Archive
Rejected experiments
Every entry is a completed experiment that failed its predefined criteria. Filter by market, open any record for the full method and numbers, or take the whole set.
Showing 31–40 of 45 experiments
| ID | Experiment | Market | Mechanism | Horizon | Result | Verdict |
|---|---|---|---|---|---|---|
| E224 | On-chain forced selling adds nothing beyond the price drop liquidations, lending-protocol, exploits, depegs, forced-flow, perpetual-futures, matched-controls, event-study | Crypto & DeFi | Forced flow | 1h-24h | -54.3 bp vs matched drops, 4 h | KILL |
| E216 | Scheduled fixing and auction flows leave no tradable trace fx, fixing, month-end, gotobi, treasury-auctions, calendar-flow, hedge-rebalancing | Rates | Calendar | 1h-1d | No fix or auction effect | KILL |
| E190 | Fading a Korean coin premium on global perpetuals loses premium, korea, funding, perpetuals, event-study, reversion | Crypto & DeFi | Market design | 4h-24h | -130 bp at 4h, funding eats the drift | KILL |
| E180 | Copying skilled prediction-market wallets loses 1.3 to 2.6c a share prediction-markets, copy-trading, wallet-skill, persistence, latency, sports-contracts | Prediction markets | Public intent | 5m-1d | -1.3c per share after a 5 minute delay | KILL |
| E131 | Same contract on two prediction markets: the gap closes in seconds prediction-markets, cross-venue, arbitrage, event-contracts, live-order-books, latency | Prediction markets | Microstructure | 15m | +1.6c on paper, +0.0-0.2c live | KILL |
| E090 | No detectable pre-CPI/FOMC crypto drift that survives costs macro-releases, cpi, fomc, pre-event-drift, bitcoin, out-of-sample, event-study | Crypto & DeFi | Calendar | 1h-4h | +47.3 bp in sample, +12.8 bp out | KILL |
| E080 | Backing favourites, draws and sure things in sports markets fails prediction-markets, sports, esports, soccer, favourite-longshot-bias, draw, near-certain-outcomes, look-ahead, calibration | Prediction markets | Behavioural | 1h-6h | -1.0c / -0.5c esports replication | KILL |
| E057 | Extreme funding rates do not predict reversals in perpetuals funding-rate, crowded-leverage, perpetual-futures, contrarian, event-study | Crypto & DeFi | Behavioural | 8h-24h | -76 to +78 bp, all intervals include 0 | KILL |
| E024 | Scaling crypto exposure by volatility does not beat buy-and-hold bitcoin, ethereum, volatility-targeting, risk-sizing, sharpe-ratio, buy-and-hold, bootstrap | Crypto & DeFi | Risk premium | 1d | 0 of 12 intervals exclude zero | KILL |
| E014 | Crypto pair ratios do not revert at 1-24 hours pairs-trading, mean-reversion, relative-value, perpetuals, hourly | Crypto & DeFi | Behavioural | 12h-24h | -6.5 bp gross, 28 bp cost | KILL |