The receipts
Most stock tools just tell you what to buy. Signal Deck measures whether its signals actually worked — and shows you, in the open.
For every signal we record, we track how that stock moved over the next week and month versus the market (the S&P 500). No cherry-picking and no hiding the misses — the table updates automatically as more results mature.
We keep which signal is which private — what matters here is the measured track record, not the recipe.
| Signal | ~1 week | ~1 month |
|---|---|---|
| Composite — multiple signals in agreement our headline: several independent signals lining up on one stock |
+0.3% · 54% beat (n=259) | +1.1% · 57% beat (n=72) |
| Signal A | -3.4% · 43% beat (n=84) | building… |
| Signal B | +3.4% · 100% beat (n=3) | -0.3% · 33% beat (n=3) |
| Signal C | +0.7% · 52% beat (n=31) | +1.1% · 67% beat (n=9) |
| Signal D | -5.0% · 0% beat (n=1) | building… |
| Signal E | +0.1% · 50% beat (n=74) | +1.0% · 57% beat (n=14) |
| Caution flag 1 a warning signal — negative is the point |
-1.2% · 30% beat (n=74) | -7.0% · 22% beat (n=36) |
| Caution flag 2 a warning signal — negative is the point |
-0.3% · 33% beat (n=9) | building… |
Each cell shows the median move versus the S&P 500 and the share of signals that beat it. “n” is how many matured examples — small samples are noisy, so early numbers wobble. Caution signals are expected to be negative — that's the point of flagging them.
Signals, not advice. This is information about how signals have behaved historically — not a recommendation to buy or sell anything, and past results don't predict the future. Everything you test inside the app runs on paper money.