Reading the House Edge and Vig on Crypto Sportsbooks

Loading...

Author: Austin Outhwaite Last updated: June 2026 Reading time : 11 min

Why “-108” Is Not a Bonus, It’s a Number to Verify

Affiliate review sites love to quote “-108 lines” as proof that a crypto book is offering sharper prices than the competition. I was reviewing exactly this claim for a reader last year and checked the underlying market. The book had -108/-108 on one specific marquee match while running -115/-115 on the same game’s secondary markets and -120/-120 on the props. The marketing used the sharp line, the actual average margin across the book was miles wider. That’s the pattern this article is about. Headline odds don’t mean anything without the arithmetic behind them.

The vig — or vigorish, or hold, or overround — is the structural margin a sportsbook charges for accepting a bet. On a two-way market priced fairly, the implied probabilities of the two sides should sum to exactly 100 per cent. On a book with 5 per cent vig, they sum to 105 per cent. That 5 per cent is the book’s expected return on every dollar of action, across a long enough sample. Understanding vig is the foundation of understanding whether any given book is actually good value.

During 2024 the average crypto bet size grew 1.4× while the average fiat bet size didn’t change, and altcoin stakes grew from 26.8 per cent of crypto volume to 47 per cent. Each of those shifts means more bettors making larger decisions, which raises the value of actually understanding what you’re paying the book per transaction. Headline odds are marketing. The vig calculation is the reality.

The Vig Formula on a Two-Way Market

Let me build up the math from scratch, because once the formula is internalised it takes 15 seconds to check any book’s margin on any market.

A two-way market has two mutually exclusive outcomes — Team A wins or Team B wins, with no draw possibility. The book posts odds on each side. Convert each odd to implied probability, sum the two, and the amount the sum exceeds 100 per cent is the overround. The hold — the book’s actual margin per bet — is the overround divided by the sum of implied probabilities.

Worked example with decimal odds. Book quotes Team A at 2.10 decimal and Team B at 1.80 decimal. Implied probabilities are 1/2.10 = 47.6 per cent and 1/1.80 = 55.6 per cent. Sum is 103.2 per cent. Overround is 3.2 per cent. Hold is 3.2/103.2 = 3.1 per cent. That’s a tight line; most crypto books run fatter.

Same market on a tighter book: 2.05 and 1.85. Implied 48.8 per cent and 54.1 per cent. Sum 102.9 per cent. Overround 2.9 per cent. Hold 2.8 per cent. Marginally tighter than the first book.

Same market on a wider book: 1.95 and 1.75. Implied 51.3 per cent and 57.1 per cent. Sum 108.4 per cent. Overround 8.4 per cent. Hold 7.8 per cent. Much worse for the bettor.

Now convert to American. -110/-110 is 52.4 per cent and 52.4 per cent. Sum 104.8 per cent. Overround 4.8 per cent. Hold 4.6 per cent. This is the standard US sportsbook line and the reference point most American bettors implicitly compare against. -108/-108 is 51.9/51.9 = 103.8 per cent sum, 3.8 per cent overround, 3.7 per cent hold. The sharper book gives back about 1 per cent of margin, which sounds small but compounds meaningfully over many bets.

The hold number is the one that actually matters. A 3 per cent hold book is sharper than a 5 per cent hold book. Across 100 bets of equal size on each, the difference in expected loss is 2 per cent of your total stake. That’s real money over a full year of active betting.

Three-Way Markets and the Draw Tax

Football, cricket, and some esports can draw, which requires three-way markets. The vig arithmetic extends naturally but you need to add three implied probabilities and the sum should be 100 per cent on a fair book.

Example: football match with home at 2.40, draw at 3.20, away at 3.00. Implied probabilities are 1/2.40 = 41.7 per cent, 1/3.20 = 31.3 per cent, and 1/3.00 = 33.3 per cent. Sum is 106.3 per cent. Overround 6.3 per cent. Hold 5.9 per cent. That’s a typical soccer market on a crypto sportsbook.

Three-way markets almost always carry higher hold than equivalent two-way markets. Books charge more for the additional pricing complexity and the difficulty of hedging three independent outcomes. A 6 per cent hold on a three-way soccer market is average; 4 per cent hold is sharp; 8 per cent or more is expensive.

The draw probability itself is where the book’s model gets most interesting. Real draw rates in top-tier European football sit around 24-26 per cent across a long sample. A book showing a draw at 3.20 decimal implies 31.3 per cent, which is a meaningful overround on the draw leg specifically. The book isn’t wrong to overshade the draw — the implied probability includes the book’s margin, and bettors are less likely to hammer the draw than the other outcomes, so the book can get away with fatter margin on the draw. But if you specifically like betting draws, you’re paying the highest concentrated margin on those bets.

Cricket three-way markets — home, away, tie — have their own quirks because actual tie probabilities in limited-overs cricket are very low (often under 2 per cent in ODIs, extremely rare in T20). Books still have to price the market, and the implied tie probability on a cricket three-way is typically 3-5 per cent. That spread of implied over actual probability is nearly pure margin on the tie leg.

Comparing Crypto Books Without Getting Fooled

Once you can calculate hold on a single market, the next step is comparing across books. The comparison is only as valid as your sample — one market on one game on one day tells you nothing about the book’s general pricing.

Sampling methodology that actually works. Pick 20-30 markets across a variety of sports — main-line winners, spreads, totals, maybe a handful of props. Calculate hold on each one for each book you’re comparing. Take the median hold across the sample. That median is a better picture of the book’s real margin than any individual market.

What goes wrong with naive comparison. A book might post one razor-sharp headline market for marketing purposes — a heavyweight boxing main-event moneyline at -102/-102, for example — while charging 7 per cent hold on every other market in the book. Reviews citing the -102 line as evidence of sharp pricing are technically accurate and practically misleading.

Hold varies predictably by market type. Main-line moneylines and spreads tend to carry the lowest hold — 4-6 per cent at typical crypto books, 2-4 per cent at the sharpest. Totals are similar. Props carry higher hold — 6-10 per cent on team props, 8-12 per cent on player props, 12-20 per cent on specials and novelties. Same-game parlays have effective hold well above 20 per cent when you correct for correlation between the legs. Knowing where each book concentrates its margin lets you interpret their advertised sharpness honestly.

Limits matter as much as prices. A book with 4 per cent hold but a A$200 max stake is effectively useless for serious action — you can’t size up enough to make the sharp price matter. A book with 5 per cent hold but A$20,000 max stake is more useful for anyone betting beyond pocket change. The hold number and the limit number need to be read together.

The cash-out margin deserves its own line. Books with live cash-out functionality typically pad the cash-out offer with an additional 5-15 per cent margin on top of the main-market vig. This is hidden margin most bettors don’t calculate. I’ve covered the cash-out math specifically in the piece on live Bitcoin betting, latency and cash-out rules, where the formula lives in detail.

Where the Margin Hides: Props, Parlays, Cash-Out

The book’s margin isn’t evenly distributed across markets. Understanding where it concentrates helps you avoid the highest-cost products and gravitate towards the cleaner ones.

Player props are the highest-margin mainstream market type. A player’s points-over-under at a crypto book might show 1.90/1.90 decimal, which looks like -110/-110 American and implies a 4-5 per cent hold. But player props often have juice distributed asymmetrically — the “easier” side at 1.85 and the “harder” side at 1.95, producing a wider effective spread once you factor in which side actually gets your action. More importantly, the book’s confidence in its prop pricing is lower, so the stated hold understates the book’s expected return because the lines aren’t sharp to true probabilities.

Same-game parlays are where the hold becomes nearly invisible and very large. The book prices each leg as if independent and multiplies the odds together. If the legs are correlated — they almost always are on same-game parlays — the true combined probability is higher than the product of individual implied probabilities, and the book pockets the correlation difference. A 3-leg same-game parlay priced as if independent can easily have effective hold of 20-30 per cent.

Cash-out is structural extra margin. Every cash-out offer carries the main-market vig plus an additional cash-out spread. The additional spread is typically 5-15 per cent and exists because the book needs to make a profit on the service of buying out your bet. Mathematically, almost no cash-out is fair to the bettor from an expected-value perspective.

Bonus bets and free bets have complex effective hold calculations. A bonus bet of A$100 that only pays winnings (not stake) has an effective value of about 70-80 per cent of the stated amount, depending on the odds you use it at. Wagering requirements on bonuses further reduce the effective value. A A$100 bonus with a 5× rollover requirement at a 5 per cent hold book has an expected net value of about A$75 minus the rollover cost, not A$100. Books advertise the gross amount.

Reduced-juice promotions are the one area where the advertised benefit is usually accurate. “Reduced juice Sundays” at some books genuinely mean -105/-105 or better during specific windows. These are loss-leaders designed to bring bettors in, and if you time your action to these promotions, you’re taking advantage of a real edge rather than a marketing gimmick. The catch is that the promotions usually run on specific markets or days, not across the entire book’s menu.