Two axes, two different questions. The Tapelab Score (0–100) answers is this a real, tradeable edge? — four dimensions of 25 (edge, robustness, practicality, evidence discipline) summed, with the sub-scores shown in every report. The Evidence Grade (A–F) answers a separate question: how much should you trust that verdict? Keeping them apart is the point. The buy-the-dip setup that works scores 70 against the board's top 76 yet carries our highest evidence grade A — it is the one we pre-registered and tested on a held-out era. A failure can carry a strong grade too: 30 at grade B means we are confident the idea is empty. A test that was not pre-registered is capped on evidence discipline, and its report says so.
Yes — a real, statistically significant edge, but a smaller one than the backtest advertises. The honest advantage is the risk-adjusted tilt (Sharpe ≈ 1.24 vs the market's 1.02) carrying ~8.8%/yr of genuine alpha (t ≈ 2.9), not the eye-catching +11-point CAGR gap, most of which is survivorship bias and a historic bull market.
The signal genuinely works out-of-sample — and that turns out not to be enough. Each individual setup carries a real, positive edge that survived a strict held-out test. But the setup fires far more often than any real account has room for, and when we're forced to choose which ones to take, the ones we skip quietly beat the ones we keep. A great signal you can't fully hold.
There is no prize to win. Before building any model, we computed the absolute ceiling — the most an ideal regime-timer could possibly add — and in the out-of-sample era that ceiling was negative. No model, however clever, can extract an edge that isn't there. We killed it in about fifteen minutes of compute, exactly as planned.
It does exactly what it promises — reduces your drawdown — and it costs you more than it's worth. Moving stops to breakeven cut our strategy's average profit per trade by more than a quarter, because it repeatedly knocks you out of the trades that were about to become your biggest winners. It doesn't protect gains; it amputates them.
No. The win rate is about 65%, not 91% — and 65% is exactly what you'd get buying any S&P 500 stock at random in the same years. The signal's forward returns are statistically indistinguishable from noise, and slightly negative relative to just being long. There is no directional edge to leverage.
Scores are score v1: a fixed, public rubric, revised only by a versioned v2 — never by re-grading history. Each report states its universe, window and whether the test was pre-registered. Numbers are before costs unless the report says otherwise, and equity studies on today's index membership are survivorship-flattered. Nothing here is investment advice.