Middles
6 min read · Free
What a Middle Is
A middle is a situation where you bet both sides of a spread (or total), with each bet placed at a different number — and a gap between them. If the final score falls inside that gap, both bets win. If the score falls outside the gap, one bet wins and one loses, typically for a small net loss (the combined vig).
Middles require two books to disagree on the same line. This happens more often than you'd expect, especially early in a betting week before sharp money has moved lines toward consensus.
How Middles Occur
Spreads move as books respond to betting volume, injury news, weather, and sharp action. Two books don't always move at the same time or by the same amount — creating brief windows where they quote different numbers.
Simplified timeline
Both Book A and Book B open Chiefs −6.5.
Sharp money hits Book B. Book B moves to Chiefs −7.5. Book A hasn't adjusted yet — still at −6.5.
Book A: Chiefs −6.5. Book B: Chiefs −7.5. A 1-point gap exists. You can bet Chiefs −6.5 at Book A and Bills +7.5 at Book B. If the Chiefs win by exactly 7, both bets win.
A Worked Example
Line: Chiefs −6.5 at Book A (−110). Bills +7.5 at Book B (−110). Both bets: $110 to win $100.
In 3 of 4 scenarios, you lose ~$10 (the combined vig from both sides). In the middle scenario (Chiefs win by exactly 7), you collect +$200.
Whether this bet has positive expected value depends on the probability that the Chiefs win by exactly 7. The NFL final margin distribution has been studied extensively — certain margins (3, 7, 10, 14) are significantly more likely than others. A 1-point gap centered on 7 has real probability.
The Math: Is a Middle Worth It?
For a middle to have positive expected value, the expected value of hitting the middle must exceed the cost of the combined vig:
Break-even probability formula
Miss: one leg wins +$100, the other loses −$110 ⇒ −$10.00
Hit: both legs win ⇒ +$200
Break-even: P × $200 = (1 − P) × $10
P(middle) = $10 ÷ $210 = 4.76%
A miss costs the vig on one bet, not both — the losing leg is paid for by the winning one, which is why the figure is −$10.00 and not double that.
So if the middle lands more than 4.76% of the time, the bet is positive-EV. For a Chiefs −6.5 / Bills +7.5 middle (a 1-point window centered on 7), the true probability is roughly 5–8% — above break-even at −110/−110 pricing. The lottery-ticket shape is real (you lose small most weeks), but the expected value is positive, which is exactly why books move these lines when they notice a gap.
At lower vig (−105/−105 at exchange-style books), break-even drops to 2.44%, widening the margin further. This is why low holds and middles are related concepts.
When Middles Make Sense
Gap is centered on a 'key number'
In NFL, scores of 3, 7, 10, 14, 17 occur most often. A gap including 7 or 3 is much more valuable than a gap between 8 and 9.
Gap is 2+ points wide
A 2-point gap (e.g., −6.5 vs +8.5) doubles the middle window and roughly doubles the probability — much better EV.
You can find low-vig pricing on both sides
At −101/−101, a miss costs about $1.00. Break-even probability drops to 0.50%, making even small-gap middles viable.
You're comfortable with variance
Middles are a lottery-ticket structure: small frequent losses, occasional large wins. If you need predictable income, standard matched betting is more appropriate.
Middles vs. Arbitrage
Arbitrage locks in profit because the combined implied probability is below 100%. A middle only locks in a loss if the score misses — but the potential upside when it hits can be worth the expected cost if you've correctly assessed the probability.
Most matched bettors use middles opportunistically: when a clear key-number gap appears at tighter pricing. They don't make middles a core strategy, but take advantage when the setup is favorable.