How a pick gets made — from a per-plate-appearance rate to published bets.
The FCM model builds ideas from scratch — does this hitter record a base, does this starter stay under his strikeout number — and then compares its own price to the sportsbook's, with the sportsbook's margin stripped out.
Almost everything interesting happens in two places:
(1) our multipliers adjust a hitter's rate into a rate for today: who's pitching, weather and wind effects, which hand he swings from and how much of the game the bullpen will cover.
(2) a calibration layer fit on the model's own graded ledger decides how much that opinion is worth in each public market
A hitter's total-bases projection is one line of arithmetic. Every term is a multiplier on his own per-plate-appearance rate. The first four are blended by starter share — the starter's terms carry the fraction of the lineup's plate appearances he is projected to cover; the bullpen term carries the rest.
| Term | What it reads |
|---|---|
| Platoon | Hitter's split severity vs. this arm's hand, leveraged by the starter's arsenal |
| Pitch-type matchup | Batter's per-pitch-type value against what this starter actually throws |
| Bullpen quality | Opposing pen's effective xFIP — leverage-corrected & fatigue-adjusted |
| Weather | Temperature & wind resolved against the hitter's handedness |
| Park geometry | Venue × handedness fixed effect, multi-season blended rate |
| Opposing lineup K | Applied to the starter's strikeout projection (not the hitter's) |
The park term never touched this pick. Yankee Stadium's short porch is real and the model knows it - the fixed effect is confined to Rice's home run number - a different bet.
This pick was carried by his own rate, the diluted platoon penalty and the weather.
This is what the gate is supposed to look like from the inside. It is not a judgment about Jacob Lopez on Aug 28 — it is a judgment about the population of picks that look like this one, made from 352 graded K unders. A gate that only ever suppresses losers is not a gate, it is hindsight. Publishing this pick because it happened to win would mean the ledger stopped governing and the vibes took over.
Every Monday the calibration layer refits on the full graded ledger. It asks one question per market: given what the model thinks and what the market thinks, what actually happened?
The weights are the interesting part! A high model weight means the model's disagreement with the price has historically been informative. A weight near zero means it hasn't. A negative weight means the model's opinion in that market has been actively worth fading — and we fade it automatically.
| Market | WEIGHT OF FCM'S OPINION |
|---|---|
| Full-game moneyline | |
| TB/H — Over | |
| K — Over | |
| K — Under | |
| ER — Under | |
| Pitcher Outs — Under | |
| TB/H — Under |
Weights current as of August 2026 — fit on 2,339 graded picks. The layer refits weekly on the growing ledger, so treat this table as a dated snapshot, not a live feed. Bars are scaled to the largest absolute weight.
We provide this data for transparency - it's something a marketing page rarely volunteers: the FCM model's opinion about unders is currently worth nothing in three markets. When our model identifies an edge for those markets, the ledger says the book was usually right.
F5 moneylines are the model's worst product — a losing record against the closing no-vig price. The FCM model's response to a losing lane is to track it until the evidence decisively produces positive EV.