Projection
MLB player prop model
Player props,
priced like probabilities.
Posterior models player markets as probabilities first, then asks whether the book price creates enough edge to publish. Batter props, pitcher strikeouts, and matchup context all pass through the same public-record discipline.
Batter props
Hits+
Hits, total bases, runs, RBIs, singles, and related markets when supported.
Pitcher props
Ks
Strikeout markets are evaluated with slate and book-depth safeguards.
Audit
CSV
Resolved public rows are downloadable from the delayed ledger.
01 / Intent
Line
A prop only matters at a price
Quality
Bad coverage is filtered out
02 / How it works
Step 01
Estimate true probability
Player-level and matchup-level inputs create a probability for the specific prop outcome.
Step 02
Remove market noise
Consensus pricing is converted into a no-vig reference so the model is compared against a cleaner market baseline.
Step 03
Score the edge
A row needs enough gap, EV, and data quality to become an actionable selection.
Step 04
Track the result
Settled rows update the ledger and calibration record instead of disappearing after the slate ends.

Public proof
The market matters as much as the projection.
A high raw probability can still be a bad pick if the line is already priced correctly.
03 / Questions
Which MLB player props does Posterior cover?
Posterior covers qualified batter and pitcher markets such as hits, total bases, runs, RBIs, singles, home-run-related markets, and strikeouts when prices and data quality are available.
Why would a prop be excluded?
A prop can be excluded if the price is stale, the edge is too small, required context is missing, or the quality gate says the row is not strong enough to publish.
Is the model record public?
Yes. Resolved rows are released through the delayed public ledger, while live actionable rows remain inside Posterior Pro.
Posterior is baseball data, not a sportsbook. The public receipt lives at /data, the methodology lives at /methodology, and today's board starts at /today.
