Vig and hold: same “fee,” totally different math
If you’ve ever compared books by checking whether a spread is -110 vs -110, you’re doing the normal thing. It’s also where a lot of bettors get quietly taxed.
Sportsbooks don’t really make their money from “being right.” They make it from pricing. Two numbers matter:
- Vig (also called juice): the built-in fee on a single market (like Bears -3 -110 / Opponent +3 -110).
- Hold: what the book expects to keep across a set of bets over time (a game, a day, a season, a sport, your entire account… depends how you define the bucket).
Here’s the clean way to think about it: vig is the markup on one price tag. hold is the store’s profit margin across all receipts. A grocery store can have a 30% markup on candy bars (high vig market) but still run a 2% overall profit margin (hold) because people also buy milk, eggs, and a bunch of low-margin stuff. Same idea with books: some markets are priced tight, others are a damn toll road.
And here’s the kicker: sportsbooks rarely advertise either number clearly. They’ll show you the odds, they’ll show you “boosts,” but they won’t show you the effective markup or what they expect to keep when the dust settles.
You can estimate both from odds, though. And once you can, comparing books becomes more than “-110 vs -110.”
Start from scratch: implied probability (the only translation you need)
Odds are just probability in a costume. To measure vig or hold, you first translate odds into implied probability (the probability the odds suggest).
For American odds:
- Negative odds (like -110): implied probability = |odds| / (|odds| + 100)
- Positive odds (like +120): implied probability = 100 / (odds + 100)
Example: a standard point spread at -110.
Implied probability = 110 / (110 + 100) = 110 / 210 = 0.5238 = 52.38%
If both sides are -110 (Team A -110, Team B -110), the implied probabilities add to:
52.38% + 52.38% = 104.76%
That extra 4.76% above 100% is the overround (also called the vig in practical conversation). It’s the book saying, “If you want to bet this market, you’re paying rent.”
One more quick example: a moneyline with asymmetrical odds.
Say: Team A -130, Team B +120.
- -130 implied = 130 / 230 = 56.52%
- +120 implied = 100 / 220 = 45.45%
Total = 56.52% + 45.45% = 101.97%
Overround ≈ 1.97%. That’s a much cheaper market than -110/-110, even though neither side is “-110.” This is why eyeballing juice without doing the translation is how you get fooled.
What vig actually is (and how to estimate it in 30 seconds)
When bettors say “vig,” they usually mean one of two things:
- Overround: the sum of implied probabilities minus 100%.
- Equivalent hold in a 2-way market: roughly what the book expects to earn if action is balanced.
For pricing comparisons, overround is your friend because it’s clean and it comes straight from the odds.
Two-way market example (spread):
Bears -3 -110
Packers +3 -110
We already did it: total implied = 104.76%, so overround = 4.76%.
Same market at a better book:
Bears -3 -105
Packers +3 -105
-105 implied = 105 / 205 = 51.22%
Total implied = 102.44%
Overround = 2.44%
That difference (4.76% vs 2.44%) is massive. If you bet a lot, it’s the difference between “maybe you can beat this” and “you’re paying a recurring subscription fee to lose.”
Three-way markets (soccer 1X2) get ugly fast. If you see:
Home +140 (41.67%)
Draw +230 (30.30%)
Away +190 (34.48%)
Total implied = 41.67 + 30.30 + 34.48 = 106.45%
Overround = 6.45%
That’s not “a little extra.” That’s a full-on tax. Recreational bettors get crushed here because they’re betting a market with a bigger built-in fee and they don’t even realize it.
If you want a quick way to sanity-check these calculations, ThunderBet’s Betting Assistant is handy for converting odds to implied probability and checking overround without fat-fingering a calculator.
Hold is not vig: it’s what happens after the bets come in
Vig lives in the odds. Hold lives in the results.
Hold (definition): the percentage of total handle (money wagered) that the book keeps after paying winners.
Formula:
Hold % = (Handle − Payouts) / Handle
Simple example. A sportsbook takes $10,000 in bets on one NFL game.
- If it pays out $9,500 total, hold = (10,000 − 9,500) / 10,000 = 5%.
- If it pays out $10,200 total (yes, books can lose), hold = (10,000 − 10,200) / 10,000 = -2%.
Here’s why you should care: a market can have a 4.76% overround and still produce a 1% hold that day if the public hammered one side and that side wins. Or it can have a 2.44% overround and produce a 10% hold if the book got a dream result.
That’s why “the vig is 4.76%” and “the book held 7% last weekend” are not contradictory statements. They’re different measurements.
Analogy that sticks: vig is the cover charge at the door. Hold is what the casino made after everyone played blackjack, roulette, and slots all night. The cover charge matters, but the casino’s nightly profit depends on where the crowd went and what happened.
As a bettor, you use vig to judge price quality and hold to judge how a book is shaping action (and which markets they’re comfortable tightening vs padding).
A simple full-picture example: same vig, different expected hold
Let’s build the whole idea with one clean scenario.
Game has a true 50/50 outcome (a coin flip). Book posts both sides at -110.
Step 1: Find the vig (overround).
Each side implies 52.38%. Total 104.76%. Overround = 4.76%.
Step 2: What’s the expected hold if action is balanced?
Assume $1,000 bet on each side (handle = $2,000).
At -110, a $1,000 bet wins $909.09 profit (because 1000 / 1.10 = 909.09). The winner gets back stake + profit = $1,909.09.
If Team A wins, book collects $1,000 from Team B bettors, pays $1,909.09 to Team A bettors.
Net to book = 2,000 − 1,909.09 = $90.91
Hold = 90.91 / 2,000 = 4.545%
Notice the small difference: overround was 4.76%, but the balanced-action hold works out to ~4.55%. Close enough that most people casually call it “the vig,” but technically they’re not identical.
Step 3: Same vig, different hold when action is lopsided.
Public piles $1,800 on Team A, only $200 on Team B (still -110 both sides). Handle is still $2,000.
- If Team A wins (the public side): payout = 1,800 + (1,800 / 1.10) = 1,800 + 1,636.36 = $3,436.36. The book only collected $2,000. Hold = (2,000 − 3,436.36) / 2,000 = -71.82%. Absolute bloodbath for the book.
- If Team B wins (the unpopular side): payout = 200 + (200 / 1.10) = 200 + 181.82 = $381.82. Hold = (2,000 − 381.82) / 2,000 = 80.91%. Christmas morning for the book.
Same odds. Same vig. Wildly different hold depending on which side wins.
That’s why books obsess over managing exposure, why they shade popular teams, and why you should stop thinking “vig” and “hold” are interchangeable.
How to estimate hold from odds (and why it’s only an estimate)
You can’t know a book’s actual hold without seeing the bets (handle distribution) and payouts. Books keep that private for a reason.
But you can estimate an expected hold from the odds under a simple assumption: balanced action across outcomes. It’s not always true, but it gives you a baseline to compare markets and books.
For a 2-way market (spread, total, moneyline without draw):
- Compute implied probabilities for both sides.
- Normalize them to remove vig (divide each by the total implied probability).
- Assume $1 of handle is split according to those normalized probabilities (a “fair” split).
- Compute expected payout using the odds, then back into expected hold.
That sounds nerdy, but here’s the practical shortcut: lower overround almost always means lower expected hold in balanced action. So if you’re shopping lines, you can treat overround as your fast proxy for “how expensive is this market?”
Where bettors screw this up is comparing only the headline number. Example:
- Book A offers NBA spread at -110/-110 (4.76% overround).
- Book B offers the same spread at -112/-108 (implied 52.83% + 51.92% = 104.75% overround, basically the same).
Those are basically equally “expensive” markets in aggregate. But if you always bet favorites, Book B quietly taxes you more. That’s not theoretical. That’s your actual betting profile getting charged extra.
This is why you should track your own behavior. If you mostly bet one side type (favorites, overs, props), the “average vig” of the market matters less than the vig you personally pay.
If you want to go one level deeper without turning this into a math class, plug a few lines into the Betting Assistant and compare overround across books for the exact market you bet. That’s line shopping with receipts.
Spotting overpriced markets (and comparing books beyond “-110 vs -110”)
If you only take one thing from this, take this: books choose where to be tight and where to be fat. Your job is to stop donating in the fat markets.
Here’s a checklist you can actually use:
- Convert odds to implied probability and add them up. If it’s a 2-way market and you’re seeing 105%+ regularly, you’re paying a premium.
- Watch for 3-way and niche markets (soccer 1X2, player props, alt lines). These often carry higher overround because fewer bettors can price them accurately.
- Compare the same market across books, not “spread vs moneyline.” If Book A is -105/-105 on NHL totals and Book B is -115/-105, Book A is your totals shop.
- Check how fast a book reprices. Books that move faster tend to offer tighter numbers more often, because they’re less afraid of getting picked off. Right now, you can see constant movement activity from places like Novig (73), ProphetX (62), Pinnacle (53), and Matchbook (41), with mainstream books like BetMGM (32) and ESPN BET (32) also showing plenty of re-pricing. Fast movers aren’t automatically “better,” but slow movers love to hide extra margin in stale lines.
- Be suspicious of “can’t lose” parlays. Parlays multiply vig and then some, especially when the book tweaks correlation rules. If you like building parlays, read Parlay Builder: Catch Correlated Legs Before You Donate (2026) and save yourself some pain.
If you want a quick refresher on sniffing out bad pricing fast, Vig vs Juice: The 30-Second Test for Bad Lines pairs perfectly with this topic.
And if you’re serious about line shopping as a long-term edge (not a hobby), you’ll like Line Shopping MLB: How 5¢ Turns Break-Even Into Profit. A few cents doesn’t sound like much until you realize you’re making that bet hundreds of times.
If you’re looking for more beginner-friendly stuff like this, the education hub at /blogs/education/ is where it lives.
Responsible gambling note: Treat betting like entertainment with a budget, not a bill you need to pay. If you’re chasing losses or stressing rent money, take a break.
