Last updated: 14 October 2026 · By Power.Win Editorial Team
Quick Answer: Sportsbooks price a player prop by turning a projection into a probability, then adding margin: a −115/−115 prop holds 6.98%. That 6.98 points of overround costs $13.04 of profit on a winning $100 bet. Power.Win publishes player prop settlement rules sport by sport for the number it posts.
A player prop price looks like an opinion about a player. It isn’t. It’s arithmetic with a markup on top, and once you can read the markup you can see exactly what the sportsbook is charging you to take the position.
Three numbers carry this whole article. First a projection, say 24.5 points for an NBA guard. An implied probability, which is what the posted odds say the chance is. And the gap between the two sides’ probabilities adding up to more than 100%, which is the book’s margin. On a −115/−115 prop that gap is 6.98 points, and it is the single biggest difference between a prop and a main line.
This page explains the mechanism. It doesn’t tell you what to bet, and nothing here finds you an advantage — the margin below is working against both sides of every market on the board.
If you’re not sure what counts as a prop in the first place, start with what is a prop bet, which covers the market types and the difference between player, game and novelty props. The American odds format itself — why +200 and −300 pay what they pay — is worked through in what +200 and −300 odds pay.
How does a sportsbook turn a projection into a probability?
A sportsbook starts with a projection of the stat, not with a price. The projection is a distribution: a central estimate plus a spread of plausible outcomes around it. The price comes from asking how much of that distribution sits above the line.
The inputs are ordinary and public. Recent per-game numbers, minutes or snaps, pace, opponent strength, and whether the player’s role changed. Official league stat pages like NBA.com’s player stats are the same box-score data the models are fed.
The shape of the distribution depends on what’s being counted:
| Prop type | Typical stat | Why the shape matters |
|---|---|---|
| Counting events | Strikeouts, goals, shots on goal, assists | Low-count events cluster, so a half-point change in the line moves the probability a lot |
| Accumulating totals | Points, rebounds, passing yards | Roughly bell-shaped around a mean, so probability changes more smoothly |
| Yes/no events | Anytime goalscorer, to record a sack | One probability, no line to cross |
For low-count props the standard model is the Poisson distribution, which NIST’s statistics handbook defines as a model for the number of events in a given interval. That’s why a strikeouts line moving from 5.5 to 6.5 can swing the price hard while a points line moving from 24.5 to 25.5 barely twitches.
Now the step that matters. Suppose the model’s projection for our guard puts his median at 24.5 points. Half the distribution sits above the line and half below, so:
- Over 24.5 points: a 50% chance
- Under 24.5 points: a 50% chance
That 50/50 split is the fair probability. No margin in it yet. Everything from here is conversion and markup.
How does that probability become a posted price?
A probability becomes American odds through one of two formulas, picked by whether the chance is above or below 50%.
- Below 50% (an underdog price, a plus number): odds = 100 × (1 − p) ÷ p
- Above 50% (a favourite price, a minus number): odds = −100 × p ÷ (1 − p)
Run both directions and you get a conversion table. These are the rows worth memorising, because every prop you see sits somewhere on this scale:
| Implied probability | American odds | $100 stake returns |
|---|---|---|
| 33.33% | +200 | $300.00 |
| 40.00% | +150 | $250.00 |
| 50.00% | +100 | $200.00 |
| 53.49% | −115 | $186.96 |
| 60.00% | −150 | $166.67 |
| 66.67% | −200 | $150.00 |
| 75.00% | −300 | $133.33 |
Going the other way is just as mechanical. A plus price converts with p = 100 ÷ (odds + 100). A minus price converts with p = |odds| ÷ (|odds| + 100). So −115 means 115 ÷ 215, which is 53.49%.
Back to our 50/50 projection. The fair price on both sides is +100: stake $100, win $100. That is the price nobody posts, because a book that posted it would take in exactly what it paid out and keep nothing.
So the book shortens both sides. It posts Over 24.5 at −115 and Under 24.5 at −115. The projection hasn’t changed. The probability hasn’t changed. The posted odds now claim each side is a 53.49% chance, when the model says 50%.
That difference, 53.49% claimed against 50% modelled on both sides at once, is the price of the bet.
How much does a sportsbook hold on a −115/−115 prop?
Add the two posted sides’ implied probabilities together and the answer falls out:
| Side | Posted price | Implied probability |
|---|---|---|
| Over 24.5 | −115 | 53.49% |
| Under 24.5 | −115 | 53.49% |
| Total | 106.98% |
A fair two-way market totals 100%. This one totals 106.98%, so the overround is 6.98 percentage points. Overround, vig and hold all point at this same gap, and it is the number to quote when someone asks what a prop costs.
Expressed as a share of what gets staked rather than as points of probability, it’s slightly smaller: 1 − (1 ÷ 1.0698) = 6.52%. Both figures are correct and they measure different things, so it’s worth being clear which one is in play. The 6.98 points is how far the posted probabilities overshoot. The 6.52% is what the book keeps out of the money if both sides attract equal action.
What it costs you, in dollars.
On a winning $100 bet at −115, your profit is 100 × (100 ÷ 115) = $86.96, and your total return is $186.96. At the fair +100 price the same winning bet would have paid $100 profit and returned $200.
So the margin costs $13.04 of profit per winning $100 bet on this prop. You still have to be right to collect anything at all; the margin decides how much being right is worth.
What it earns the book, in dollars.
Now run it from the other side, with action balanced:
| Amount | |
|---|---|
| Staked on the over | $115 |
| Staked on the under | $115 |
| Total taken in | $230 |
| Paid to the winning side ($115 stake + $100 profit) | $215 |
| Retained | $15 |
$15 out of $230 is 6.52%, the same figure as the share-of-stake calculation above, arrived at through cash instead of probability. A balanced two-sided market is the only case where a book’s margin is locked in regardless of the result, which is why books would rather move a price than carry one-sided risk.
For context on how this margin compares with casino games, Wizard of Odds keeps a reference set of house edge figures across games. The comparison is a useful sanity check: a prop’s margin is not a small number.
Why does a player prop line move?
A posted prop moves for one of two reasons, and they’re worth telling apart because they mean opposite things.
Reason one: new information. The projection itself changed. An injury report, a player ruled out for rest, a role change, a lineup leak, a change in expected minutes. The model is re-run and comes back with a different distribution.
Reason two: money on one side. The projection didn’t change at all. The book is holding lopsided risk and wants the other side of the market filled.
The book can move the line, the odds, or both
This is the distinction most explanations skip. There are two separate dials:
| What moves | Example | What it changes for you |
|---|---|---|
| The line (the number) | 24.5 points → 26.5 points | What has to happen for the bet to win |
| The odds (the price) | −115/−115 → −190/+145 | What you’re paid for the same outcome |
| Both | 24.5 at −115 → 26.5 at −120 | The threshold and the payout together |
Worked through on our prop. Say the other starting guard is ruled out and the model’s projection rises from 24.5 to 27.0 points. The over at 24.5 now looks like roughly a 64% chance, which converts to a fair price of about −178. The book has three moves:
- Hold the line at 24.5 and reprice: post the over near −190 and the under near +145, which totals 106.33% and keeps the book’s cut where it was.
- Move the line to 26.5 and hold the price: the over at the new number sits near a coin flip again, so it goes back out near −115.
- Do some of each, the common outcome in practice.
All three are the same information expressed differently. Only the first one leaves the number alone, which is why “the line moved” and “the price moved” are not interchangeable.
Now the money case. Suppose 80% of the stake lands on the over at −115 and the projection hasn’t budged. The book can post Over −125 / Under −105 to make the under more attractive. Implied probabilities are then 55.56% and 51.22%, totalling 106.78% — an overround of 6.78 points, near enough the same as before. The midpoint of the market shifted while the book’s cut stayed in the same band.
One consequence worth knowing: props built from the same game interact. When legs make each other more likely, a book prices them as a group rather than multiplying them, which is explained in same game parlays and correlated legs.
How does Power.Win settle the number it posted?
A price is only half of a prop. The other half is the rule that decides what the posted number resolves against, and Power.Win writes those rules per sport rather than once for everything.
The sportsbook rules carry dedicated Player Props sections at Basketball 3.7, American football 4.13, Ice hockey 5.6 and Baseball 6.10, plus a longer soccer block. Three soccer rules do most of the work:
- Soccer 1.5: if a player isn’t in the starting XI, that player’s prop markets are voided. The rule carves out anytime, first and last goalscorer, player to receive a card, player to score x+, and team first goalscorer, which stand regardless.
- Soccer 1.32: soccer player props are settled using data supplied by OPTA. The settlement source is named, not left to a screenshot of a scoreboard.
- Soccer 1.38: a definitions section for the soccer prop terms themselves: goal and own goal, shots on goal, goal assist.
That third one is easy to skip and shouldn’t be. A “shots on goal” prop is only as precise as the definition of a shot on goal, and the price you took was built on the book’s definition, not yours.
No minimum time, once a player appears
The rule that catches people out is the runner rule, and Power.Win writes it per sport. For basketball (3.7.8) and ice hockey (5.6.6) there’s no minimum duration of play required for settlement: any player who appears on the field of play during active play counts as a Runner, and bets on all that player’s Player Prop markets stand.
Read the two together and the shape is clear. Whether your prop is live at all is decided by the lineup rule for that sport. In basketball and ice hockey, once the player gets on, no amount of early substitution or limited minutes gets your stake back — a player who plays ninety seconds has played. Soccer runs the other way: under 1.5 a player who isn’t in the starting XI has his props voided even if he comes on, outside the goalscorer and card markets the rule carves out.
So the number on the slip and the rule behind it are one product. A price built on a 30-minute projection and a rule that settles on a 90-second appearance are both true at the same time, and only one of them is printed next to the odds.
What does player prop pricing look like on Power.Win?
On Power.Win, a prop price and the rule that settles it are published in the same place: the odds on the sportsbook board, the resolution rule in the sportsbook rulebook, numbered by sport.
Props on the Power.Win sportsbook appear on the same boards as the main markets, both before a game and in play, across its major sports and esports listings. Stakes and payouts run entirely on crypto rails, with no fiat route in or out.
What’s worth checking before you read any prop price there, or anywhere:
- Find the two-way total. Convert both sides, add them, subtract 100%. That’s the overround on the market in front of you.
- Find the settlement rule for that sport. Power.Win’s rules name the section number, which means you can check it rather than assume it.
- Find the data source. For soccer props it’s OPTA, stated in rule 1.32.
- Check the line and the price separately. A number you liked at 24.5 is a different bet at 26.5, even at the same odds.
Power.Win is operated by 3-102-968317 SRL, Costa Rica, licensed by the Tobique Gaming Commission, licence 0000209. Power.Win isn’t available to players in the United States: section 5 of its Terms and Conditions lists the United States of America among the countries whose players aren’t eligible to deposit or play real-money games. Read this page as an explanation of how prop pricing works, not as an invitation to sign up.
Frequently Asked Questions
What is a no-vig price?
A no-vig price is what the odds would be with the margin stripped out. Divide each side’s implied probability by the two-sided total: 53.49% ÷ 106.98% = 50%, which converts back to +100. It’s the book’s own numbers with its cut removed.
Does a prop price tell you what the sportsbook expects to happen?
Not exactly. A −115 price implies 53.49%, but the book’s own projection behind it was 50% — the extra 3.49 points are margin, and the same markup is applied to the other side at the same time. A posted price is a projection plus a charge.
Does a player need to play a full game for a prop to stand?
It depends on the sport. In basketball (3.7.8) and ice hockey (5.6.6), no — Power.Win’s No Minimum Time rule treats any player who appears during active play as a Runner, and that player’s Player Prop markets stand on the official result. In soccer, rule 1.5 voids the props of a player who isn’t in the starting XI, so appearing as a substitute doesn’t make them stand.
Is the hold the same on the over and the under?
Not once a book shades a price. At −115/−115 both sides carry an identical 53.49%. Move to −125/−105 and they split to 55.56% and 51.22%, so the over is now carrying more of the 106.78% total than the under.
Which sports does Power.Win publish separate player prop rules for?
Power.Win’s sportsbook rules give Player Props their own sections for basketball (3.7), American football (4.13), ice hockey (5.6) and baseball (6.10), with soccer handled across rules 1.5, 1.32 and 1.38, including a definitions block for the prop terms.
Sources (Citations)
- Power.Win Sportsbook Rules (power.win/sportsbook-rules): Player Props sections Basketball 3.7, American football 4.13, Ice hockey 5.6, Baseball 6.10; Soccer 1.5 (starting XI void and its named exceptions), Soccer 1.32 (settlement on OPTA data), Soccer 1.38 (definitions: goal and own goal, shots on goal, goal assist); the “No Minimum Time” runner rule under Basketball 3.7.8 and Ice hockey 5.6.6. Read 14 October 2026.
- Power.Win Terms and Conditions (power.win/terms): section 5 (“Who Can Play”), which names the countries whose players aren’t eligible to deposit or play for real money. Read 14 October 2026.
- NIST/SEMATECH e-Handbook of Statistical Methods, 1.3.6.6.19 Poisson Distribution (itl.nist.gov): the Poisson distribution as a model for the number of events occurring in a given interval — the standard model behind low-count props. Read 14 October 2026.
- Wizard of Odds, House Edge (wizardofodds.com/gambling/house-edge/): reference house-edge figures, used only as a comparison scale for the prop margin computed here. Read 14 October 2026.
- NBA.com player stats (nba.com/stats/players/traditional): official per-game statistics, the public box-score inputs a projection is built from. Read 14 October 2026.
- Odds and probability conversions in this article computed from the standard American-odds formulas: p = 100 ÷ (odds + 100) for plus prices, p = |odds| ÷ (|odds| + 100) for minus prices.
Take the Lead, Gamble Responsibly
Knowing a prop’s overround tells you what the market costs to enter; it tells you nothing about which side of it lands, and the 6.98 points sit on top of both.
Work out what a bet is worth risking before the bet slip is ever open, and use the deposit limits, reality checks and self-exclusion tools on Power.Win’s responsible gambling page. If the betting has stopped being fun, treat that as the signal to stop; the support resources are listed on that same page.
By the Power.Win Editorial Team