Polymarket BTC Spot Lead/Lag
BTC spot-to-contract latency capture
Rejected — 98 actual-print entries produced PF 0.97 and −0.81% mean return. The positive 25-trade holdout is statistically unresolved and does not rescue a negative combined result.
- Category
- Prediction markets
- Window
- 2026-05-27 → 2026-06-24
- Instruments
- Polymarket BTC Up/Down
- Timeframe
- 15-minute event markets
- Tested
- 2026-06-24
Measured equity history
Recorded directly from the chronological test ledger.
Shown in the backtest's native equity or cumulative P&L units. It is not scaled to an investment amount because the published artifact does not provide a defensible capital denominator.
How it works
BackfillingThe bet
What market behavior this strategy is wagering on.
Plain-English explainer is backfilling — check back tomorrow. We publish the explanation only once it's written and reviewed, never auto-filled.
How it decides
What makes it enter, size, and exit a position.
Plain-English explainer is backfilling — check back tomorrow. We publish the explanation only once it's written and reviewed, never auto-filled.
How it can break
The regimes and failure modes that turn the edge negative.
Plain-English explainer is backfilling — check back tomorrow. We publish the explanation only once it's written and reviewed, never auto-filled.
Gate scorecard — 0 / 11
auto-imported from results.json| # | Gate | Result | Pass |
|---|---|---|---|
| 01 | Minimum sample | 98 fills; below the 100-trade floor. | ✗ |
| 02 | Profit factor ≥ 1.20 | Combined PF 0.974 after recorded taker fees. | ✗ |
| 03 | Sharpe ≥ 0.6 | Not evaluated on a four-week binary-event sample. | ✗ |
| 04 | Max drawdown ≤ 12% | No account sizing model. | ✗ |
| 05 | Positive ≥ 60% of periods | Insufficient history. | ✗ |
| 06 | Bootstrap LB Sharpe > 0 | Combined bootstrap interval crosses zero. | ✗ |
| 07 | Placebo beats p95 | Not run because the base rule failed. | ✗ |
| 08 | 2× cost stress PF > 1.0 | Base-cost PF is already below 1. | ✗ |
| 09 | Deflated Sharpe positive | Not evaluated. | ✗ |
| 10 | No component > 40% | Not evaluated after rejection. | ✗ |
| 11 | Walk-forward OOS ≥ 0.9× IS | Positive holdout is too small and its interval crosses zero. | ✗ |
VERDICT: REJECTED — NO ROBUST NET EDGE
The hypothesis was that Binance-style spot movement reaches Polymarket’s BTC 15-minute contract with a short delay. The selected rule followed a move of at least 3 basis points when the matching contract could be bought for 70 cents or less.
Using the next actual same-side taker-buy print within ten seconds as the fill, the rule produced 98 trades, PF 0.97, and −0.81% mean return after recorded fees. Training was negative. The 25-trade holdout was positive, but its 95% interval ranged from −21.3% to +30.2% per trade. That is noise, not validation.
| Sample | Trades | Win rate | Mean return | PF |
|---|---|---|---|---|
| Training | 73 | 67.1% | −3.07% | 0.91 |
| Holdout | 25 | 72.0% | +5.81% | 1.21 |
| Combined | 98 | 68.4% | −0.81% | 0.97 |
The likely problem is latency and adverse selection: by the time another taker print confirms an executable price, the apparent spot lead has mostly been incorporated. Retire this simple threshold version.
Tested on 7,267 15-minute markets from 2026-05-27 through 2026-06-24.
Frequently asked
Is Polymarket BTC Spot Lead/Lag profitable in 2026?
In this pre-registered backtest (2026-05-27 → 2026-06-24), Polymarket BTC Spot Lead/Lag returned a profit factor of 0.97 and passed 0/11 validation gates (placebo N/A). Verdict: REJECTED. Every result is published, pass or fail.
Has Polymarket BTC Spot Lead/Lag been backtested honestly?
Yes — through The Validation Gauntlet, a pre-registered 11-gate framework (profit factor, deflated Sharpe, a random-permutation placebo, cost-stress and walk-forward) with the specification locked before any out-of-sample metric is computed. It failed and is published anyway.
Methodology: 11-gate validation — pre-registered spec, 11-gate battery, real market data. Full reproducible report: backtests/polymarket_spot_leadlag/REPORT.md in the source repository.Author: Validated Research Team (Methodology v1.0 — 11-gate validation). Backtests are not investment advice.