LATE-12/10 — Late-Window Threshold Entry

In-Sample Discovery Archive Now Live: Dynamic Shadow Forward-Validating 2026-07-09T12:24:34+00:00 UTC
IN-SAMPLE DISCOVERY RESULTS. This page documents the backtest results on the canonical feature panel (May 25-Jun 5 2026, 928 markets) that led to this candidate's deployment. The KISS distance_only logit model used by some strategies was trained on a subset of this same panel, so these in-sample figures are not forward-looking and cannot be used to infer live performance. This candidate was deployed as a live dynamic-shadow paper monitor on 2026-06-13 (fresh inception, no backfill) -- see the live deep dive for current forward fires and realized PnL.

kiss_late_sniper_12_10
Enters only at TTE=12, 11, or 10 -- never at TTE=14. Requires edge >= 0.10 (tighter threshold than KISS). Different entry timing from existing live strategies -- additive, not redundant. Second-best in-sample dynamic candidate on canonical panel.

PnL q=10 (IS)
+209.66
Trades
486
Win Rate
45.9%
Max Drawdown
-115.2
Sharpe (ann)
10.90
Sortino (ann)
14.18

Strategy Description

Entry logic: Enter only TTE=12-10; edge >= 0.10; skip TTE=14 entirely
Entry TTEs: TTE 12, 11, 10 only
Median entry TTE: 12 min
Side: NO (buy NO contract)
Quantity: q=10 contracts (backtest convention)
Model: KISS distance_only logit (2-feature: distance_to_strike_bps, abs_distance_to_strike_bps)

Comparison vs Benchmarks (In-Sample)

All figures below are in-sample on the canonical panel. Not live realized PnL.
StrategyTotal PnL q=10Trades Win Ratevs This Strategy
FTE-14 NO (unconditional TTE=14) +385.22 92853.7% -175.56
LATE-12/10 — Late-Window Threshold Entry (THIS STRATEGY) +209.66 48645.9% --
DOVE-14 (edge >= 0.07, TTE=14) +112.23 40849.8% +97.43
SIEVE-14 (edge >= 0.10, TTE=14) +104.08 27050.4% +105.58

Risk Metrics (In-Sample)

MetricValueNotes
Max Drawdown (q=10)-115.16 Peak-to-trough cumulative PnL drop
Sharpe (annualized)10.90 mean_daily / std_daily * sqrt(365); in-sample only
Sortino (annualized)14.18 mean_daily / downside_std * sqrt(365); in-sample only
Best day48.80 Highest single-day PnL
Worst day-52.07 Worst single-day PnL
PnL excl. best day160.86 Robustness check: not one-day-dependent
Avg entry price0.4157 Mean NO VWAP at entry
Median entry price0.3500 Median NO VWAP at entry
Promotion verdictPROMOTE_CANDIDATE Gate: beat KISS, DD ok, positive excl best day, no overfitting

Deployment Status

The figures below are in-sample only. This strategy never fires at TTE=14, so it avoids the highest-liquidity window; the higher edge threshold (0.10 vs 0.07 for KISS) adds some selectivity, but these results predate deployment. It was deployed as a live dynamic-shadow paper monitor on 2026-06-13 (fresh inception, no backfill) to begin forward validation -- see the live deep dive for current fires and realized PnL. Promotion to champion/benchmark status still requires confirmed forward (OOS) performance.

Forward (OOS) Gate for Champion/Benchmark Graduation

Gate: Deployed as a live dynamic-shadow paper monitor since 2026-06-13 (fresh inception, no backfill). Champion/benchmark graduation still requires sustained forward (OOS) outperformance over more trading days -- in-sample results below predate deployment and are not validated alpha.
What to check forward: Total PnL vs KISS on forward (OOS) days; drawdown vs KISS; win rate stability; no evidence of regime-specific overfit (check performance across BTC up/flat/down days separately).
Candidate registry: reports/consistent_panel_dynamic_yesno/shadow_strategy_registry_proposed.csv

Panel and Split Used

Panel: feature_panel_v2.parquet (local only, gitignored)
Date range: May 25-Jun 5 2026
Markets: 928 (all markets with valid TTE=14 snapshot)
Alignment: btc_left_forward_causal, 5s tolerance
Columns: 254 (158,601 rows total across all TTE snaps)
Split used for evaluation: Full 928-market pool (same as all published benchmark results)
ML split: Documented separately (train/val/retro-test); NOT applied to strategy evaluation here
Script: scripts/research/run_consistent_dynamic_yesno_v1.py