Research in the open
11,821,273 tries.
11,821,273 tries.
One winner.
We don't sell a magic bot — we sell the discipline of testing everything and keeping almost nothing. Here is every strategy configuration we evaluated between 2026-06-21 and 2026-07-28, and the single one that survived out-of-sample. Nothing here is live-trading performance.
The funnel
Each row is a harder test than the one above it. Real counts from our strategy lab — the survivors shrink at every step.
11,821,273
Config-evaluations
every parameter combo tried — grid sweeps, optimizer trials, exit-mechanics runs
12,909
Full backtests
real candles, next-open fills, fees + slippage, in-sample / out-of-sample split
5,408
Positive out-of-sample
made money on data they were never tuned on
482
Passed the gate
positive OOS + Sharpe ≥ 1 + at least 30 trades
1
Walk-forward robust
survived rolling walk-forward + multiple-testing correction
The one that survived
✓ Walk-forward robust
See it live on Investing →
EPS Acceleration
+15.38%forward return · 45 days
+3.10%vs. just holding the market
15names in the basket
What didn't make it
Published, not buried. These looked plausible and failed the same out-of-sample test — that's the point.
Momentum ∧ Quality
+2.50%
vs +3.10% hold — refuted
ROE
+0.15%
vs +3.10% hold — refuted
How to read this. Every figure binds to our live strategy-lab aggregates (2026-06-21 → 2026-07-28). Backtests use real candles, next-open fills, fees + slippage, and an in-sample / out-of-sample split. "Passed the gate" is a basic filter; only the walk-forward-robust winner is one we stand behind — and even that is research, not investment advice. No strategy is cleared for live trading.
Run the same tests yourself — free
Open the Strategy Lab and put any coin through the exact pipeline above, risk-free on paper.