Guide · 6 min read
What is a middle bet? Middling in sports betting explained
A middle lets both of your bets win if the result lands between two lines. Here is how the maths works, when a middle is worth taking, and why it behaves very differently from a sure bet.
Updated 21 October 2026
Contents
What a middle is
A middle is a pair of bets on opposite sides of the same market, placed at different lines at different bookmakers. If the final result lands between the two lines, both bets win.
Two typical examples:
- Goal totals: Over 2.5 goals at Bookmaker A and Under 3.5 goals at Bookmaker B. A match with exactly three goals wins both bets.
- Handicaps: Dortmund −1.5 at Bookmaker A and Stuttgart +2.5 at Bookmaker B. If Dortmund wins by exactly two goals, both bets win.
The same idea works for ice hockey totals in the ICE Hockey League or the DEL, for handball and for basketball, where points are spread more evenly and windows are narrower.
Middles appear because bookmakers move lines at different speeds. One reacts to team news or heavy money and moves its main line from 2.5 to 3.0 goals, another has not caught up yet, and for a few minutes you can stand between them.
How the outcomes work
A standard middle has three possible results:
- Below the window: the Under wins, the Over loses.
- In the window: both bets win.
- Above the window: the Over wins, the Under loses.
Outside the window you have one winner and one loser, and the winner usually does not quite cover the loser, so a miss costs a small amount. When the middle hits, you collect two winning bets and the profit is large compared with the stake. That lopsided payoff is the whole point.
Worked example: two goal lines
Bookmaker A still offers Over 2.5 goals at 1.95. Bookmaker B has moved and offers Under 3.5 goals at 1.50. You split €200 so that both misses cost the same, using the same formula as for a sure bet (stake = total × (1/odds) / S, with S = 1/1.95 + 1/1.50 = 1.17949):
| Goals | Over 2.5: €86.96 | Under 3.5: €113.04 | Net |
|---|---|---|---|
| 0, 1 or 2 | −€86.96 | +€56.52 | −€30.44 |
| Exactly 3 | +€82.61 | +€56.52 | +€139.13 |
| 4 or more | +€82.61 | −€113.04 | −€30.43 |
A miss costs about €30. A hit pays about €139. Whether that is a good trade depends only on how often a match like this ends with exactly three goals.
Hit rate and expected value
Because a miss costs a little and a hit pays a lot, the question is how often the result lands in the window. The break-even hit rate is:
break-even p = miss loss / (hit profit + miss loss)
For the example: 30.44 / (139.13 + 30.44) = 30.44 / 169.57 ≈ 17.95%. If the match ends with exactly three goals more often than about one game in 5.6, the middle has positive expected value.
Expected value per middle is:
EV = p × hit profit − (1 − p) × miss loss
In a match where the market expects close to three goals, a simple goal model puts exactly three goals at roughly 22%. Then EV = 0.22 × 139.13 − 0.78 × 30.44 = 30.61 − 23.74 = +€6.87 per €200 staked, about 3.4%. At a 20% hit rate it is only +€3.47; at 17% it is already negative. The number moves quickly with the estimate, which is why the estimate matters more than anything else.
Hit rates depend heavily on the sport and the numbers involved. In football one whole goal is a big window, because low-scoring sports concentrate results on a few numbers. In basketball a three-point window lands only a few percent of the time, so the prices have to be much better to pay off.
ArbVantage shows middles with an estimated hit rate and expected value, not as sure bets. The estimates come from a statistical model of how results are distributed around the market line. They are estimates, and a single middle tells you almost nothing about whether the model is right. Judge them over hundreds of bets.
Negative middles
Most middles are "negative middles": both outcomes outside the window lose money, as in the example above. That is not a problem in itself. The problem is a negative middle whose hit rate is too low for its price.
Over 2.5 at 1.80 and Under 3.5 at 1.40, €200 split as before (€87.50 and €112.50). A miss now costs €42.50 and a hit pays €115.00.
Break-even: 42.50 / 157.50 ≈ 27%. With the same 22% hit rate, EV = 0.22 × 115 − 0.78 × 42.50 = 25.30 − 33.15 = −€7.85. Same match, same window, but a bet you should skip.
The window alone tells you nothing. Always compare the break-even hit rate with a realistic estimate before you place anything.
Whole and quarter lines
Half-goal lines like 2.5 cannot push, so the outcomes are clean. Whole-number lines can. If you hold Over 2.5 and Under 3.0, a match with three goals wins the Over and pushes the Under, so you get that stake back. That is a half-middle: better than a miss, not as good as a full hit.
Asian quarter lines split your stake in two. Over 2.75 is half on 2.5 and half on 3.0, so on exactly three goals the 2.5 half wins and the 3.0 half pushes. Quarter and whole lines create extra partial outcomes, so write down every possible result and its payout before you bet. A table like the one above takes a minute and removes most mistakes.
Arb-middles
The best case is when both prices are high enough that the outcomes outside the window still make a small profit. This is an arb-middle: a sure bet with an extra chance of a large win.
Say you find Over 2.5 at 2.10 at Bookmaker A and Under 3.5 at 1.95 at Bookmaker B. S = 1/2.10 + 1/1.95 = 0.98901, below 1. Split €200 as a normal two-way sure bet: €96.30 on the Over and €103.70 on the Under.
- 0 to 2 goals: the Under returns about €202.22, profit +€2.22.
- 4 or more goals: the Over returns about €202.23, profit +€2.23.
- Exactly 3 goals: both win, profit +€204.45.
That only holds if both legs are placed at these odds and stand. Arb-middles are rare and short-lived, so the usual sure-bet checks apply: confirm both lines and prices on the bet slip, and place the leg most likely to move first.
Risks and variance
A sure bet pays a small, roughly fixed amount whatever happens, if every leg stands. A middle does the opposite: most of the time you lose a little, and now and then you win a lot. Even with a 22% hit rate, ten misses in a row happen about once in every twelve runs of ten. That is normal, not a sign that something is broken.
- Variance. Size middles as a small share of your bankroll, not by how confident you feel. Judge results over hundreds of bets, not fifty.
- Model error. The edge lives entirely in the hit-rate estimate. Red cards, weather and team news shift goal distributions, so treat a middle more like a value bet than a sure bet.
- Line moves. If the second line moves before you place it, the window can shrink to nothing and you are left with a single bet.
- Settlement rules. Check that both bookmakers settle on 90 minutes plus stoppage time and treat abandoned matches the same way.
- Limits. Bookmakers notice accounts that keep betting stale lines, just as they notice sure bettors. See how bookmaker limits work.
Information, not betting advice. 18+. Betting involves risk and middles can lose for long stretches. If betting stops being fun, see our responsible gambling page.