Why
A model that outputs a probability cannot be graded with accuracy, and the substitution is the most common quiet mistake in applied work. Accuracy, precision, recall and F1 all require a threshold: somewhere the 0.63 becomes a yes. That threshold is a decision about costs, not a property of the forecast, and once it is applied the number you actually sized a position with has been discarded. A forecaster that is right 100% of the time on the questions it answers has not demonstrated skill — it has demonstrated that it only answers easy questions.
The correct grade for a probability is a proper scoring rule, meaning one whose expected score is optimised by reporting your true belief. Two are worth knowing. The Brier score is mean squared error on probabilities, it is bounded, and it decomposes cleanly, which is why it is the useful one here. The log score is the only local proper rule, and it punishes confident wrongness without limit — assign 0.001 to something that happens and the score is unbounded. The choice between them is a choice about whether one catastrophic call should dominate the record, and for a trading agent it usually should.
Then the trap, and it is the reason this card is not simply "calibrate your model". Calibration alone is worthless. A forecaster that ignores every question and always reports the base rate is perfectly calibrated — when it says 30%, the thing happens 30% of the time — and it is useless, because it never distinguishes one case from another. Murphy's decomposition of the Brier score names the two halves separately: reliability is whether your stated probabilities match observed frequencies, resolution is whether you push probabilities away from the base rate in the right direction. Calibration without resolution is a thermometer that always reads the seasonal average. Both numbers are needed, and only the decomposition shows which one you lack.
And there is a baseline sitting right there that most projects never use. On a prediction market the price is a probability, produced by everyone else's money, on exactly the questions being forecast, at exactly the moment of the decision. Scoring the agent against the market on identical events costs one extra logged column and settles the only question that matters — not "is the forecaster good" but "is it better than the number that was free". priced-by-the-wrong-thing argues that an instrument can answer a different question than the one being asked; this is the same discipline applied to your own output.
Finally, the reason it is not academic. Position sizing under Kelly is a function of the probability itself, not of a classification. A forecast that is directionally right but systematically overconfident produces bets that are systematically too large, and the account dies while the hit rate looks fine. Miscalibration does not show up as wrong answers. It shows up as the wrong bet size, which is a different failure and a worse one.
How it works
What each score can and cannot see
| Metric | Needs a threshold | Rewards honesty | Sees overconfidence | Note |
|---|---|---|---|---|
| Accuracy / F1 | Yes | No | No | Discards the magnitude you sized the bet with |
| AUC | No | No | No | Only ranking. A model that outputs p/10 for every p has identical AUC and is unusable |
| Brier | No | Yes | Yes | Bounded, and decomposes into reliability + resolution + uncertainty |
| Log score | No | Yes | Unboundedly | Local; one confident miss can dominate the whole record |
The AUC row is the one that catches people. Ranking metrics are invariant to any monotone squashing of the probabilities, so a systematically overconfident model and a well-calibrated one can score identically — and only one of them can be traded.
The decomposition, and why one number is not enough
Brier = reliability − resolution + uncertainty.
- Uncertainty is a property of the questions, not of you. It is the base-rate variance, and it sets the score you get for saying nothing.
- Reliability (lower is better) is the calibration term: when you said 0.7, did it happen 70% of the time?
- Resolution (higher is better) is the skill term: did you move away from the base rate, and correctly?
The always-say-the-base-rate forecaster has perfect reliability and zero resolution, which is exactly the profile of a system that has learned to be safe rather than useful. The opposite profile — high resolution, poor reliability — is the more common one for an LLM asked for a probability: it discriminates cases well and states the odds too confidently. The two failures need opposite fixes, and a single Brier number tells you neither.
The measurement, in the shape this project can actually run
The agent card already produces the input. One decision row is enough:
| Column | Where it comes from |
|---|---|
p_model |
The forecast the agent acted on |
p_market |
Implied probability from the order book at the same instant |
outcome |
0 or 1, after resolution |
From those three: a reliability diagram, both Brier scores, and their decompositions. The comparison against p_market is the experiment; everything else is diagnostics. And it has an unusual property for a research measurement — it cannot be gamed by picking easier questions, because the baseline gets the same questions.
One honest caveat to state up front: an order book gives a bid-ask spread, not a point probability, and the mid is a convention rather than a truth. On a thin market the spread can be wider than the edge being claimed, so the spread belongs in the log next to the mid — otherwise the comparison silently flatters whichever side of it the agent traded on.
Why this sits in the Future section
It is not a forecasting technique. It is the measurement layer that any probabilistic output needs, and the outputs are multiplying — an LLM asked for a likelihood, a policy's confidence in a grasp, a router deciding whether to escalate. the-harness-not-the-model argues that what you build around a model matters more than the model. Scoring is the part of that harness that tells you whether anything inside it is working, and it is usually the part nobody builds because the model already returns a number that looks like an answer.
Where it lands in Jayverse
- Verex: log every agent probability against the market's own price. For any pricing or resolution agent that outputs a probability, persist forecast p, the market's implied price at the same instant, and the outcome — the Brier score against the market price is the only test that proves the agent has edge.
- OFA: grade solver/router confidence with Brier or log score, not accuracy. A solver that reports a win-probability for a route needs a proper scoring rule, since a threshold-based accuracy number discards the size the auction actually bid.
- Rabbit: decompose calibration from resolution before sizing anything. Any agentic component that sizes an action off a probability (session-key risk, position size) should log the reliability/resolution split, since an overconfident-but-well-ranked model sizes wrong even when it looks accurate.
Key expressions
| Expression | 뜻 · 쓰이는 자리 |
|---|---|
| quiet mistake | 눈에 잘 안 띄는(조용한) 실수 · 흔히 저지르지만 티가 안 나는 오류를 말할 때. "the most common quiet mistake in applied work" |
| discard | 버리다, 폐기하다 · 계산에 쓰인 정보를 결과 발표 후 버릴 때. "has been discarded" |
| punish without limit | 한없이(무한정) 벌하다 · 과신한 예측이 틀렸을 때 점수가 무한히 나빠지는 경우. "it punishes confident wrongness without limit" |
| dominate the record | 기록 전체를 좌우하다, 압도하다 · 한 번의 큰 실패가 전체 평가를 결정할 때. "one catastrophic call should dominate the record" |
| sitting right there | 바로 거기 있는, 누구나 쓸 수 있는 · 이미 존재하는데 활용 안 되는 자원을 가리킬 때. "a baseline sitting right there" |
| settle the only question that matters | 유일하게 중요한 질문에 답을 내다 · 부차적 논의를 걷어내고 핵심만 확정할 때. "settles the only question that matters" |
| catch people | (사람들이) 걸려 넘어지다, 헷갈려하다 · 흔히 실수하는 포인트를 가리킬 때. "The AUC row is the one that catches people." |
| invariant to | ~에 영향받지 않는, 불변인 · 어떤 변환을 가해도 값이 안 바뀔 때. "invariant to any monotone squashing of the probabilities" |
| flatter | 실제보다 좋아 보이게 하다 · 편향된 비교가 결과를 유리하게 왜곡할 때. "silently flatters whichever side of it the agent traded on" |
| game (something) | 부정하게 이용하다, 조작하다 · 시스템을 악용해 유리한 결과를 얻을 때. "it cannot be gamed by picking easier questions" |
| proper scoring rule | 정직한 답을 하도록 설계된 채점 규칙 · 확률 예측을 평가하는 통계 용어. "is a proper scoring rule, meaning one whose expected score" |
| AUC | 곡선 아래 면적(Area Under the Curve) · 순위 매기기 성능만 보는 지표, 확률 보정을 전혀 반영하지 못함. "The AUC row is the one that catches people" |
| Brier score | 브라이어 점수 · 확률 예측의 평균제곱오차, 신뢰도·해상도·불확실성으로 분해되는 지표. "The Brier score is mean squared error on probabilities" |
| log score | 로그 점수 · 확신에 찬 오답을 무한히 벌하는 유일한 로컬 적정 채점 규칙. "The log score is the only local proper rule" |
| Kelly | 켈리 공식(포지션 사이징) · 확률값 자체로 베팅 크기를 정하는 방법, 과신하면 베팅 규모가 커짐. "Position sizing under Kelly is a function of the probability" |
| Murphy's decomposition | 머피 분해 · 브라이어 점수를 신뢰도와 해상도 두 항으로 나눠 어느 쪽이 부족한지 보여줌. "Murphy's decomposition of the Brier score names the two halves" |