gap intelligence: model #16 vs market (consensus with #17)

← lab · 21,280 model-vs-market gaps dissected · survival AUC 0.6895 · ran 7/4/2026

What this is: Dissects every disagreement between our models and the market: who was right in which situations, whether two engines agreeing means anything, and whether a classifier can predict when OUR side of the gap wins.

Every model-vs-market disagreement dissected: who was right, where, and can we predict it? model advantage = market log-loss minus model log-loss in the bucket (positive = our number was closer to reality there).

Gap size	Direction	n	Model advantage	Reality	Model said	Market said	Stable
large	model_higher	1,628	-0.04576	0.3888	0.528	0.3946	⏳ yes
large	model_lower	1,379	-0.0433	0.4627	0.3354	0.4662	⏳ yes
medium	model_lower	3,483	-0.01611	0.3793	0.308	0.3791	⏳ yes
medium	model_higher	3,026	-0.0153	0.3827	0.463	0.3915	⏳ yes
small	model_lower	3,550	-0.0062	0.3521	0.3108	0.3452	⏳ yes
small	model_higher	2,840	-0.00227	0.3908	0.4136	0.3792	⏳ yes
tiny	model_higher	2,450	-0.00094	0.342	0.3617	0.3521	⏳ yes
tiny	model_lower	2,681	-0.00041	0.3469	0.332	0.3419	⏳ yes

Gap-survival classifier holdout AUC: 0.6895 · top features: rel_gap, prob_market, gap, price_open

Hurricane test over 4,560 shared predictions — Brier: engine A 0.19595, engine B 0.19611, engine mean 0.19402, market 0.19016. Averaging wrong models beats both members, but the market still beats the ensemble.

Reading the columnswhat each number actually means

Model advantage	market log-loss minus model log-loss in the bucket; positive = our probability was closer to what actually happened
Gap size	\|model p − market p\|: tiny <2pp, small <5pp, medium <10pp, large <20pp, huge beyond
Survival AUC	can gap features predict when our side of the gap is right? >0.5 = learnable
Hurricane test	are individually-wrong engines nearly-right on average? Brier: lower = better

Spec · the reproducible recipe

{
  "kind": "gap_intel",
  "model_run": 16,
  "second_run": 17
}