番茄蛋炒面

I studied everything very thoroughly on TermMax—interest rates, tenors, liquidation lines. I checked every number, and in the end I still lost 41,000 u. The loss came from one thing I didn’t research: how long my own money would be unused.

At the time, I bought a 250,000 u, 6-month FT with a locked 13% rate. In the fifth month, my family needed the money. I couldn’t redeem the FT early—only transfer it. But for an FT that isn’t yet due, the transfer has to be discounted. I ended up getting about 8% less than face value, and 20,000 u was gone.

What’s even more painful is that when I looked back, the market interest rate during the same period had already risen to 18%. I was locked at 13%, so for the remaining one year I earned about another 21,000 u less. Add them up: 41,000 u.

I spent two weeks researching this project, but it took zero minutes to research “when I would need this money.” With fixed-income products, if you pick the wrong tenor, all the in-depth research in the world is useless.

Now, whenever I buy anything with a fixed term, my first step is to write down “how long I won’t need this money.” If I can’t write it out, I don’t buy. Interest rates are only the second part of the research—the tenor is the first.

@TermMax #TermMax

98000u is the extra money I earned in DUSK staking by "choosing the right validators"—with the same staking, choosing differently makes a difference of this amount after three years.

Research: I ranked DUSK validators by two metrics—commission rate and missed-block rate (stability). Top validators have around a 5% commission, while bottom validators may pay commissions of 20% or more. Some validators also have high missed-block rates, so part of the staking rewards gets slashed/penalized. I staked 300000u with the validator that had the lowest commission and the lowest missed-block rate.

After three years, compared to staking the same funds with "high commission but high missed blocks" tail validators, my net return was about 98000u higher (the commission difference minus the penalty difference—about a 10% gap per year, compounded over three years).

This money doesn’t require any market timing—just one choice. But it’s often overlooked, because most people only look at "what the annualized return is," not "who they are staking with." In the staking track, choosing the validator is part of your returns: for the same protocol, picking the right validator versus the wrong one leads to two completely different annualized yields.

@Dusk $DUSK #dusk

15000u is the money I made on TermMax by tracking whales’ wallets—not people, but data.

Method: I watched on-chain actions of a few top liquidity addresses on TermMax—what markets they bought FTs in, when they added to the Vault, and when they withdrew. I wasn’t copy-trading; I was logging their behavior patterns, then using my own judgment to decide whether to follow. In six months, I caught three high-quality signals: once they concentrated their add to a certain-term FT—I followed and captured the tail end of the interest rate increase; once they exited a market early—I reduced my position alongside them to avoid a round of interest rate pullbacks. Net profit over six months: 15000u.

This money proves one thing: on-chain funds don’t lie—KOLs’ mouths might, but whales’ on-chain behavior doesn’t. Follow the data, not the people. What you earn is “behavioral science” money, not “persona” money. And it echoes a lesson from earlier when I lost 54000u following a KOL: it’s better to watch what people do than to listen to what they say.

@TermMax #TermMax

54000u is the money I lost on DUSK following a KOL—three signals, two losses and one win; the net loss comes to this number.

First, he said “a bottom structure has appeared,” and I bought 20,000u with him, making 8,000u—felt great. Second, he said “there will be a wave before and after the mainnet is activated,” so I added to 30,000u, and the result was: the good news was cashed in the very day and then it dropped. I panicked and cut out, losing 32,000u. Third, he said “delisting is a case of overreaction—buy more the more it falls,” so I went in again with 20,000u. It kept drifting downward; in the end I stopped out, losing 30,000u. In total across the three trades: 8,000 minus 32,000 minus 30,000 equals a net loss of 54,000u.

But this loss also made me see a structural problem in DUSK’s information environment: the project’s information supply heavily depends on external interpretation—there are few official announcements, the pace is slow, and key milestones (Dusk Trade, DLT-TSS) have been hanging unresolved for a long time. So during the vacuum period, the KOL’s interpretation becomes the market’s only “compass.” The scarcer the information, the more weight call-signals carry; the more followers there are, the more volatility gets amplified. DUSK’s high volatility is partly provided by liquidity and partly created by the information vacuum. This is the project’s responsibility, not market randomness.

My current rule is: anyone’s viewpoint can only be used as research input, never as a trading instruction. You can borrow ideas, but you can’t outsource decisions—especially in projects with an information vacuum. Outsourcing decisions means handing your judgment to someone who isn’t financially aligned with you.

@Dusk $DUSK #dusk

82000u is the biggest amount of money I “lost” on TermMax—note the quotation marks, because I never actually put this money in; it’s the portion I failed to earn.

Three missed opportunities. First: the pre-market auction in March, with a scale of 6 million u. The chance to participate at a discounted rate was right there. I hesitated for a week because “the details weren’t published, and the valuation couldn’t be calculated accurately”—afterward, based on how the auction was settled, I can estimate that I missed about 17,000u. Second: the deeply discounted FT when market sentiment hit a low point. I observed that discount in the market, did the calculations, and ultimately didn’t buy—based on how that batch of FTs redeemed at maturity, I estimate I missed about 50,000u. Third: the RLUSD deposit phase. I studied data from the Leaderboard, judged that the official team was buying deposits, but I didn’t switch positions because it felt like too much hassle—missed about 15,000u. Total: 82,000u.

These three have one thing in common: it wasn’t that I couldn’t understand. It’s that I understood it but didn’t take action. The value of analysis is in decision-making, not in reading. I got the direction right quite a few times, but the number of times I actually acted was too few—the difference in between is the missed opportunity ledger.

I pasted this account on the first page of my memo, and next to it I wrote one line: “The distance from analysis to decisions is the distance from profit to cash in hand.” Next time I finish tallying the books, first ask myself, “Did I act?”

@TermMax #TermMax

When you pull up the list of DUSK holders, you’ll see a glaring number: the first five addresses together hold 66% of the coins. The top two each account for more than 20%.

What does this mean? Plainly put: the coin’s “public sentiment” is determined by five wallets. If they stake, the staking rate rises; if they don’t move, the market stays put. Even the bullish or bearish chatter among retail traders carries very little weight in front of these five addresses.

There’s a detail worth thinking about: DUSK’s outward story is about compliance, institutions, and securitization—an institutional- and regulator-facing narrative. But what it pairs with is a highly concentrated token structure. The story says “let more people participate,” while the structure says “a few people hold it.” When these two lines clash, the latter usually wins.

I’m not saying this must be a bad thing. High concentration sometimes means someone is genuinely holding and doing things long-term. But it changes how you should watch the chart: when this coin’s price makes unusual moves, many times it’s not market behavior—it’s wallet behavior. So now when I monitor it, besides the price, I also watch what the top addresses are doing. If their holding proportions start to fall, that means the tokens are being dispersed. If 66% stays locked up for the long term, it suggests the structure hasn’t changed—then any price volatility is just the brush strokes of the big players.

@Dusk $DUSK #dusk

63000u is the principal I used to do interest rate arbitrage between two chains on TermMax—moved it 13 times in a month. I made less money than I expected, but I learned more than I expected.

The operations aren’t complicated: TermMax deployed 10 chains, and the FT markets each price independently. In June, I noticed that the USDC FT prices for the same maturity differed by 1.5% between Ethereum and Arbitrum. Buy on the cheaper side, cross back to the more expensive market, and in theory it’s a near risk-free price spread. In practice, it’s a different story: cross-chain bridge fees, the Gas on each chain, and—most importantly—the FT market liquidity is thin. Splitting 63000u into 13 entries means each trade nudges the price up a bit, so the actual spread I capture shrinks to only 1.1%.

Total for the month: gross profit about 945u. After deducting bridge fees and Gas of about 210u, net profit is about 735u—total return 1.2%, annualized roughly 14%. It’s pretty similar to just putting the money into a single-chain FT market to earn fixed yield, but I took on the bridge risk and the effort of 13 separate operations.

Conclusion: the real room for interest rate arbitrage is approximately equal to how fragmented the market is. 10 chains means 10 pricing points—sounds like lots of opportunities, but each opportunity gets whittled down heavily by bridge fees, Gas, and the thin order book. What arbitration earns is "money others don’t want to move"—for my scale of principal, moving it is worse than not moving it.

In the end I stopped: I moved the 63000u back into a single market for fixed yield. The yield is the same, and the risk is about half. The multi-chain story is exciting, but the multi-chain costs are very real.

@TermMax #TermMax

I researched DUSK’s fee economics and ended up with an interesting conclusion: the network’s fees aren’t a cost—they’re part of the newly minted supply.

First, the mechanism: Gas is priced in LUX, and 1 DUSK equals 1 billion LUX. Transaction fees are neither burned nor sent back to the protocol; instead, they’re added into the block reward for redistribution—70% to block producers, 10% to the development fund, and 5% each to two committees. In plain language: the fee you pay ultimately becomes a reward paid to stakers.

This design is the opposite of Ethereum: ETH burns fees, acting as a deflationary mechanism, while DUSK sends its fees back as a “reward amplifier.” Which is better depends on how you view it. For stakers, it’s an additional income stream—more transactions mean higher staking returns. For pure holders, it’s an implicit tax: your share gets diluted by the extra rewards.

I ran calculations using a hypothetical scenario in the DuskEVM assumptions: if, after Dusk Trade goes live, its daily transaction volume reaches one-tenth of a mainstream L1, then after fees are folded into rewards, the actual staking effective annualized return would be noticeably higher than the advertised 12%. Conversely, if the chain keeps seeing little to no activity, stakers only receive the emissions themselves. So the staking returns curve is, in essence, another way of writing the network usage curve.

Conclusion: DUSK’s fee design is unique within L1, but it’s not a free lunch—it ties “network growth” directly to “staking returns” on the same curve. If you believe in the staking annualized return, you’re really betting on the transaction volume of Dusk Trade; these two things are, in practice, one and the same.

@Dusk $DUSK #dusk

Last night I went through Dusk’s documentation, trying to figure out how “selective disclosure” is actually triggered in practice. The docs sounded very polished: default privacy, disclose on demand, and regulators can verify via licensed records. But when I got to the chapter on permission management, I stopped. The document only says: “the granting party holds the viewing key.” It doesn’t say whether that authorization can be revoked, whether it has an expiration time, or who records the authorization activity.

So I went and tested it on the testnet: I created an identity and granted view permission to a simulated auditor. The workflow worked—four steps total: generate the authorization key, submit the transaction, the other party verifies, and that’s it. It took about two minutes. But when I came back to check the authorization records, I found that there’s no documentation about revoking an authorization, and the testnet console also has no corresponding entry. It’s a toggle switch that has only “on” and no “off.” In a compliance scenario, that’s dangerous—once auditing permission is granted, if it can’t be revoked, then “disclosure on demand” effectively becomes “one-time authorization, permanently visible.”

I then went to GitHub and looked up the contract code for the authorization module. I searched for “revoke” and “expire,” and the results were empty. Maybe I wasn’t looking at the right repository branch, or maybe these features are still planned. But as a user, I can only judge based on the existing documentation and existing code.

Later I realized something: this isn’t a Dusk-only problem. It’s a shared lesson along the “programmable privacy” track—granting permissions is cryptography, while revoking permissions is governance. Granting can be handled with math; revocation can only be handled with process. And right now, the documentation has no process.

So my testnet wallet has a few hundred DUSK, and I haven’t moved any mainnet assets—for now. I’ll wait until the official team writes authorization management (revocation, expiration, audit logs) into the documentation and code, and then I’ll put the real money in. Until then, I admit that “selective disclosure” is a good design, but good design can’t be only an “on” button with no “off” button.

@Dusk $DUSK #dusk

30000u is the unrealized loss I'm currently carrying on DUSK. I’m writing this not to sell sympathy, but to review the mistakes I made.

In mid-January, it surged nearly twofold over two days. I couldn’t resist and chased into the position at 0.09 for 90,000u—only to buy right at the emotional peak. Now the price is 0.06, and my unrealized loss is about 33%, which comes to exactly 30,000u. Looking back, I violated a rule I had set for myself three times: "Don’t chase on a single day’s big surge".

My review has three lessons. First, a surge driven by news moves you to buy based on emotion rather than value. The January jump of 422% was driven by mainnet activation plus partner news—if I had studied that before the announcements, there would have been no need to chase. Second, any averaging-down / adding during a pullback needs logical anchors. I added once in February at 0.07; it’s not the worst entry, but it’s not really a position that fits my plan either. Third, during the drawdown, I did one thing right: I didn’t cut at the absolute low, and I didn’t add leverage.

What I’m doing now: I’m not in a rush to get back to break-even on the 30,000u loss. I’m placing my stop-loss at 0.045 (the key support I researched). If it breaks below, I’ll exit. If not, I’ll hold and earn the 12% collateral/pledge yield—using time to create room.

@Dusk $DUSK #dusk

The DUSK with financing of $8 million in 2018 and a unit price of 0.0404 took a full six years to get the mainnet running on January 7, 2025. Back when I was earlier, such a project would have been blacklisted—"If it hasn’t taken off in 6 years, it definitely won’t." Only after I mapped out its timeline did I realize that "slow" itself is a filter.

So what did it do in those six years? It took a 10% equity stake in the Dutch exchange NPEX in 2020; released an economic model audit report in 2024; launched its first irreversible block on the mainnet in January 2025; upgraded DuskDS in December; and officially activated the mainnet in January 2026, with DuskEVM going live, Hedger private transactions becoming available, and NPEX’s €300M+ securities tokens poised to be issued.

Compared with those “projects” that “go live” in just three months, a project that takes six years to turn in its assignment has, at minimum, eliminated 90% of the rug-pull risk—because scams can’t afford to last six years.

Of course, being slow also means patience is the threshold: at the current price of $0.06, the 422% frenzy spike in January has already passed, but the 94% drawdown still lies ahead. My approach: treat DUSK as an “annual observation position”—don’t watch the daily chart, just check whether quarterly milestones are being兑现.

@Dusk $DUSK #dusk

5,000 U——This is the cost I paid during my BABY research because of “methodological mistakes.” Today I write these three errors in full. It matters more than the gains.

First mistake: Look at the conclusion before the evidence. Before I entered in June, I spent three days browsing KOL analyses (all bullish) and only then looked up sources. I went into the research with the preset of “it’s doing well,” filtering out all unfavorable information—until I was down 20%, forcing me to face the risks. The correct order is: first look at the whitepaper and on-chain data, then see what others are saying. Assumptions decide bias, and data decides judgment.

Second mistake: Use “price action” to validate the “fundamentals.” In July, BABY went sideways for a month. For a while I doubted my research—if the price doesn’t move, does that mean the fundamentals are weak? Later I realized: the sideways action actually means the market hasn’t priced the project yet (FDV/TVL at 0.039x), not that the fundamentals are bad. Using price to verify fundamentals is like using noise to test a signal. Fundamental validation can only come from on-chain data, protocol revenue, and the progress of catalysts—those three were improving at the time.

Third mistake: Treat “research once” as “research complete.” After I finished that deep dive in July, I felt that everything was settled. I didn’t update my information for the following two weeks—result: I almost missed the verifier’s signature anomaly event. Later I established a weekly loop: research isn’t a one-time action; it’s an ongoing system. Information is flowing, and research must be continuous.

Together, these three mistakes cost me about 5,000 U (one chase, one almost cutting at the low, and one exposure to risk)—but this tuition bought me three methodological upgrades: evidence comes first, fundamentals independent of price, and research continuous by design. Since then, my investment decisions have never again had any “gut-feel” component.

5,000 U is the most worthwhile tuition I’ve paid in this market. Methods are more expensive than returns—and they never depreciate.

@BabylonLabs_io $BABY #baby

I staked 20,000 U on BABY, then spent a month answering a more fundamental question: where exactly is this project’s ceiling? The answer is bigger than I expected, but the path to reach it is narrower than I imagined.

The ceiling is determined by the product of three variables: the size of BTC staking × the revenue per connected chain × the number of chains connected. Let’s break it down:

First variable: the BTC staking size. Babylon is currently locking 52,000+ BTC (about $3.9B). There are 21 million BTC in total supply. Of that, the portion that can be staked (excluding exchange holdings, cold-wallet dormant assets, etc.) conservatively still amounts to several million coins. If Babylon can increase staking penetration from 0.25% to 1%—that’s 200,000 BTC—there’s 5x room for TVL growth.

Second variable: revenue per chain. Osmosis has already agreed to share 50% of Bitcoin LST transaction fees with stakers. If this "revenue-sharing" becomes a standard clause for BSN, then every connected chain would effectively be working to pay stakers. The higher the chain revenue, the higher the BTC staking yields—and the more attractive staking becomes. That’s the flywheel.

Third variable: the number of connected chains. This is the narrowest bottleneck: for each additional chain, you have to go through the full governance proposal → software upgrade → integration testing process, measured in months. After Phase 3 multi-staking is rolled out, the same batch of BTC can serve multiple chains at the same time—which will change the entire growth curve, turning "linear chain onboarding" into "exponential reuse."

After running the numbers, my conclusion is this: the ceiling depends on how quickly Phase 3 gets pushed forward. If multi-staking launches on schedule, Babylon would evolve from a "staking protocol" into a "BTC security wholesale marketplace"—that’s when the real value uplift of 20,000 U truly happens.

Of course, the prerequisite is that the lessons from the failures of the past few years don’t get repeated—people may have connected, but nobody actually uses it. So in the end there’s one more variable I didn’t calculate, and it’s the most important: demand. Whether protocol revenue can truly take off ultimately depends on the number of real users on each chain.

@BabylonLabs_io $BABY #baby

The principal of 30,000 U took me four months to figure out how to allocate. Today I’m writing out this whole configuration thought process in full—this is currently the draft that underpins all my actions on BABY.

First, the result: 30,000 U is split into three parts—10,000 U in spot BABY, 10,000 U in staked BABY, and 10,000 U in a reserve fund. After four months, total assets are about 34,830 U, with earnings of 4,830 U (+16%), of which the staked cashflow contribution is about 650 U.

Why split it like this? Spot takes on “price elasticity”: if the price is revalued (Aave passes, TVL grows), spot captures the full upside and can be reduced at any time. Staking takes on “cashflow”: 4–6 U comes in every day; reinvesting compounds the amount—so even if the price doesn’t rise, you can still accumulate chips. The reserve fund takes on “opportunity cost”: it prevents me from missing out (I can add anytime) without going all-in (if things drop, there’s still ammo). The risk exposure of the three roles is completely different and they don’t interfere with each other.

Over these four months, this configuration was stress-tested twice. The first was in June when it fell to 0.007: spot was down about 25% on paper, but staking rewards still came through as expected, and the reserve fund kept me from panic-selling at the bottom. The second was last month when it briefly spiked upward: when it was up 20% on paper, I wanted to reduce, but according to the rules—no cutting when there’s no substantive positive catalyst that triggers a reduction—I held. The next day when it pulled back, I didn’t regret it.

The key of the configuration isn’t the act of “splitting into three parts,” but that each part has its own rules. Spot reduces only when trigger conditions are met; staking only looks at rewards and validators; the reserve fund moves only in two situations (when it falls into the DCA target range, or when a catalyst lands). Each chunk of money has a clear job to do, so I don’t have to make emotional, day-to-day decisions for the entire position.

30,000 U isn’t a huge amount, but this allocation turns it into a system that doesn’t keep me up at night.

@BabylonLabs_io $BABY #baby

Write a complete retrospective: the four months I spent on BABY—from entry, getting stuck, research, averaging up, to where I am now. Written for people who want to copy homework, and also for myself three months from now.

Entry (early June): The day the Upbit listing news dropped, I chased the price on the exchange at a cost of 0.0092. Bad luck— I entered at the stage high. Then it fell to 0.007, and I was down more than 20%. The first trade was a failed lesson.

Getting trapped (mid-June to early July): The two most painful weeks. I did two things: first, I pledged half of my position—using the “unlock in two days” mechanism to force myself to stay calm; second, I started reading documentation and proposals to understand the underlying logic: EOTS, dual consensus, and where the yield comes from. Research replaced anxiety.

Turning point (early July): After the news of Binance Labs’ strategic investment came out, I started focusing on fundamentals instead of price: TVL stayed stable at 52,000 BTC, staking rewards hit on schedule every day, the Aave V4 proposal was in progress, and Osmosis went through a BSN proposal. In mid-July, I added more to bring my total BABY position to 20,000 U, averaging down/up the cost to 0.0095.

Now (August): Price is 0.0113. Total assets are about 34,830 U, with cumulative gains of about 4,830 U (+16%). Staking rewards accumulated to 650 U, all of which was reinvested. Daily cash flow is 4–6 U. Position structure: 10,000 U in spot, 10,000 U pledged, 10,000 U in reserve cash, and 3,000 U per month for DCA.

If I could do it again, I would only change one thing: I wouldn’t chase the high—I’d buy in batches the first time. Everything else—staking, research, averaging up, and DCA—I wouldn’t change, because they were proven: when you’re trapped, the most effective move isn’t watching the charts; it’s doing research.

@BabylonLabs_io $BABY #baby

I realized a detail that many people haven’t noticed: in the BTCFi space, Babylon is an early mover—but being first doesn’t mean you win. What truly determines the winner is the “last mile”—liquidity.

Over the past year, BTCFi’s TVL fell from a peak of $9.1B to about $3.3B, a drop of 64%. If we look at Babylon’s own TVL, it fell from $5.6B to $3.9B—down 30% as well. But this isn’t Babylon’s problem; it’s a problem with the sector. BTCFi’s growth logic is “staking → liquidity → applications,” and the liquidity leg hasn’t been fully connected yet.

Staking solves “BTC can earn yield,” but that yield is locked up and can’t be used. Liquidity solves “the yield can be taken out and put to work”—liquid staking tokens like stBTC and LBTC enable staked BTC to be used as collateral in DeFi, provide liquidity, and be traded.

Babylon’s response is stBTC (to be launched in Q1 2026) and the TBV system. Lombard’s LBTC has already proven this model: $1.5B TVL, 260,000 users, 60% market share, and integration with 70+ protocols. But LBTC belongs to Lombard, not Babylon—Babylon needs its own liquidity flywheel.

Once this flywheel is in place, the value proposition changes: BTC stakers won’t get “paper gains,” but “usable yield.” Usable yield will attract more people to stake, more staking will bring more liquidity, and more liquidity will enable more applications. This is a positive feedback loop.

But in the other direction, if liquidity keeps failing to connect, TVL growth will stall—because assets that are locked in and can’t be withdrawn have limited growth potential.

So the “last mile” determines BTCFi’s endgame. Babylon’s advantage is being early—52,000 BTC staked, and 55% of the sector’s share. The downside is that the liquidity flywheel is still being built. What I’m watching is: when will stBTC’s DeFi integrations (lending markets, DEXs, derivatives) exceed LBTC’s 70 protocols—that will be the signal that the “last mile” has been connected.

@BabylonLabs_io $BABY #baby

When I studied Babylon’s staking structure, I found a point that very few people distinguish: it actually has two completely different staking tracks, hidden inside the single word “staking.”

The first track is BTC staking: you lock your BTC into a Taproot script and delegate it to the Finality Provider (FP), to provide Bitcoin finality for the chains being connected. The yield comes from the secured chain—for example, after Osmosis integration, 50% of the Bitcoin LST transaction fees are distributed to BTC stakers. BTC always stays on the Bitcoin chain, constrained by Bitcoin Script.

The second track is BABY staking: delegate BABY to 100 CometBFT validators to produce blocks on the Babylon Genesis chain itself and maintain its liveness. The reward is BABY inflation, with an annualized rate of roughly 15–18%. This is standard Cosmos-style staking.

The risk models for these two tracks are entirely different. The risk of BTC staking is EOTS slashing—if the FP behaves maliciously, your BTC will truly be slashed. The risk of BABY staking is validator double-signing or going offline and being jailed—the one being slashed is BABY, not BTC. The sources of yield are also different: one relies on income from an external chain, and the other relies on inflation from the chain itself.

Most people confuse these two tracks, leading to two kinds of mistakes: believing that staking BABY lets you share in the BTC security “dividends” (which actually requires staking BTC), or believing that BTC staking can achieve the high annualized returns of BABY staking (when in reality BTC staking yield depends on how many chains it is connected to).

What I keep focusing on is the difference in yield between these two tracks. When the real annualized yield of BTC staking starts to exceed that of BABY staking, it means external-chain revenue has already become the main source of returns—at which point Babylon no longer relies on inflation subsidies.

@BabylonLabs_io $BABY #baby

After finishing BABY’s unlocked calendar, I found something that runs counter to most people’s intuition—the most persistent supply pressure may not come from any specific unlock event.

At first, I assumed BABY’s supply pressure mainly came from news about “X million tokens unlocked on a certain day of a certain month,” which the market focuses on. But after reviewing Tokenomist’s data, I found that early private investors account for 30.5%, the team 15%, and advisors 3.5%. Starting in May 2026, they’ll be linearly unlocked over 36 months, ending in 2029. It’s not a one-time spike—it’s a steady release of 130–190 million tokens every month for three years, regardless of market conditions.

The next known unlock is August 10, at about 136 million tokens (total value about $1.5M), which is less than 2% of circulating supply. The single-event size isn’t huge—but when you factor in 8% annual inflation (4% to BTC stakers, 4% to BABY stakers), BABY’s average monthly added supply could be in the range of 150–250 million tokens, lasting until 2029.

This isn’t FUD—it’s a known timetable. Whether the fee model introduced in Phase 3 and the BSN demand can generate enough on-chain activity to absorb that supply is the key question that needs to be tracked.

Babylon’s tokenomics may not be meant to create supply shocks—it may be about designing a long-term release curve, giving the market and the protocol enough time to match supply and demand.

So what I keep watching isn’t “how many tokens were unlocked in a given month”—but the trend in the gap between average monthly added supply and average monthly on-chain consumption (Gas fees + BSN payments + the staking needed for governance participation).

@BabylonLabs_io $BABY #baby

Sometimes I wonder whether BTC delegators have the freedom to choose FP—but do most users really have the information needed to make that choice?

At first, I thought the FP list is sorted by APY, so users just pick one, much like choosing a validator. After reading the relevant materials on FPs, I found that what users can mainly see when choosing an FP is APY, the name, and the logo. What they can’t see are things like the FP’s historical uptime, voting participation rate, how frequently its commission is adjusted, and whether it has been slashed. These details determine whether an FP is reliable when the market is volatile—but they aren’t shown in the staking UI.

Choosing an FP is not like choosing a validator—on ETH, third-party tools like rated.network and beaconcha.in provide complete validator scoring and tracking. For BTC staking, there currently isn’t anything comparable in terms of publicly available standards for evaluating FPs.

Babylon may be doing quite well at the FP layer—Fps need to stake BABY in order to register, which is, in itself, a filtering threshold. But the information environment users have when selecting an FP may still be insufficient to support the assumption that “users will rationally choose reliable FPs.”

So what I’m really focused on may not be the total number of FPs or their distribution—but whether the operational transparency standards for FPs will improve as the mainnet runs longer.

@BabylonLabs_io $BABY #baby

Trustless isn’t “I don’t need to trust anyone.”

It can’t.
No one can.

TBV lowers trust in the Bitcoin scripting layer to the minimum—withdrawal conditions are hard-coded, guaranteed by cryptography. But on that basis:

The Covenant committee will require you to trust that 14/20 of the members do no harm. The liquidation path requires you to trust that the liquidator will execute it. The oracle requires you to trust that the quotes haven’t been manipulated.

Trust hasn’t been eliminated. It has been broken into pieces that you can evaluate separately.

The Bitcoin scripting layer is trustless. The committee layer is trust-reduced. The liquidation layer is trust-delegated.

With all three layers written together, it’s more honest than just saying “trustless.”

@BabylonLabs_io $BABY #baby