Post-Rejection Follow-up Sampling: A Methodology for Counterfactual Outcome Measurement in Algorithmic DEX Trading

Algorithmic trading systems on decentralised exchanges (DEXs) reject the majority of candidate tokens they evaluate, yet the counterfactual outcome of rejected candidates, what would have happened had the system entered, is rarely measured. We introduce Post-Rejection Follow-up Sampling (PRFS), a methodology in which a separate tracking subsystem samples the post-rejection price and liquidity trajectory of each rejected token at a fixed cadence over a follow-up horizon of up to twenty-four hours. PRFS produces the data necessary to evaluate filter precision against the actual market outcomes of rejected candidates rather than against synthetic backtest reconstructions. We describe the methodology, its data architecture, and the deposit format. The companion dataset contains 67,000 forward-outcome observation rows across 2,997 rejection events spanning 457 unique mints, collected over a continuous eight-day observation window (2026-04-10 to 2026-04-19, UTC). Approximately fifty-five percent of rejection events receive at least one forward-outcome observation; coverage at the mint level is complete. The methodology is dataset-independent and generalises to any algorithmic trading system in which rejections substantially outnumber executions.