Pro trading strategies,
just pick one.

Every strategy here is tested against years of real Binance price history — not flattering "what if" curves built to look good. We report each strategy on its own, honestly, instead of blending them into one rounded-up number:

• Momentum — tested over 6.3 years and 2,681 trades. About +31% growth per year, with a worst dip of roughly 27% along the way. It wins only ~15% of its trades on purpose: it takes many small losses and lets a few big winners run.
• Neutral — 5.6 years, 475 trades. About +22% per year, worst dip ~25%, and it wins roughly half its trades.
• Carry (funding) — 3.6 years. About +10% per year with a worst dip of just 0.5%, because it isn't betting on price going up or down.
• Carry (quarterly) — 6.3 years. About +7% per year, worst dip ~5%.

"Worst dip" (you'll often see it called drawdown) simply means how far your balance fell from its highest point before recovering — the number that tells you how bumpy the ride was.

Every result above uses real Binance data and plain 1× leverage — no hidden borrowing quietly inflating the numbers. And they're reproducible: run the same strategy with the same settings on the same data and you get the same result. We re-checked every figure above from scratch.

One honesty note: trading is never free — exchange fees and the small gap between the price you want and the price you actually get ("slippage") eat into returns, and for Carry the funding cost is fully built into the result. We're finishing a public write-up of exactly how those costs are modeled in each test, so read these as our measured research figures, refined as we document the final details.

Most "diversified" bots just run the same basic bet in three outfits. These three make money in genuinely different ways, so they tend not to struggle at the same time — and holding them together smooths out the ride.

• Momentum rides strong trends. It bets that when a market really gets moving, it keeps moving long enough to pay for the many small false starts along the way. It shines when markets are trending and bleeds slowly when they drift sideways.
• Neutral trades two related coins against each other. When two coins that normally move together drift apart, it bets they'll snap back to their usual gap. It doesn't care whether the overall market goes up or down — it shines when markets are choppy or range-bound.
• Carry collects a kind of structural "rent" that crypto markets pay to keep themselves balanced (the funding payments between traders, and the gap between futures and spot prices). It's set up to stay neutral to price direction — which is why its worst dip is near zero.

The key idea: one strategy needs trends, one needs range-bound markets, and one needs neither. So when Momentum is having a rough sideways month, Neutral and Carry are usually in their element — and the other way around. That's real diversification by design, not three coats of paint on the same idea.

Every strategy gives you a small set of plain-language sliders — things like how much to risk per trade, how much leverage to use, and how strong a trend must be before the bot acts. They're labeled in everyday terms with built-in help, not raw code.

What you can't touch are the safety rules — on purpose. They're written into the software and automatically re-tested before every release, so they can't be switched off by accident:

• Auto-stop on trouble. If a live bot hits 5 errors in a row, or its balance drops more than 10% in a single cycle, it stops itself and stays stopped until you restart it by hand — so the first sign of trouble is something you see, not something quietly swept under the rug.
• No half-done trades. For strategies that open two positions at once, if only one side fills, the bot automatically closes the leftover and flags it — you never end up accidentally exposed on one side.
• Safe mode by default. New bots run in the safer "isolated" mode, where a single position can't drain your whole account. The more advanced modes are off by default and unlock only behind an explicit setting, with a clear Beta warning and a small-capital confirmation first.

For the parts that handle real money — placing and sizing orders on the exchange — the code carries 83–99% test coverage across 391 automated tests. The dials you can move are kept within safe limits by design; the rules that keep you out of trouble simply aren't yours to override.