Moneyline Bets Explained: How the Price Maps to Win Probability, and Where the Vig Sits
A moneyline bet asks one question: who wins? No spread. No total. The price is the whole bet, and every price carries two things inside: the win probability the book charges you for, and the margin the book keeps.
This guide reads the DraftKings moneyline boards for games October 24 to 26, 2026, across the NFL, NBA, and NHL, plus the World Series price. Three findings come out of the numbers.
- The margin runs between 3.91% and 4.50% per dollar on all 28 two-way moneylines checked, from a near pick'em to a -625 favorite.
- The 20-cent gap bettors learn first only describes lines near even money. On a -520 favorite the gap is 130 cents. On a -625 favorite the gap is 170.
- The split of the vig between favorite and underdog depends on a model choice. On one real line, the choice moves the underdog's cost from 4.10% to 14.11% per dollar.
How a Moneyline Pays
A minus sign marks the favorite. The number is what you risk to win $100. A plus sign marks the underdog. The number is what a $100 stake wins. Your stake comes back on top of the profit, so a winning $100 bet on the Giants at +390 returns $490.
The four formulas behind every moneyline
Favorite at -F: profit on a $100 stake = 100 × 100 / F
Underdog at +D: profit on a $100 stake = D
Break-even win rate at -F = F / (F + 100)
Break-even win rate at +D = 100 / (D + 100)
Texans -520: 520 / 620 = 83.87%
Giants +390: 100 / 490 = 20.41%
Here are real prices from the DraftKings boards on ESPN's odds pages, read about 10:25 p.m. ET on October 23, 2026.
| Team | Price | Profit on $100 | Break-even win rate |
|---|---|---|---|
| Texans (NFL) | -520 | $19.23 | 83.87% |
| Vikings (NFL) | -290 | $34.48 | 74.36% |
| Steelers (NFL) | -162 | $61.73 | 61.83% |
| Flyers (NHL) | -122 | $81.97 | 54.95% |
| Ducks (NHL) | +102 | $102.00 | 49.50% |
| Saints (NFL) | +136 | $136.00 | 42.37% |
| Colts (NFL) | +235 | $235.00 | 29.85% |
| Giants (NFL) | +390 | $390.00 | 20.41% |
The break-even column is the price written as a probability. A Texans ticket at -520 pays off over time only if Houston wins more than 83.87% of the time. The odds converter runs this conversion for any price. For heavy chalk in football, the juice trap breakdown covers the cost of laying big numbers.
The Mirror Price and the Cents Gap
Every favorite price has a mirror. At -150, the favorite implies 60%. At +150, the underdog implies 40%. The two sum to exactly 100%, so a -150 / +150 market carries zero vig. The vig on any moneyline is the distance between the underdog's posted price and the mirror of the favorite's price.
Measuring the gap in cents
Mirror check: F / (F + 100) + 100 / (F + 100) = 1
Gap with a plus-money underdog: F - D
Gap with two minus prices -A and -B: (A - 100) + (B - 100)
Flyers -122 / Ducks +102: 122 - 102 = 20 cents
Bengals -115 / Ravens -105: 15 + 5 = 20 cents
A Covers guide published by NBC Sports says most moneylines go up as 20-cent lines, with -150 / +130 as the example, and notes the convention varies by book and sport. The live boards confirm the rule near even money. The rule breaks as soon as the favorite grows.
| League | Market | Favorite | Underdog | Gap | Booksum | Cost per $1 |
|---|---|---|---|---|---|---|
| NFL | Bengals at Ravens | -115 | -105 | 20 | 104.71% | 4.50% |
| NHL | Ducks at Flyers | -122 | +102 | 20 | 104.46% | 4.27% |
| NBA | Heat at Hornets | -148 | +124 | 24 | 104.32% | 4.14% |
| NFL | Steelers at Saints | -162 | +136 | 26 | 104.20% | 4.04% |
| NHL | Kings at Devils | -166 | +140 | 26 | 104.07% | 3.91% |
| NHL | Sharks at Bruins | -192 | +160 | 32 | 104.21% | 4.04% |
| MLB | World Series winner, Brewers vs Rays | -215 | +175 | 40 | 104.62% | 4.41% |
| NFL | Broncos at Cardinals | -238 | +195 | 43 | 104.31% | 4.13% |
| NFL | Colts at Vikings | -290 | +235 | 55 | 104.21% | 4.04% |
| NFL | Giants at Texans | -520 | +390 | 130 | 104.28% | 4.10% |
| NBA | Bucks at 76ers | -625 | +455 | 170 | 104.22% | 4.05% |
Booksum and cost per dollar
Booksum B = implied favorite + implied underdog
Cost per $1 with the vig split in proportion = 1 - 1 / B
Giants at Texans: 83.87% + 20.41% = 104.28%
1 - 1 / 1.0428 = 4.10%
Read the gap column against the cost column. The gap runs from 20 cents to 170 cents. The cost stays between 3.91% and 4.50% on all 11 markets in the table, and the eight lines past -150 sit between 3.91% and 4.41%. The table shows 11 of the 28 two-way markets on those boards: 13 NFL games, 12 NHL games, two NBA games, and the World Series price. All 28 fall between 3.91% and 4.50%.
The pairs repeat across leagues. A -148 favorite comes with a +124 underdog four times on those boards, in the NFL, NHL, and NBA. A -135 favorite comes with +114 three times. DraftKings prices the margin as a percentage and lets the cents float.
Tadgh Hegarty and Karl Whelan of University College Dublin use the same 1 - 1/B definition in their 2023 paper on bookmaker margins. Their example: a -110 / -110 market holds 4.54% by this formula, against 4.76% for the shortcut of booksum minus 100. The exact figure is 4.545%, which the 20-cent table below rounds to 4.55%.
Why the Cents Widen on Big Prices
A cent buys less probability as the price grows. Ten cents near even money moves the break-even rate 2.38 points. At -500, the same ten cents moves the rate 0.28 points.
| Price change | Break-even before | Break-even after | Shift |
|---|---|---|---|
| -110 to -100 | 52.38% | 50.00% | 2.38 pts |
| -150 to -140 | 60.00% | 58.33% | 1.67 pts |
| -200 to -190 | 66.67% | 65.52% | 1.15 pts |
| -300 to -290 | 75.00% | 74.36% | 0.64 pts |
| -500 to -490 | 83.33% | 83.05% | 0.28 pts |
| +150 to +160 | 40.00% | 38.46% | 1.54 pts |
| +200 to +210 | 33.33% | 32.26% | 1.08 pts |
| +300 to +310 | 25.00% | 24.39% | 0.61 pts |
| +500 to +510 | 16.67% | 16.39% | 0.27 pts |
A book holding a fixed 20 cents at every price would give its margin away on big favorites. Here is what a strict 20-cent line keeps per dollar at each level.
| 20-cent line | Booksum | Cost per $1 |
|---|---|---|
| -110 / -110 | 104.76% | 4.55% |
| -150 / +130 | 103.48% | 3.36% |
| -200 / +180 | 102.38% | 2.33% |
| -300 / +280 | 101.32% | 1.30% |
| -520 / +500 | 100.54% | 0.53% |
A 20-cent line at -520 keeps 0.53% per dollar. DraftKings posts -520 / +390 and keeps 4.10%. The 110 cents taken off the Giants' price lift the margin from 0.53% to 4.10%.
The same math tells you where to shop. At a true 40% win chance, moving from +150 to +160 turns a $0.00 bet into +$4.00 per $100 staked. At a true 60%, moving a favorite from -150 to -140 adds $2.86 per $100 staked. Compare prices hardest between -200 and +200, where each cent moves the most probability.
Where the Vig Sits: Three Models, Three Answers
The booksum tells you how much the book keeps. The booksum does not say which side pays. You need a model for the split. Three common ones:
- Proportional: divide each implied probability by the booksum.
- Additive: subtract half the overround, in percentage points, from each side.
- Power: raise each implied probability to one exponent k, chosen so the two sides sum to 100%.
The proportional model charges both sides the same
Fair probability = implied / B
Decimal payout = 1 / implied
Return per $1 = (implied / B) × (1 / implied) - 1 = 1 / B - 1
Giants at Texans, both sides: 1 / 1.0428 - 1 = -4.10%
The proportional model, the method our no-vig calculator uses, assumes every side loses the same amount per dollar. Hegarty and Whelan call this the overround formula and write the assumption out: the expected payout is equal on every outcome. The other two models move the cost onto the underdog. The de-vig guide walks through each method in probability terms. Here is the same split in dollars.
| Market | Model | Fair favorite | Fair underdog | Favorite cost per $1 | Underdog cost per $1 |
|---|---|---|---|---|---|
| Bengals -115 / Ravens -105 | Proportional | 51.08% | 48.92% | 4.50% | 4.50% |
| Additive | 51.13% | 48.87% | 4.40% | 4.60% | |
| Power | 51.16% | 48.84% | 4.35% | 4.65% | |
| Bruins -192 / Sharks +160 | Proportional | 63.09% | 36.91% | 4.04% | 4.04% |
| Additive | 63.65% | 36.35% | 3.21% | 5.48% | |
| Power | 63.93% | 36.07% | 2.78% | 6.21% | |
| Vikings -290 / Colts +235 | Proportional | 71.36% | 28.64% | 4.04% | 4.04% |
| Additive | 72.25% | 27.75% | 2.83% | 7.05% | |
| Power | 72.73% | 27.27% | 2.19% | 8.64% | |
| Texans -520 / Giants +390 | Proportional | 80.43% | 19.57% | 4.10% | 4.10% |
| Additive | 81.73% | 18.27% | 2.55% | 10.48% | |
| Power | 82.47% | 17.53% | 1.67% | 14.11% |
Power split on Giants at Texans
Solve (520/620)^k + (100/490)^k = 1: k = 1.0957
Texans fair: (520/620)^1.0957 = 82.47%. Giants fair: (100/490)^1.0957 = 17.529%
Giants cost: 1 - 0.17529 × 4.90 = 14.11%
Texans cost: 1 - 0.8247 × 1.1923 = 1.67%
On Bengals at Ravens, each side's cost moves 0.15 points across the three models. On the Giants, the cost of one +390 ticket runs from 4.10% to 14.11%. The model you pick decides whether the Giants side is an ordinary bet or the worst price on the board.
What the Evidence Says About Who Pays
A large modern sample from fixed-odds markets outside the U.S. points at the longshot. Hegarty and Whelan studied 84,230 European soccer matches from 22 leagues, 2011/12 through 2021/22, plus 58,112 ATP and WTA tennis matches, at average closing odds across bookmakers. Their results:
- Soccer bets in the lowest decile of win probability lost 17% on average. Bets in the highest decile lost 2%.
- Tennis bets in the lowest decile lost 23%. The highest decile lost 3%.
- Across all soccer bets, the realized loss rate was 7.8% against 6.5% predicted by the overround. In tennis, their Table 2 shows 7.4% against 5.4%.
The power split on Giants at Texans, 14.11% against 1.67%, is the same order as the soccer gap of 17% against 2%. Whelan's 2024 paper in Economica explains the mechanism. Adding a margin drives more favorite bettors out of the market than longshot bettors. A risk-averse book then worsens longshot odds to cut risk, and the bias appears even when bettors judge probabilities correctly on average.
The U.S. moneyline studies cited here are mostly older, and they split.
- Linda and Bill Woodland's 1994 study of MLB betting in the Journal of Finance found the racetrack pattern in reverse: favorites, not underdogs, were the worse bet. The gap was too small to profit from after commission.
- Their 2003 update added the 1990 to 1999 seasons and found the reverse bias virtually unchanged.
- Gandar, Zuber, Johnson, and Dare (Applied Economics, 2002) re-examined the reverse-bias finding with corrected tests. Slight underdogs still looked underbet, but returns rose with the odds-implied win probability for favorites and underdogs alike, a pattern closer to the standard bias. They found too little evidence to call the pattern a true inefficiency.
- The Woodlands' 2001 NHL study, covering 1990 to 1996, found hockey bettors overbet favorites, and simple betting strategies turned a profit.
- Nutaro's 2023 baseball paper in Risks, using run differential as the measure of team strength, found the standard bias: bettors rate favorites too low and longshots too high.
Read together, these studies do not settle which split fits a 2026 DraftKings hockey or baseball board. The honest position: the split is unknown. Price both ends.
The Worst-Case Test
When the split is unknown, test each bet against the model least kind to the bet. For an underdog, the power split assigns the lowest fair probability. For a favorite, the proportional split does. A bet worth making clears both.
Suppose a second book posts these prices against the DraftKings lines above. Expected value uses each model's fair probability from the DraftKings market.
| Offered price | Proportional | Additive | Power | Verdict |
|---|---|---|---|---|
| Giants +450 | +7.64% | +0.48% | -3.59% | Pass |
| Texans -450 | -1.70% | -0.11% | +0.80% | Pass |
| Sharks +175 | +1.49% | -0.03% | -0.80% | Pass |
| Ravens +105 | +0.28% | +0.17% | +0.12% | Thin edge under all three |
Giants +450 at a second book
Decimal payout at +450: 1 + 450 / 100 = 5.50
Proportional: 0.19571 × 5.50 - 1 = +7.64%
Power: 0.17529 × 5.50 - 1 = -3.59%
The +450 Giants ticket looks like a 7.6% edge on a proportional no-vig tool. Under the power split, the same ticket loses 3.6%. A bet surviving only under the friendliest model is a bet on the model, not on the game. Ravens +105 clears all three models by a hair, because near even money the model fight disappears. The test bites on big prices, where the cents gap widens and the split matters most.
What a Moneyline Settles On
A price only compares to another price with the same settlement rule. DraftKings' Massachusetts house rules, implementation date October 7, 2026, and listed as the active version on the Massachusetts Gaming Commission site, set these terms:
- A 2-way moneyline offers no tie or draw selection. If the teams tie, the bet settles as a push and your stake comes back.
- A 2-way moneyline includes overtime and shootouts. Under the general rules, a 3-way moneyline, which the rules also call the 60 Minute Line in hockey, excludes overtime and shootouts. Football is the exception: the football section says 3-way moneyline bets include overtime.
- Soccer bets settle on regulation time, including injury and stoppage time. Extra time does not count, except in extra-time and futures markets.
- In MLB, the projected starting pitchers on the board are informational. A starter change does not void a moneyline bet, except in markets labeled Pitcher Must Start.
Three practical rules follow. Never de-vig a 3-way hockey or soccer line as a 2-way market, because the draw takes its own share of the probability. Expect the favorite's price on a 60-minute line to sit longer than on the 2-way line, since a regulation tie loses the bet. And in baseball, a late scratch of the listed ace leaves your bet live at the old price, so bet early only when your number holds without the starter.
Calculator: Where the Vig Sits on Your Line
Enter both sides of a two-way moneyline from the sharpest book you trust. Add a price you found elsewhere for one side. The calculator shows the booksum, each model's fair odds and cost per dollar, and the expected value of your price under all three models, with the worst case.
Moneyline vig split and worst-case test
The Moneyline Checklist
- Convert the price to a break-even rate before you bet. If your own win estimate sits below the break-even rate, pass.
- Check the booksum. 23 of the 28 DraftKings two-way moneylines checked here sit between 104.07% and 104.34%. A market well above 105% carries a worse margin than the board.
- On any line past -200 or +200, run the worst-case test. Underdogs need an edge over the power fair price. Favorites need an edge over the proportional fair price.
- Shop hardest near even money. Ten cents there moves the break-even rate up to 2.38 points.
- Match the settlement rule before you compare two prices: 2-way or 3-way, overtime or regulation, action or listed pitchers.
- Size the bet from the edge you trust, not the edge the friendliest model shows. The Kelly calculator handles the sizing.
- Track your price against the close. The CLV guide explains the tracking.
Read the free picks and see how the calls get made in the free Discord room. No card needed.
The verdict: a moneyline price is a win probability plus a margin, and DraftKings holds the margin between 3.9% and 4.5% per dollar whether the game is a near pick'em or a -625 mismatch. The cents gap widens from 20 to 170 to keep the percentage flat. Where the margin sits is a model choice, and on a +390 underdog the choice moves the cost from 4.10% to 14.11%. Near even money, ignore the split and shop cents. On big prices, bet only what clears the least friendly model.
Sources and Method
Moneyline prices come from DraftKings boards on ESPN's NFL, NBA, NHL, and MLB odds pages, read twice about 10:25 p.m. ET on October 23, 2026, for every game October 24 to 26 the pages listed with both sides posted, plus the World Series winner market. Lines move, so recheck any price before you bet. Settlement terms come from DraftKings' Massachusetts house rules, implementation date October 7, 2026, hosted by the Massachusetts Gaming Commission. The 20-cent convention comes from the Covers guide on NBC Sports. Hegarty and Whelan's 2023 paper supplies the margin definition and the soccer and tennis loss rates by decile. Whelan's 2024 Economica paper supplies the risk-aversion mechanism. The Woodlands (1994, 2001, 2003), Gandar and coauthors (2002), and Nutaro (2023) supply the U.S. evidence. Every break-even rate, booksum, cost, fair price, and expected value above comes from the formulas shown, and you recompute each one in a spreadsheet. For the power split, use Goal Seek on k.
Limits: the second-book prices in the worst-case table are hypothetical examples, not posted lines. The three models do not exhaust the field. The Shin method, covered in the de-vig guide, matches the additive fair prices to the hundredth of a point on all four two-way lines in the model table. Cost per dollar assumes the fair line comes from the same market you bet into. If you hold a better estimate than the market, use your own probability instead of any fair line.