Hit rate (+120 min)
52.8%
1212 winners / 2296 measured
Honest scorecard — every recorded signal, every outcome
How Sigmatic's public signals have actually performed
Every BUY, WAIT, and SELL the app recorded for the public intraday scope, scored against the actual stored minute bars 30, 60, and 120 minutes after the call. WAIT counts are surfaced separately because a WAIT is not a directional bet — it's the absence of one. Disclosures · Back to the dashboard
Expectancy per call
+0.0%
Average signed return per directional call. Hit rate alone hides this.
Worst single session
-30.7%
Worst sum of signed returns from any one trading day in this window.
Window: Jun 22, 2026 → Aug 21, 2026 (NY) · 44 sessions. 2296 of 3039 directional calls have +120 min outcomes (price 2 hours later), 11,276 WAIT calls, 5 distinct symbols. Data last refreshed Aug 21, 4:37 PM ET (page rendered Aug 23, 12:42 AM ET).
Best + worst session in window
Best session+50.4%
Aug 4, 2026 · 89 of 97 directional calls measured · hit rate 100.0%
Worst session-30.7%
Jul 8, 2026 · 92 of 106 directional calls measured · hit rate 38.0%
Daily roll · last 44 sessions
| Session (NY) | Directional calls | With outcomes | Hit rate | Signed return |
|---|---|---|---|---|
| Aug 21, 2026 | 14 | 13 | 15.4% | -2.6% |
| Aug 20, 2026 | 73 | 36 | 58.3% | +2.0% |
| Aug 19, 2026 | 31 | 27 | 0.0% | -9.8% |
| Aug 18, 2026 | 31 | 16 | 12.5% | -3.1% |
| Aug 17, 2026 | 99 | 63 | 87.3% | +4.3% |
| Aug 14, 2026 | 34 | 33 | 45.5% | +0.2% |
| Aug 13, 2026 | 52 | 51 | 9.8% | -10.0% |
| Aug 12, 2026 | 21 | 16 | 0.0% | -3.0% |
| Aug 11, 2026 | 125 | 77 | 85.7% | +8.4% |
| Aug 10, 2026 | 30 | 14 | 21.4% | -2.0% |
| Aug 7, 2026 | 52 | 28 | 0.0% | -8.6% |
| Aug 6, 2026 | 42 | 36 | 33.3% | -5.5% |
| Aug 5, 2026 | 112 | 83 | 77.1% | +7.7% |
| Aug 4, 2026 | 97 | 89 | 100.0% | +50.4% |
| Aug 3, 2026 | 58 | 57 | 100.0% | +22.4% |
| Jul 31, 2026 | 104 | 88 | 69.3% | -0.9% |
| Jul 30, 2026 | 41 | 36 | 44.4% | -3.0% |
| Jul 29, 2026 | 5 | 5 | 100.0% | +2.5% |
| Jul 28, 2026 | 42 | 40 | 52.5% | -3.1% |
| Jul 27, 2026 | 83 | 78 | 65.4% | +9.5% |
| Jul 24, 2026 | 75 | 30 | 0.0% | -15.5% |
| Jul 23, 2026 | 73 | 36 | 13.9% | -7.5% |
| Jul 22, 2026 | 20 | 15 | 33.3% | -0.3% |
| Jul 21, 2026 | 41 | 40 | 85.0% | +3.6% |
| Jul 20, 2026 | 103 | 73 | 35.6% | -5.0% |
| Jul 17, 2026 | 47 | 42 | 78.6% | +11.0% |
| Jul 16, 2026 | 83 | 27 | 40.7% | -2.4% |
| Jul 15, 2026 | 101 | 85 | 34.1% | -12.3% |
| Jul 14, 2026 | 18 | 18 | 16.7% | -1.9% |
| Jul 13, 2026 | 131 | 76 | 68.4% | +1.8% |
| Jul 10, 2026 | 134 | 128 | 62.5% | -14.3% |
| Jul 9, 2026 | 49 | 49 | 67.3% | +5.1% |
| Jul 8, 2026 | 106 | 92 | 38.0% | -30.7% |
| Jul 7, 2026 | 120 | 87 | 21.8% | -27.1% |
| Jul 6, 2026 | 18 | 18 | 33.3% | -1.5% |
| Jul 2, 2026 | 172 | 149 | 53.0% | +26.1% |
| Jul 1, 2026 | 66 | 49 | 46.9% | +2.6% |
| Jun 30, 2026 | 55 | 50 | 100.0% | +15.2% |
| Jun 29, 2026 | 83 | 78 | 46.2% | -25.7% |
| Jun 26, 2026 | 20 | 15 | 80.0% | +0.9% |
| Jun 25, 2026 | 65 | 41 | 22.0% | -11.7% |
| Jun 24, 2026 | 116 | 89 | 49.4% | +2.5% |
| Jun 23, 2026 | 68 | 42 | 4.8% | -11.6% |
| Jun 22, 2026 | 129 | 81 | 50.6% | -6.2% |
Hit rate counts BUY + SELL calls whose price 2 hours later moved in the predicted direction. Sessions with zero directional calls don't appear (WAIT-only days are summarized in the headline counts above).
Data quality · tick counts in this window
Full ticks5,450
Every symbol the scanner attempted at this minute returned price data. Most ticks in a healthy window land here.
Partial ticks0
At least one symbol failed at this minute, but the broad-market posture symbols (QQQ + SPY) did return data. Signals are honored.
Outage ticks1,713
The scanner couldn't reach the data provider at this minute. Signals from these ticks are surfaced but flagged.
In flight1
The tick is still recording — the cron started but hasn't logged its end yet. A persistent nonzero count here signals stuck handlers.
All four buckets count ticks, not signals — each scanner run looked at every symbol we watch. 21 of 44 sessions in this window pre-date the per-tick history table (started 2026-05-02) — those sessions don't contribute to these counts.
1,640 provider ticks in the last 90 days were degraded (partial completion, handler error, or timeout). The most recent 50 are surfaced below.
- Aug 21, 3:55 PM handler-error 1 of 2 symbols affected
- Aug 21, 3:20 PM handler-error 1 of 2 symbols affected
- Aug 21, 3:10 PM handler-error 1 of 2 symbols affected
- Aug 21, 9:36 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:35 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:35 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:34 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:34 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:33 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:33 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:32 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:32 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:31 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:31 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:30 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:30 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:29 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:29 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:28 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:28 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:27 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:27 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:26 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:26 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:25 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:25 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:24 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:24 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:23 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:23 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:22 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:22 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:21 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:21 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:20 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:20 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:19 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:19 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:18 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:18 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:17 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:17 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:16 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:16 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:15 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:15 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:14 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:14 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:13 AM handler-error 1 of 1 symbols affected
- Aug 21, 9:13 AM handler-error 1 of 1 symbols affected
Why we don't lead with win rate
Win rate is the easiest number to inflate. A signal generator that calls a 0.1% move a “win” can hit 70% hit rate while losing money on every losing trade because the average loss is bigger than the average win. That's why this page leads with three numbers in equal weight:
- Hit rate — share of directional calls (BUY + SELL) that moved the predicted direction 120 minutes later. Easy to read but easy to game.
- Expectancy — average signed return per directional call. A 60% hit rate paired with +0.05% expectancy is a worse engine than a 45% hit rate paired with +0.4% expectancy. This is the number that pays the bills.
- Worst session— the worst single trading day's sum of signed returns. Tells you what the engine looks like on a bad day, not just an average day.
Read all three together. A scorecard that surfaces only one of them is a marketing pitch.
Corrections log
Sigmatic stores any after-the-fact correction to a signal or outcome (with operator email, before/after payload, and rationale) in an append-only audit table. The dashboard never silently rewrites history — if a number on this page changes after the fact, there's a row in the corrections log explaining why.
0 corrections recorded since the audit table started accumulating data on 2026-05-02. Zero corrections so far — the dashboard's recorded history matches what was shown live.
Not investment advice.This scorecard reflects what the public intraday scanner has done so far — it's a measurement, not a forecast. Past performance does not predict future results. Every signal on this page was generated by the same engine that runs the public dashboard, with no override or curation. R-based math (entry / stop / target / risk-reward) becomes available once the new outcomes materializer (deployed 2026-05-02) has run for a full window of trading sessions; until then, hit rate and signed-return metrics are the load-bearing numbers.