LABELED FAIR REFERENCES · LIVE MARKET PRICES
Know what a bet is worth before you make it.
FairLine labels the source behind each fair reference. The evaluator prefers Pinnacle, then sportsbook consensus or Kalshi. Game boards place current book prices beside the Pinnacle no-vig line when it is available. You can also price Kalshi orders and track whether your logged bets beat the close.
Models built on free public data · calibration published openly · no sportsbook affiliation
Methods, limits, and tracked records stay visible.
Labeled references. Fair odds support price comparison. The tracked MLB pocket carries its limits and full record.
Fractional-Kelly sizing. The evaluator labels its fractional-Kelly stake and the probability assumption behind it.
Complete records. Every graded result stays on the board: wins, losses, and the sample-size warnings.
Published methods. Model inputs use free public data. Methodology and point-in-time backtests remain available.
Every tool helps you decide whether a price is worth taking.
TOOLS
01
See the fair reference
The no-vig break-even from the best available labeled reference for each listed game and market, updated throughout the day.
Dashboard →02
Evaluate any price
Type your book's odds and get the no-vig break-even from the labeled reference, your edge (even when it's negative), and the best number on the board.
Evaluate →03
Shop the books we track
Book-by-book prices against fair on one screen, with line movement. Take the best number, not the first one.
Game pages →04
Price Kalshi orders
The most each contract is worth paying after fees, and whether a passive order still has room to fill.
Kalshi finder →05
Log bets, watch CLV
Each logged bet keeps its entry price and closing reference, so you can see whether the market moved toward your number.
My Bets →PROJECTION MODELS
Independent model context
FairLine publishes independent MLB, NHL, and EPL projections built from free public data. The fixed MLB study measured a 4.2-point typical gap from the Pinnacle close. The accuracy page also shows live graded calibration by sport. Use the models as labeled context when a market is thin or moving.
We publish our misses.
Full accuracy dashboard →55.8%
moneyline accuracy, 2025 point-in-time backtest
+0.026
Brier skill vs the coin-flip baseline
4.2 pts
mean absolute model-to-Pinnacle gap · fixed MLB study
+8.8 pts
model-favored dogs vs Pinnacle's close · 2026-04-12 to 2026-06-15, n=291, forward confirmation pending