose is lender verification of deposits, receipts and loan status through a shared ledger. The available reports do not describe a public market in which investors can buy and sell claims on individual grain deposits.
Bank participation and rollout
Execution details remain limited. The Crypto Times reported that three major banks are preparing to join the dedicated chain, but did not name the banks.
The amount already migrated onchain has not been disclosed, nor has a timetable for scaling the system. As a result, the announcement establishes the intended architecture and the warehouse-collateral base cited by Arya.ag, while leaving the pace of lender adoption and live onchain volume unresolved.
Disclaimer: This article is provided for informational purposes only. It is not offered or intended to be used as legal, tax, investment, financial, or other advice.

Positron AI announced on September 10 that it raised $875 million in a Series C financing at a $5 billion post-money valuation, providing capital for its next-generation AI inference hardware. Qatar Investment Authority confirmed that it participated in the round, saying the funding will support development of Positron’s Asimov silicon and commercialization of its Titan systems.
The financing puts a sharply higher price on the AI-chip startup only months after its prior raise. The proceeds are earmarked for chip design and manufacturing work, engineering data-center capacity, and expanding production of its hardware systems.
Series C structure and $5 billion valuation
The financing is structured as a $375 million Series C at a $3.5 billion pre-money valuation and a Series C-1 tranche of up to $500 million, according to Positron.
The disclosed headline amount is $875 million, including as much as $500 million from the contingent C-1 tranche. The announcement associates the $5 billion post-money valuation with the Series C and provides no additional C-1 financial terms beyond its potential size.
Reuters reported via Investing.com that Positron’s valuation was about $1.06 billion after a $230 million funding round in February 2026. It is now $5 billion, an increase of nearly five times in a matter of months.
Funding Asimov tapeout, data-center capacity and Titan production
Positron said the proceeds will fund tapeout of Asimov silicon, a more than 2-megawatt engineering data center, and a scale-up in Titan production. Tapeout is a critical step in moving a chip design toward manufacturing, making the stated uses of capital directly tied to bringing the company’s inference silicon and systems to market.
QIA’s confirmation focused on the same execution priorities. In its statement on the investment, the sovereign wealth fund said its participation would support Asimov’s development and Titan’s commercialization.
That allocation places the new financing across the chip-development, testing-infrastructure and system-production stages. Positron’s September 10 announcement did not provide a timetable for the Asimov tapeout, completion of the engineering data center, or Titan production ramp.
Disclaimer: This article is provided for informational purposes only. It is not offered or intended to be used as legal, tax, investment, financial, or other advice.

Bitwise Investment Advisers will liquidate and close the Bitwise Dogecoin ETF (BWOW), ending the fund less than a year after its November 2025 launch. The firm announced the decision on September 10, saying it was optimizing its product range for evolving investor needs; BWOW’s final expected trading day on NYSE Arca is October 14, 2026.
The closure removes one of Bitwise’s Dogecoin-focused exchange-traded products from the market. Remaining shareholders are scheduled to receive cash tied to the fund’s October 21 net asset value on or around October 22.
BWOW liquidation schedule and shareholder cash payment
Bitwise said in a Form 8-K dated September 10 that it would liquidate and close BWOW as part of product-range optimization for evolving investor needs.
Under the schedule posted on the official BWOW website, October 14 is the final expected NYSE Arca trading day. Shareholders still holding shares are due cash based on the October 21 NAV, with payment expected on or around October 22.
BWOW’s assets and trading activity after its November 2025 launch
BWOW began trading on November 25, 2025. It held approximately $687,713 in net assets as of September 9, according to the fund’s official website, leaving it with a relatively small asset base less than 12 months after launch.
The Block reported that trading interest faded after the opening session: BWOW generated about $3 million in volume at launch but did not come close to that level afterward.
The sub-$700,000 asset total and the post-launch drop in trading activity provide context for Bitwise’s decision to streamline its product lineup. The firm’s stated rationale, however, was its effort to adapt the range to evolving investor needs.
Disclaimer: This article is provided for informational purposes only. It is not offered or intended to be used as legal, tax, investment, financial, or other advice.

A casino advertising twenty studios sounds like twenty different design philosophies. It is frequently fewer than that, because the industry consolidated while nobody was counting.
Provider range still matters. It just measures something narrower than the number suggests.
What a Provider Count Actually Predicts
Two columns, and the second is the one people assume the number covers.
Provider count predicts
It does not predict
Mechanical variety across the lobby
Title count, which reflects licensing volume
Category coverage: Megaways, cluster pays, hold-and-win
Independence, because of consolidation
The range of return figures available to select from
Quality, since every studio ships good and poor titles
The third row on the left is worth pausing on. Studios ship the same title in several certified builds and the operator chooses which runs, so more studios means a wider menu of figures the platform could have selected from. That is genuinely useful, and the figure that applies is in each game's panel.
Consolidation Nobody Mentions
Here is the correction that changes how you read every roster.
Evolution ownership runs deep: it owns Ezugi, NetEnt, Red Tiger and Big Time Gaming. A lobby listing all four alongside Evolution itself is listing five names and one corporate group.
That matters because commonly owned studios share technology stacks, release calendars, and increasingly design conventions. Two brands under one roof produce more similar output than two genuinely independent developers, so the effective diversity of a roster is lower than its length.
So the useful question is not how many names appear, but how many distinct houses they represent, and whether the roster reaches categories no single group dominates.
1. Stake
The widest effective range of the seven, combining a large licensed roster with its own in-house originals documented at 99%.
Those originals are the differentiator here, not the licensed breadth. A platform building its own titles adds a genuinely independent design source and a lower house edge in the same move, which no amount of licensing achieves.
Balances are custodial and held between sessions.
2. BC.Game
Comparable scale under a long Curacao trading record, again pairing an extensive licensed catalogue with house-built originals at 99%.
Its licensed roster covers the major houses, and the in-house layer does the same work as Stake's: independence from the consolidation described above, plus a price advantage.
3. Dexsport
Dexsport draws on roughly twenty slot studios, which is a substantial roster by any measure.
Named across it are Pragmatic Play, NetEnt, Play'n GO, Hacksaw, Nolimit City, BGaming, Playson, Yggdrasil, Red Tiger, Endorphina, Spinomenal, 3 Oaks, PG Soft, Novomatic, Big Time Gaming, Spribe, TaDa and Turbo Games.
That reaches Megaways through Big Time Gaming, high-volatility design through Nolimit City and Hacksaw, and crash formats through Spribe and TaDa.
The platform runs no in-house originals, which means no 99% floor anywhere in the lobby and every return figure is a studio build the operator selected. Non-custodial, with an Anjouan licence lighter than Curaçao or Malta.
4. Mega Dice
Around 50 providers feed the wider catalogue, which is the largest raw number here.
Raw count is exactly what the consolidation point warns about, and a roster assembled through aggregators frequently includes many small studios producing similar content. Breadth of names, less breadth of approach. Telegram-first access, with documentation thinner than the platforms above.
5. Vave
Solid coverage of the major houses without an originals suite.
That absence is the defining feature: every title carries a studio-set return, so the whole lobby sits in the 95.6% to 98.2% band typical of licensed content with no house-built alternative alongside it.
6. Cloudbet
A conventional roster covering the main providers, trading since 2013 with its company named on the licence.
Fewer novelty studios than the platforms above, which suits its orientation toward larger stakes on established formats more than toward mechanical variety.
7. Rollbit
The narrowest roster of the seven, with wallet-forward access and on-chain elements outside the casino.
Adequate coverage of mainstream content, with noticeably less reach into specialist categories.
A Check Worth Running
Ignore the headline number and look for categories.
Does it reach the main mechanics: Megaways, cluster pays, hold-and-win and crash
Does it include a high-volatility specialist and a studio known for low-variance content
Does it carry in-house originals, which are independent of the consolidation and usually better priced
A lobby answering yes to all three has genuine range whatever the count says, and who supplies a lobby determines everything downstream of it.
Confirm what is legal where you live, keep stakes within a set budget, and play only if you are of legal age, since KYC or AML checks may apply. Responsible gambling is unaffected by provider range, and a lobby with twenty studios offers twenty routes to the same house edge.


Disclaimer: The information here is provided for general purposes only and is not legal, tax, investment, or financial advice. Provider rosters, ownership structures and catalogue composition vary by operator and change over time, so confirm current details on the platform itself. Betting carries risk, and rules vary by country, so check the law where you live. Please gamble responsibly, within your means, and only if you are of legal age.

UK monthly gross domestic product grew by 0.4% in July 2026, extending the 0.3% expansion recorded in June after no growth in May. The reading stands out for the strength of computer programming and related services, while the Office for National Statistics said high-turnover businesses showed activity linked to artificial intelligence and cloud computing.
The ONS’s assessment adds context to the July services performance, but it does not provide a measured causal explanation for the monthly result: the agency did not quantify how much AI- or cloud-related activity added to business turnover or GDP.
Data Snapshot
MetricCurrentPreviousChangePeriodAs ofSourceReal gross domestic product (GDP)0.4%0.4% in the three months to June 2026; 0.6% in the three months to May 2026grew by 0.4% in the three months to July 2026, compared with the three months to April 2026three months to July 20262026-09-11Office for National StatisticsMonthly GDP0.4%0.3% in June 2026; no growth in May 2026grew by 0.4% in July 2026July 20262026-09-11Office for National StatisticsServices output0.6%0.5% in the three months to June 2026; 0.6% in the three months to May 2026grew by 0.6% in the three months to July 2026, compared with the three months to April 2026three months to July 20262026-09-11Office for National StatisticsComputer programming, consultancy and related activities3.5%—grew by 3.5% in July 2026July 20262026-09-11Office for National StatisticsInformation and communication2.4%—grew by 2.4% in July 2026July 20262026-09-11Office for National Statistics
July GDP growth followed 0.3% in June
Monthly GDP growth of 0.4% in July 2026 followed 0.3% growth in June 2026 and no growth in May 2026, according to the ONS release published on 11 September.
The monthly measure should not be read as directly equivalent to the rolling three-month series. Real GDP grew by 0.4% in the three months to July 2026 compared with the three months to April 2026; the corresponding growth rates were 0.4% in the three months to June 2026 and 0.6% in the three months to May 2026.
Computer programming added 0.12 percentage points to July GDP
Computer programming, consultancy and related activities grew by 3.5% in July 2026. The category contributed 0.14 percentage points to services output and 0.12 percentage points to real GDP during the month.
Information and communication grew by 2.4% in July 2026 and was the second-largest positive contributor to services sector output. Those figures identify a concrete source of momentum within the services economy, rather than attributing the full GDP result to a single technology-related activity.
AI and cloud activity in the services data
The ONS said there was evidence that many of the businesses with the largest turnover in July 2026 were involved in artificial-intelligence and cloud-computing-related activities, although it could not quantify their exact impact on turnover.
Services output grew by 0.6% in the three months to July 2026 and was the main contributor to GDP growth over that period. The rate was 0.5% in the three months to June 2026 and 0.6% in the three months to May 2026.
July’s figures are an early estimate, with the full time series open to revision in the next release on 15 October 2026.
Disclaimer: This article is provided for informational purposes only. It is not offered or intended to be used as legal, tax, investment, financial, or other advice.

The open ECDSA.Fail project published its preprint on September 9. It describes a smaller logical-circuit benchmark for a secp256k1 operation used in Shor’s algorithm, which is relevant to Bitcoin and Ethereum signatures.
At the July 26 cutoff, the best submission used 1,151 logical qubits and 1,299,453 average executed Toffoli gates. Its Q×T score was about 1.496 billion.
That score was more than 50% below Google Quantum AI’s published benchmark, according to the ECDSA.Fail preprint; the accounting conventions differed.
The result measures an improvement in the circuit itself. It does not provide a new estimate of the hardware required for a complete attack on Bitcoin.
ECDSA.Fail cuts the secp256k1 circuit benchmark
The project focused on point-addition circuits for secp256k1, the elliptic curve used by Bitcoin and Ethereum signatures. Point addition is a component of the computation required when applying Shor’s algorithm to the underlying elliptic-curve cryptography.
Its reported best circuit combined 1,151 logical qubits with 1,299,453 average executed Toffoli gates. Multiplying the project’s qubit and gate measures produced the roughly 1.496 billion Q×T score, a compact benchmark intended to track the circuit-resource trade-off.
Logical qubits are an abstraction used in quantum computing resource estimates; they are not the same as physical qubits in a machine. The distinction matters here because converting a logical design into a fault-tolerant physical implementation requires error correction, a cost not included in the new benchmark.
More than 100 contributors drove an 86.1% decline
The public ECDSA.Fail challenge ran for roughly two months and involved more than 100 participants and AI agents, according to an Eigen Labs account of the effort.
Its Q×T benchmark fell from 10.75 billion by 86.1%. The work used an open, collaborative process to optimize a cryptographic circuit relevant to long-term quantum risk to digital-asset signatures.
That result concerns the project’s own benchmark and baseline, not a matching percentage reduction in the resources required for an operational quantum attack. It is also an optimized component of a full quantum-attack circuit, rather than an end-to-end cryptanalytic system.
Official ECDSA.Fail research visual accompanying the benchmark announcement. — Source: Eigen Labs
Google’s estimate used tens of millions of Toffoli gates
Google Quantum AI’s March 2026 secp256k1 estimates were about 1,200 logical qubits and 90 million Toffoli gates, or 1,450 logical qubits and 70 million Toffoli gates, with fewer than 500,000 physical qubits under its stated superconducting-hardware assumptions. The figures come from Google’s resource-estimate paper.
ECDSA.Fail’s 1.299 million average executed Toffoli gates can look far lower, but its headline Q×T score combines logical qubits and executed Toffoli gates for the point-addition circuit. Google reported alternative resource estimates with explicit physical-qubit assumptions, so the two results differ in accounting and measurement scope and are not formally interchangeable. A lower Q×T score is consequently not a direct replacement for Google’s full set of figures.
One Shor-attack component, not a complete Bitcoin attack
From the challenge’s original 10.75 billion Q×T, the logical-circuit benchmark for secp256k1 point addition fell to approximately 1.496 billion.
That is a result for one major component of a prospective fault-tolerant Shor attack, not for a complete attack implementation. The ECDSA.Fail preprint does not include the physical error-correction costs needed to determine the number of real qubits a machine would require.
The comparison also has a defined limit: CoinDesk reported that the researchers acknowledge their benchmark and Google’s result are not formally like-for-like. Any effect on practical risk timelines would still depend on the unmeasured full-attack work and fault-tolerant hardware capable of running it.
Disclaimer: This article is provided for informational purposes only. It is not offered or intended to be used as legal, tax, investment, financial, or other advice.

Uniswap Labs launched StablePair Hook on September 10, 2026, introducing a dynamic-fee mechanism for stablecoin trading on Uniswap v4. The initial Ethereum mainnet rollout is limited to USDC/USDG and USDC/USDT pools, a narrow deployment in a segment where Uniswap says it handled $43.4 billion in stablecoin-to-stablecoin swaps during the second quarter.
The hook changes fees based on a pool’s relationship to a reference rate and on whether a trade moves the pool toward or away from that rate. The design puts the focus on a common issue for stablecoin markets: pricing can shift even when the assets being traded are intended to maintain closely aligned values.
StablePair Hook launches in two Ethereum pools
StablePair Hook is built for Uniswap v4, whose hook framework allows pool-level logic to be applied to trading activity. Uniswap Labs’ first pools pair USDC with USDG and USDC with USDT on Ethereum mainnet.
The two specified pools are the full scope of the launch at the outset: the mechanism is not being used across every stablecoin market available through the protocol, nor does the launch represent a network-wide conversion of stablecoin pools to the new fee model.
Although the initial configuration provides a live implementation of the model, it does not show how broadly the model will be adopted: the deployment announcement names USDC, USDG and USDT, and Uniswap Labs did not provide a timetable for further StablePair pools in the supplied announcement.
The fee mechanism rewards rate-restoring trades
StablePair sets fees according to the extent of a pool’s deviation from a reference rate and the direction of an incoming trade, according to The Defiant’s report on the launch. In practical terms, trades that help restore the reference relationship are treated differently from trades that push the pool further from it.
For corrective trades made when a pool is outside its defined price band, the fee declines block by block through a Dutch-auction mechanism. That is a different approach from applying one static fee to every swap regardless of market conditions or trade direction.
The mechanism is intended to make rate-restoring flow more economically attractive when a stablecoin pair has moved outside the band. Its actual effect on liquidity, execution and trading behavior will depend on how the two initial pools operate after launch; no performance results were included in the supplied material.
Official StablePair Hook illustration showing dynamic pricing curves for stable pairs. — Source: Uniswap Labs
A $43.4 billion quarterly stablecoin-swap market
Uniswap reported $43.4 billion in stablecoin-to-stablecoin swap volume during Q2 2026. The company said that volume was greater than the combined volume of the next three onchain venues.
StablePair is a Uniswap v4 dynamic-fee mechanism for stablecoin pairs, linking the rollout to that existing flow of trading activity.
The figures are company-reported, and the supplied material does not independently verify them. It also does not provide a separate Q2 volume figure for the newly launched pairs.
Pool creation is initially controlled by Uniswap Labs
Access to the hook is restricted in the early stage. Only Uniswap Labs can currently create pools using StablePair Hook, The Defiant reported, meaning other parties cannot yet independently deploy a StablePair pool under the current arrangement.
The restriction also distinguishes the launch from a fully open pool template. It gives Uniswap Labs control over the initial set of deployments while the company tests the mechanism in USDC/USDG and USDC/USDT on Ethereum.
Fee parameters and the hook’s logic can, however, be adjusted through Uniswap governance. That leaves the economic settings and functioning of StablePair subject to a governance route even as creation of new StablePair pools remains centrally controlled for now.
Disclaimer: This article is provided for informational purposes only. It is not offered or intended to be used as legal, tax, investment, financial, or other advice.