Two thirds of my strategy's profit came from one year. That is not an edge, it is an event
Two thirds of my strategy's profit came from one year. That is not an edge, it is an event. The headline number was +134.12 USDT from short breakouts. Then I split it by year. 2022 - the year of the LUNA collapse and the FTX failure - produced 88.54 of it. Strip that year out and four years of trading returned 45.59 USDT, most of it from 2025. 2026 so far is flat. WHAT IT MEANS This is tail capture, not income. Realistically it does nothing in calm markets and pays during sharp selloffs. Expecting it to earn every month is how people quit a strategy right before it works. Part of a public experiment: 5,063 USDT to 1,000,000 USDT, every result published - including the ones that make me look bad. Run on $BTC $LTC #Binance #Futures #Backtesting #TradingBot #CryptoData
MAGIC tăng +83.33% — tôi đã thử cả hai cách giao dịch mã này
CÁC MÃ TĂNG MẠNH NHẤT HÔM NAY MAGIC: +83.33% KAIA: +60.74% BAT: +35.35% RLC: +28.23% STRK: +24.53% Mọi tài khoản trên ứng dụng này đều hiển thị danh sách này. Đây là phần họ bỏ qua. Tôi đã kiểm tra bằng dữ liệu quá khứ cả hai cách giao dịch một biến động như mức tăng +83.33% của MAGIC: mua khi giá phá vỡ ngưỡng kháng cự và đặt cược ngược chiều với giả định giá sẽ quay đầu. Cả hai đều thua lỗ. Không phải chỉ thua chút ít — mức lỗ xấp xỉ chi phí giao dịch, 0.240% cho mỗi lượt mua rồi bán, đúng như những gì ta thấy ở một tín hiệu không chứa thông tin hữu ích khi tính phí giao dịch. Một biến động lớn trong 24 giờ cho bạn biết điều gì đó đã xảy ra. Nó không cho bạn biết điều gì sẽ xảy ra tiếp theo, và dữ liệu cho thấy điều đó đúng theo cả hai chiều.
Funding rate predicts where price goes. It just changes which direction, by year.
Funding rate predicts where price goes. It just changes which direction, by year. I sorted every funding print into quintiles and measured the forward 72-hour return of the highest versus the lowest. In bear markets, high funding preceded WORSE returns - crowded longs getting squeezed. From 2024 onward it preceded BETTER returns - momentum. The signal is real. The sign is not stable. Average it across all years and you get -0.08%, below the 0.240% it costs to trade. WHAT IT MEANS This is genuinely different from a useless signal - there is information here. But to use it you would need to know the regime in advance, which is the same circular problem that kills everything else. Part of a public experiment: 5,063 USDT to 1,000,000 USDT, every result published - including the ones that make me look bad. Run on $BTC $ETH $DOGE #Binance #Futures #Backtesting #TradingBot #CryptoData
MAGIC is up +88.50% - I tested both ways to trade it
TODAY'S TOP GAINERS MAGIC: +88.50% KAIA: +66.13% BAT: +38.56% STRK: +29.63% PIXEL: +28.70% Every account on this app will show you this list. Here is the part they skip. I backtested both ways of trading a move like MAGIC's +88.50%: buying the breakout, and fading it on the assumption it reverts. Both lost money. Not narrowly - they lost roughly the transaction cost, 0.240% per round trip, which is what a signal carrying no information looks like once you charge it fees. A large 24-hour move tells you something already happened. It does not tell you what happens next, and the data says so in both directions. Part of a public experiment: every strategy I test gets published, working or not. Movers above: $MAGIC $KAIA $BAT #Binance #Altcoins #Futures #MAGIC #CryptoData
Nếu tín hiệu chứa bất kỳ thông tin nào về hướng đi, kết quả giữa các cấu hình hẳn đã khác nhau. Một vài cấu hình sẽ vượt qua mức chi phí. Nhưng không có cấu hình nào làm được. Đây không phải là lợi thế yếu cần tinh chỉnh — mà là hoàn toàn không có lợi thế, nhưng vẫn phải trả phí vào lệnh.
Đây là những gì dẫn đến kết quả đó.
Tôi đã thử nghiệm 378 chiến lược đường trung bình động. Chiến lược nào cũng lỗ đúng bằng chi phí giao dịch.
Hai nhóm tín hiệu, bốn khung thời gian, EMA và SMA, có và không có bộ lọc xu hướng, chỉ mua và chỉ bán. Tổng cộng 378 tổ hợp, tất cả đều áp dụng phí thực tế và mức trượt giá đo được trong thực tế.
Chi phí khứ hồi cho một giao dịch bằng 0,240% quy mô vị thế. Kết quả trung bình mỗi giao dịch, tính trên tất cả các cấu hình, nằm trong khoảng từ -0,361% đến -0,401%.
Chỉ có chi phí, không có gì khác.
Một phần của thử nghiệm công khai: từ 5.063 USDT lên 1.000.000 USDT, mọi kết quả đều được công bố — kể cả những kết quả...
Momentum inverted in 2026. The obvious fix: watch recent performance, and flip to the opposite rule when momentum stops paying.
I tested that over six different lookback windows. The best one returned +1.783% per rebalance against a +1.737% baseline - a difference indistinguishable from noise. Every other window was worse. None repaired 2026.
The reason is simple: by the time trailing performance tells you the regime flipped, you have already taken the losses that told you.
THE NUMBERS Baseline: +1.737%/reb Best adaptive rule: +1.783%/reb Difference: noise Best fix for 2026: still -0.66%
Adaptive rules feel like the answer to regime change. They are usually just a slower version of the same bet, with the losses already paid for the signal.
Part of a public experiment: 5,063 USDT to 1,000,000 USDT, every result published - including the ones that make me look bad.
MAGIC is up +87.80% - I tested both ways to trade it
TODAY'S TOP GAINERS MAGIC: +87.80% KAIA: +64.96% BAT: +31.74% STRK: +31.73% PIXEL: +24.43% Every account on this app will show you this list. Here is the part they skip. I backtested both ways of trading a move like MAGIC's +87.80%: buying the breakout, and fading it on the assumption it reverts. Both lost money. Not narrowly - they lost roughly the transaction cost, 0.240% per round trip, which is what a signal carrying no information looks like once you charge it fees. A large 24-hour move tells you something already happened. It does not tell you what happens next, and the data says so in both directions. Part of a public experiment: every strategy I test gets published, working or not. Movers above: $MAGIC $KAIA $BAT #Binance #Altcoins #Futures #MAGIC #CryptoData
I backtested a martingale system - add to losers, close when it recovers. 75 of 78 trades were winners.
It still lost 519.83 USDT, because the three losing trades gave back more than all 75 winners made combined.
Peak margin on those trades reached 4x the entire account.
THE NUMBERS Win rate: 96% Winning trades: 75 of 78 Net result: -519.83 Worst 3 trades combined: -1,098.53
Win rate is the easiest number to make look good and the least informative one to report. Anyone showing you a win rate without the size of the losses is showing you the half of the picture that flatters them.
Part of a public experiment: 5,063 USDT to 1,000,000 USDT, every result published - including the ones that make me look bad.
I swept 40 symbols and found three that looked profitable. I called them validated and put them live.
Then I tested the exact same settings on 751 days of price history the search had never touched. All three reversed. Not marginally - completely.
Three survivors out of 40 is roughly what pure chance produces at that sample size.
THE NUMBERS In-sample result: +1.74 Out-of-sample result: -208.72 Days of untouched data: 751 Symbols that survived: 0 of 3
Searching hard enough through random data will always produce a winner. The out-of-sample test is the only thing that tells you whether you found an edge or just found noise wearing a costume.
Part of a public experiment: 5,063 USDT to 1,000,000 USDT, every result published - including the ones that make me look bad.
MAGIC is up +77.53% - I tested both ways to trade it
TODAY'S TOP GAINERS MAGIC: +77.53% KAIA: +56.99% BAT: +35.01% STRK: +24.93% PIXEL: +23.33% Every account on this app will show you this list. Here is the part they skip. I backtested both ways of trading a move like MAGIC's +77.53%: buying the breakout, and fading it on the assumption it reverts. Both lost money. Not narrowly - they lost roughly the transaction cost, 0.240% per round trip, which is what a signal carrying no information looks like once you charge it fees. A large 24-hour move tells you something already happened. It does not tell you what happens next, and the data says so in both directions. Part of a public experiment: every strategy I test gets published, working or not. Movers above: $MAGIC $KAIA $BAT #Binance #Altcoins #Futures #MAGIC #CryptoData
I ran 3,209 breakout trades. The longs lost money. The shorts made all of it.
I ran 3,209 breakout trades. The longs lost money. The shorts made all of it. Donchian 20 channel, ATR stop, 10 symbols, four and a half years. If this were just riding a bull market, the longs would be the winners. They are not - they are negative. The shorts even made money on coins that ROSE over the period. BTC gained 94% buy-and-hold, and shorting its breakdowns still returned +22.83 USDT. XRP gained 87%, shorts returned +24.27 USDT. WHAT IT MEANS Crypto falls faster than it rises - liquidation cascades are asymmetric. That is documented market structure, which is why this result is more believable than one that only works when everything goes up. Part of a public experiment: 5,063 USDT to 1,000,000 USDT, every result published - including the ones that make me look bad. Run on $BTC $XRP $LTC #Binance #Futures #Backtesting #TradingBot #CryptoData
Backtest giả định rằng lệnh cắt lỗ của bạn luôn tồn tại. Nhưng khi giao dịch thật, đó là một lệnh có thể bị từ chối. Khoảng cách giữa hai giả định ấy là nơi tiền thật bị mất.
Đây là những gì đã xảy ra.
Lệnh cắt lỗ của tôi bị từ chối giữa lúc thị trường lao dốc. Sau đó, bot mở thêm 3 vị thế chỉ trong 6 giây.
Trong đợt giảm giá nhanh, Binance từ chối lệnh cắt lỗ của tôi với lỗi -2021: giá thị trường đã vượt qua mức kích hoạt trước khi lệnh được gửi đến.
Mã của tôi không xử lý trường hợp bị từ chối đó. Vị thế vẫn mở mà không có lệnh cắt lỗ, và vì trạng thái nội bộ của bot không ghi nhận điều này, tín hiệu tiếp theo lại mở thêm một vị thế. Cả thảy ba lần.
Mức rủi ro dự kiến là 3 USDT. Khoản lỗ thực tế là 11.75 USDT.
Đây là một phần của thử nghiệm công khai: biến 5,063 USDT thành 1,000,000 USDT, công bố mọi kết quả — kể cả những kết quả khiến tôi trông thật tệ.
MAGIC is up +86.41% - I tested both ways to trade it
TODAY'S TOP GAINERS MAGIC: +86.41% KAIA: +47.61% BAT: +32.07% PIXEL: +30.25% STRK: +24.42% Every account on this app will show you this list. Here is the part they skip. I backtested both ways of trading a move like MAGIC's +86.41%: buying the breakout, and fading it on the assumption it reverts. Both lost money. Not narrowly - they lost roughly the transaction cost, 0.240% per round trip, which is what a signal carrying no information looks like once you charge it fees. A large 24-hour move tells you something already happened. It does not tell you what happens next, and the data says so in both directions. Part of a public experiment: every strategy I test gets published, working or not. Movers above: $MAGIC $KAIA $BAT #Binance #Altcoins #Futures #MAGIC #CryptoData
I measured real execution costs with 48 controlled trades. One number decides everything.
I measured real execution costs with 48 controlled trades. One number decides everything. I stopped assuming what trading costs and measured it: 48 controlled round trips, entering and exiting immediately, recording the actual fill against the reference price. Fee was constant at 0.04% per side. Slippage was not - ETH filled at essentially zero, while DOGE cost 50 to 100 basis points per round trip. On a 150 USDT position, a round trip costs 0.36 USDT before the market moves at all. WHAT IT MEANS Any strategy must clear 0.240% of position size per trade just to break even. Most published setups never mention this number, which is why most published setups lose money. Part of a public experiment: 5,063 USDT to 1,000,000 USDT, every result published - including the ones that make me look bad. Run on $ETH $DOGE #Binance #Futures #Backtesting #TradingBot #CryptoData
MAGIC is up +88.22% - I tested both ways to trade it
TODAY'S TOP GAINERS MAGIC: +88.22% KAIA: +45.48% BAT: +36.83% STRK: +24.61% ZK: +22.88% Every account on this app will show you this list. Here is the part they skip. I backtested both ways of trading a move like MAGIC's +88.22%: buying the breakout, and fading it on the assumption it reverts. Both lost money. Not narrowly - they lost roughly the transaction cost, 0.240% per round trip, which is what a signal carrying no information looks like once you charge it fees. A large 24-hour move tells you something already happened. It does not tell you what happens next, and the data says so in both directions. Part of a public experiment: every strategy I test gets published, working or not. Movers above: $MAGIC $KAIA $BAT #Binance #Altcoins #Futures #MAGIC #CryptoData
Being long and short at the same time means the market's direction cancels out. What is left is whether the winners keep beating the losers - which is a real question, not a bet on the market going up.
Here is what that came from.
After 378 failures, one strategy finally passed the test that killed all the others.
Rank 45 perpetuals by 5-day return. Long the top 5, short the bottom 5, hold 15 days, rebalance. Equal weight, dollar-neutral.
The out-of-sample control is the one that had destroyed everything before it: fix the settings on old data, then run them on data the search never touched. In-sample +1.711% per rebalance, out-of-sample +1.801%. The mean held.
It also survives costs far worse than real: still +0.48% per rebalance even at a 1.50% round trip, six times what Binance actually charges.
Part of a public experiment: 5,063 USDT to 1,000,000 USDT, every result published - including the ones that make me look bad.
Same universe, same ranking rule, same holding period, five different years.
2024 and 2025 carried the whole strategy. 2026 is the worst year of the five - a 25% win rate over roughly nine months.
Buying the strongest coins and shorting the weakest has stopped working this year. Doing the opposite would have returned +1.47% per rebalance at a 69% win rate.
A strategy is not a fact about markets, it is a fact about a period. Anyone showing you a backtest without the year-by-year breakdown is hiding the years that did not work.
Part of a public experiment: 5,063 USDT to 1,000,000 USDT, every result published - including the ones that make me look bad.
MAGIC is up +81.90% - I tested both ways to trade it
TODAY'S TOP GAINERS MAGIC: +81.90% KAIA: +45.28% BAT: +39.73% ZK: +26.13% STRK: +21.87% Every account on this app will show you this list. Here is the part they skip. I backtested both ways of trading a move like MAGIC's +81.90%: buying the breakout, and fading it on the assumption it reverts. Both lost money. Not narrowly - they lost roughly the transaction cost, 0.240% per round trip, which is what a signal carrying no information looks like once you charge it fees. A large 24-hour move tells you something already happened. It does not tell you what happens next, and the data says so in both directions. Part of a public experiment: every strategy I test gets published, working or not. Movers above: $MAGIC $KAIA $BAT #Binance #Altcoins #Futures #MAGIC #CryptoData
Runners stopped while I rebuild the strategy from the backtest evidence. Nothing open.
Running on a Binance testnet account, so this is demo capital - I would rather say that than let you assume otherwise. The strategy, the costs and the losses are all real.
No signals for sale, no course, no affiliate link. Just the account and the data behind it.