Statball

Track Record

Every pick the model has made this season, scored against the final. The predictions are recomputed from the frozen algorithm exactly as they stood pregame — the same numbers the slate showed — so there is nothing to curate and nowhere to hide. † marks pitcher-adjusted calls.

Season 2026, through 2026-09-27 · back to the picks

1,348–1,082
season record
55.5%
pick accuracy
0.2499
Brier score
347–237 (59%)
strong picks

$100 at real market prices

The season back-test, on honest terms: simulated bankrolls, real prices. Every paper position is bought at the actual traded Kalshi price at first pitch and settled by the real result — no model prices, no assumptions. The strategies trade only where the model genuinely disagrees with the market, sized quarter-Kelly with a 10% cap, constants chosen before any market data was seen. Coverage: 885 of 2,430 games (36%).

$450.81
strong edges · 650 bets
$322.67
all edges · 778 bets
$158.23
every pick · 885 bets
0.244 vs 0.257
market vs model Brier
$95$251$407MayJunJul$451$323$158
strong edges (≥5 pts vs price)all edges (≥2 pts)every pick at any price

In aggregate the market's prices are sharper than the model — which is exactly why the strategies bet only the biggest disagreements instead of every game. Exchange fees are not modeled; thin early markets can move between the priced snapshot and first pitch. Hover the chart for any day’s balances.

The Stake

real moneyday 106

On July 8, 2026 we put $100 of our own cash in a Kalshi account and handed it to the strategy: it buys the published strong signals automatically, and everything below is read from the exchange — we can’t retouch it.

$96.17
account value · -3.8% since the start
$96.17 cash
$0.00 in 0 open positions
36 bets placed
$100 · start$250 · target (+150%)

Strong signals only, limit orders at the ask, quarter-Kelly with hard caps — the full rules, every fill, and the paper benchmark live on The Stake.

By month, by confidence tier

MonthAll picksLeanSolidStrong
September58.9%212–14858%114–8255%69–5776%29–9
August56.6%236–18152%127–11965%82–4460%27–18
July55.5%206–16552%84–7855%79–6465%43–23
June57.0%225–17049%53–5664%56–3259%116–82
May52.7%221–19850%58–5952%54–5055%109–89
April51.5%202–19048%114–12358%72–5353%16–14
March60.5%46–3061%23–1555%16–1378%7–2

Tiers: lean = 50–56% stated confidence, solid = 56–62%, strong = 62%+.

›September212–148 · 58.9%
›August236–181 · 56.6%
›July206–165 · 55.5%

Thursday, July 16

0–1
✗Mets 4, Phillies 158%†—
›June225–170 · 57.0%
›May221–198 · 52.7%
›April202–190 · 51.5%
›March46–30 · 60.5%

Wednesday, March 25

0–1
✗Yankees 7, Giants 051%†—