Signal Backtests
How Orrery's price-based signals performed historically, replayed on the top 15 markets by volume. Transparent methodology, no cherry-picking.
History
1M CLOB
Universe
Top 15
Cooldown
1h
Cost model
-2pp
Orrery replays the same live rules minute by minute, then measures 1h, 6h, and 24h follow-through. Flow and resolution-risk need event-time history, so they are excluded for now.
Momentum
239 firings across 8 marketsPrice + 1h direction + 24h direction all aligned, with ≥3pp 24h move. Expected to continue in the same direction.
Expected · 24h, after spread
-3.69pp
raw avg − 2pp typical spread
Median · 24h
+0.00pp
robust to outliers
Win rate · 24h
42%
212 measured
Raw avg · 24h
-1.69pp
before spread / fees
Average move by horizon
- 1h-1.44pp
- 6h-1.78pp
- 24h-1.69pp
Centred at zero. Bar length is proportional to the largest of the three windows, so the decay shape is visible without re-scaling.
By category (4 categories fired)
| Category | Firings | Measured | Win 24h | Avg 24h |
|---|---|---|---|---|
| Politics | 69 | 58 | 26% | -2.85pp |
| Macro | 63 | 59 | 56% | +1.18pp |
| Sports | 59 | 47 | 34% | -5.26pp |
| Geopolitics | 48 | 48 | 54% | +0.60pp |
Win rate hidden until ≥10 measured firings — thin samples can flip with one outlier.
Per-market breakdown (8 markets fired)
| Market | Firings | Win 24h | Avg 1h | Avg 24h |
|---|---|---|---|---|
| Will there be no change in Fed interest rates after the September 2026 meeting? | 63 | 56% | +0.38pp | +1.18pp |
| Will Nigel Farage win 60–70% of votes in the Clacton parliamentary by-election? | 51 | 27% | -0.15pp | -3.98pp |
| Will CF América win on 2026-08-13? | 49 | 36% | -4.48pp | -6.02pp |
| Strait of Hormuz traffic returns to normal by August 31? | 48 | 54% | +0.06pp | +0.60pp |
| Will Nigel Farage win at least 80% of votes in the Clacton parliamentary by-election? | 18 | 21% | +1.37pp | +0.36pp |
| Milwaukee Brewers vs. Los Angeles Dodgers | 5 | 0% | +0.00pp | -3.00pp |
| Cincinnati Open: Katie Boulter vs Katie Volynets | 3 | — | -7.00pp | +0.00pp |
| Cincinnati Open: Rinky Hijikata vs Gael Monfils | 2 | — | -74.40pp | +0.00pp |
Divergence
125 firings across 9 markets1h move running against the 24h trend. Expected to continue short-term in the 1h direction before 24h reasserts.
Expected · 24h, after spread
-2.39pp
raw avg − 2pp typical spread
Median · 24h
+0.00pp
robust to outliers
Win rate · 24h
44%
105 measured
Raw avg · 24h
-0.39pp
before spread / fees
Average move by horizon
- 1h+0.45pp
- 6h+1.43pp
- 24h-0.39pp
Centred at zero. Bar length is proportional to the largest of the three windows, so the decay shape is visible without re-scaling.
By category (4 categories fired)
| Category | Firings | Measured | Win 24h | Avg 24h |
|---|---|---|---|---|
| Sports | 42 | 34 | 32% | -2.10pp |
| Politics | 32 | 22 | 59% | +1.31pp |
| Macro | 32 | 30 | 55% | +0.24pp |
| Geopolitics | 19 | 19 | 32% | -0.54pp |
Win rate hidden until ≥10 measured firings — thin samples can flip with one outlier.
Per-market breakdown (9 markets fired)
| Market | Firings | Win 24h | Avg 1h | Avg 24h |
|---|---|---|---|---|
| Will CF América win on 2026-08-13? | 39 | 32% | -0.31pp | -2.26pp |
| Will there be no change in Fed interest rates after the September 2026 meeting? | 31 | 55% | +0.27pp | +0.29pp |
| Will Nigel Farage win 60–70% of votes in the Clacton parliamentary by-election? | 20 | 71% | +2.12pp | +2.29pp |
| Strait of Hormuz traffic returns to normal by August 31? | 19 | 32% | +0.05pp | -0.54pp |
| Will Nigel Farage win at least 80% of votes in the Clacton parliamentary by-election? | 12 | 38% | +2.42pp | -0.31pp |
| Milwaukee Brewers vs. Los Angeles Dodgers | 1 | — | +0.00pp | +0.00pp |
| Cincinnati Open: Katie Boulter vs Katie Volynets | 1 | — | -12.00pp | +0.00pp |
| Will the Fed decrease interest rates by 25 bps after the September 2026 meeting? | 1 | — | -0.25pp | -1.25pp |
| Cincinnati Open: Rinky Hijikata vs Gael Monfils | 1 | — | +0.00pp | +0.00pp |
Honest caveats
- Small samples. CLOB returns ~30 days of minute-level history; short-dated markets may have very few firings. Anything below 10 measured firings shows a dash instead of a win rate so we don't overstate confidence.
- Survivorship. This backtest runs on markets currently live — not on markets that already resolved and disappeared. Persistent historical coverage is planned so the full universe can be evaluated once enough live history has accumulated.
- No execution costs. Avg-move figures are mid-price to mid-price. Real entry/exit would pay spread, slippage, and Polymarket fees — so a +1.5pp 24h move is not +1.5pp of profit.
- Signal drift. If we tune the live rule, these numbers change with it. That's the point — the page always reflects the rule that's currently running, not a frozen historical version.