NHL Preseason ·
Dallas Stars – Minnesota Wild: the real probabilities
NHL Preseason. The chance of each outcome once the bookmaker margin is removed — what a bet is truly worth.
Reference: Pinnacle (margin removed)
Winner
| Outcome | Probability | Fair odds |
|---|---|---|
| Dallas Stars | 66.3 % | 1.51 |
| Minnesota Wild | 33.7 % | 2.97 |
Total points
| Line | Over | Under | Fair odds |
|---|---|---|---|
| 5.5 | 46.3 % | 53.7 % | 2.16 / 1.86 |
Updated 1 min ago · Probabilities: margin removed from Pinnacle's odds (power method), blended with Betfair Exchange (40 %) when it is liquid enough. This is not a prediction: a 60 % favourite loses 4 times out of 10.
Is a bookmaker wrong about Dallas Stars – Minnesota Wild?
Babeth checks about forty bookmakers against these probabilities every minute. When one pays more than the fair odds, she tells you where, at what price and how much to stake.
How are these probabilities computed?
Bookmakers build a margin into their odds: the implied probabilities add up to more than 100 %. We start from the sharpest bookmaker on the market — the one whose prices best predict results — and remove that margin (power method) so the probabilities sum to exactly 100 %.
When Betfair Exchange also prices the market with enough liquidity (at least €1,000 matched, spread ≤ 6 %), we blend the two: 60 % Pinnacle, 40 % Betfair. A calibration correction is added only if it has improved accuracy out of sample. The result is a fair probability and a fair odds figure: the price at which a bet neither wins nor loses money in the long run.
It is not a prediction. A 65 % favourite loses more than one time in three. The only way to make money is to be paid more than the real chance justifies — which is exactly what the app looks for.