Skip to content
BAL strategy avatar
RejectedBAL

Polymarket BTC-to-Altcoin Lead/Lag

Cross-asset latency capture

Rejected — training PF 1.32 collapsed to PF 0.38 and −37.3% mean return in the chronological holdout. This threshold family overfit the selection period.

Category
Prediction markets
Window
2026-05-27 → 2026-06-24
Instruments
Polymarket BTC, ETH, SOL, XRP Up/Down
Timeframe
15-minute event markets
Tested
2026-06-24

Measured equity history

Recorded directly from the chronological test ledger.

24 points

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.

1.13Profit factor
Not measuredSharpe
Not measuredMax drawdown
57.1%Win rate
35Trades
0/11Gates passed
Not runPlacebo

How it works

Backfilling
  1. The 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.

  2. 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.

  3. 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
#GateResultPass
01Minimum sample35 trades.
02Profit factor ≥ 1.20Combined PF 1.129; holdout PF 0.378.
03Sharpe ≥ 0.6Not evaluable.
04Max drawdown ≤ 12%No sizing model.
05Positive ≥ 60% of periodsInsufficient history.
06Bootstrap LB Sharpe > 0Holdout interval crosses zero.
07Placebo beats p95Not run after holdout failure.
082× cost stress PF > 1.0Not run after holdout failure.
09Deflated Sharpe positiveNot evaluable.
10No component > 40%Only five holdout trades.
11Walk-forward OOS ≥ 0.9× ISPF fell from 1.316 in training to 0.378 in holdout.

VERDICT: REJECTED — HOLDOUT COLLAPSE

The rule bought the corresponding direction in ETH, SOL, or XRP after BTC moved at least 15 basis points while the altcoin had moved no more than half as far. It required an actual same-side taker-buy print and capped the observed entry near 65 cents.

Training looked promising, but the chronological holdout failed decisively: PF fell from 1.32 to 0.38, with a −37.3% average return across five holdout trades. The combined PF of 1.13 is below the validation floor and is driven by selection-period performance.

Sample Trades Win rate Mean return PF
Training 30 60.0% +12.66% 1.32
Holdout 5 40.0% −37.33% 0.38
Combined 35 57.1% +5.52% 1.13

This is a clean overfit warning. Retire the threshold family unless a future test begins with a longer, independently collected dataset and fixed parameters.

Tested across BTC, ETH, SOL, and XRP 15-minute markets from 2026-05-27 through 2026-06-24.

Frequently asked

Is Polymarket BTC-to-Altcoin Lead/Lag profitable in 2026?

In this pre-registered backtest (2026-05-27 → 2026-06-24), Polymarket BTC-to-Altcoin Lead/Lag returned a profit factor of 1.13 and passed 0/11 validation gates (placebo N/A). Verdict: REJECTED. Every result is published, pass or fail.

Has Polymarket BTC-to-Altcoin 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/btc_alt_leadlag/REPORT.md in the source repository.Author: Validated Research Team (Methodology v1.0 — 11-gate validation). Backtests are not investment advice.