Backtests

Signal Backtests

How Orrery's price-based signals performed historically, replayed on the top 15 markets by volume. Transparent methodology, no cherry-picking.

top 15

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.

updated 02:54:00 UTC·src:Polymarket CLOB

Momentum

359 firings across 11 markets

Price + 1h direction + 24h direction all aligned, with ≥3pp 24h move. Expected to continue in the same direction.

VerdictWeakSampleHigh339 firings · 11 marketsHistorically negative after spread — treat as research, not a profitable signal.

Expected · 24h, after spread

-5.85pp

raw avg − 2pp typical spread

Median · 24h

-0.65pp

robust to outliers

Win rate · 24h

41%

339 measured

Raw avg · 24h

-3.85pp

before spread / fees

Average move by horizon

  • 1h
    -0.55pp
  • 6h
    -2.66pp
  • 24h
    -3.85pp

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)
CategoryFiringsMeasuredWin 24hAvg 24h
Geopolitics26325537%-4.84pp
Sports473534%-1.96pp
Science393964%-2.54pp
Politics101090%+8.20pp

Win rate hidden until ≥10 measured firings — thin samples can flip with one outlier.

Per-market breakdown (11 markets fired)

Divergence

200 firings across 11 markets

1h move running against the 24h trend. Expected to continue short-term in the 1h direction before 24h reasserts.

VerdictStrongSampleHigh194 firings · 11 marketsWorth attention; verify on a per-market basis.

Expected · 24h, after spread

+2.55pp

raw avg − 2pp typical spread

Median · 24h

+0.75pp

robust to outliers

Win rate · 24h

53%

194 measured

Raw avg · 24h

+4.55pp

before spread / fees

Average move by horizon

  • 1h
    +0.04pp
  • 6h
    +2.63pp
  • 24h
    +4.55pp

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)
CategoryFiringsMeasuredWin 24hAvg 24h
Geopolitics15315057%+6.39pp
Sports292654%+0.92pp
Science131315%+0.98pp
Politics55-21.36pp

Win rate hidden until ≥10 measured firings — thin samples can flip with one outlier.

Per-market breakdown (11 markets fired)

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.