Connor Faulkner

The Fallen · dead ends

The Fallen

Negative results rarely get written up. These are mine — experiments killed with evidence, kept here so I don't quietly re-run them. Each one names the test that killed it.

Raise no monuments to victory. Victory forgets.

Carve instead the names of the fallen hypotheses,

that the living may not repeat their marches.

— Testimonies of the Dry Run

Copy trading on Polymarket

fell 2026 · cause: a 28.3% win rate, negative every month

Followed profitable-looking leaders with paper money for four months: 3,122 trades, −$224, a 28.3% win rate, negative every single month. A leader-quality filter didn't restore the edge. Retired. The dry run's main value was making those losses fake.

Overnight seasonality in BTC

dead on arrival · cause: fees six times the edge

Tested every fixed 2-hour window across the sample. Best gross edge: about one basis point. Fees: six times that. Dead on arrival.

Funding-rate carry

shelved 2026 · cause: nothing to harvest

Gross carry ran between −3% and +2% annualised in 2026. There's nothing to harvest. Shelved unless the funding regime changes materially.

Fade-yesterday reversion

fell 2022, discovered 2026 · cause: the year split

Full-sample t-stats looked healthy. Split by year, the edge died in 2022 and never came back. Lesson kept: never trust a full-sample statistic without a year split.

ETH/BTC pairs trading

never lived · cause: a random walk in a spread's clothing

The ratio is a random walk with a 241-day mean-reversion half-life. There is no spread to trade.

Sub-daily microstructure

fell 2026 · cause: real, and 15× below taker fees

Real structure exists — a BTC-to-altcoin lead-lag with a t-statistic of 7.7 — but it sits roughly 15× below taker fees. Real and unreachable: the worst kind of result.

Multi-outcome arbitrage on Polymarket

fell 2026 · cause: zero candidates in four months

Buy every outcome when the set prices under $1.00. Four months of scanning found zero qualifying candidates. The market is tighter than the strategy assumed.

Sports betting signal-mining

killed at the research stage · cause: priced before it was found

Team form, referee histories, fantasy-football sentiment: public information the bookmakers price before you do, behind a 2–8% vig. Killed at the research stage without building the pipeline.

What survived marches on — Dispatches.