Skip to content

Middle Calculator

Two opposite bets at different numbers. This maps the window where both of them win, and what every other result costs.

Over — wins when the number goes higher

$

Under — wins when the number stays lower

$

The middle

43.5 – 47.5

4 whole numbers land inside, and both bets win on every one of them — $181.82. Every other result costs you $9.09 or less.

Best case
+$181.82

Both bets win.

Worst case
−$9.09

One bet wins, the other loses. This is the normal outcome.

Total at risk
$200.00
Middle hit rate to break even
4.76%

How often the number has to land inside for this position to be worth taking.

What each region of the result pays
ResultOverUnderNet
Total below 43.5−$100.00+$90.91−$9.09
Total inside the middle+$90.91+$90.91+$181.82
Total above 47.5+$90.91−$100.00−$9.09

This is a settlement table, not a probability. How often a game lands in a four-point window depends on the sport, the pace and the scoring distribution — and this calculator will not invent that number for you. Take the break-even rate above and judge it against what you actually know about the matchup.

Runs entirely in your browser · nothing is sent or saved

What is a middle bet?

Middling means holding both sides of a market at different numbers, so there is a range of results where both bets win.

Take Over 43.5 early, and after the line moves take Under 47.5. Any final total of 44 through 47 wins both tickets. Outside that window one side wins and the other loses, which costs only the margin on the loser.

That is the shape: a small, frequent loss against a large, occasional double win. Two −110 bets at $100 lose $9.09 when the middle misses and win $181.82 when it hits, so it only needs to land about one time in twenty-one to be worth having.

How it works

A middle is holding both sides of a market at different numbers, so there is a range of results where both bets win. Take Over 43.5 and Under 47.5 and any final total of 44 through 47 pays both tickets. Outside that window one wins and the other loses, which costs you only the margin on the losing side.

That is the whole appeal: a small, frequent loss against a large, occasional double win. Two bets at −110 with a $100 stake each lose $9.09 when the middle misses and win $181.82 when it hits, so the position only needs to land inside about one time in twenty-one to be worth having.

Middles usually appear over time rather than all at once. You bet an early number, the line moves several points, and the other side is now available at a number that leaves a gap. Chasing a middle by taking both sides at once almost never works, because the two prices together include two margins and the window is rarely wide enough to pay for them.

The rows to read carefully are the push rows. On whole-number lines a result landing exactly on your number is a genuine third outcome — one bet pushes, the other settles — and it pays differently from both the hit and the miss. That is also why the break-even rate is sometimes withheld: a single break-even number only exists when every non-middle result pays the same, and an exact-line push breaks that. Rather than inventing a probability split across the outside regions, the calculator says so.

The formula

each leg settles independently:
  win   → stake × (decimal − 1)
  push  → 0        (stake returned)
  loss  → −stake

net at any result = leg A P/L + leg B P/L

the middle window = (lower line, higher line)
whole numbers inside = the results that pay both

when every non-middle row pays the same L,
and the middle pays M:

  break-even middle rate = −L / (M − L)

otherwise it is undefined without assuming
how the outside results split.

Modelled as a settlement table rather than a formula, because a middle has three or four distinct regions and collapsing them hides the push rows.

A worked example

Over 43.5 −110 for $100, and later Under 47.5 −110 for $100. The middle is 44 to 47 — four numbers.

A total of 45 pays both: $90.91 twice, for +$181.82. A total of 40 wins the Under and loses the Over: −$9.09. A total of 52 does the reverse: also −$9.09.

So the break-even rate is 9.09 / 190.91 = 4.76%. The question is whether four specific numbers in an NFL total come up more than one time in twenty-one — and given how much of the scoring distribution sits in the low-to-mid forties, that is often a comfortable yes.

Change the Over to a whole number, 44, and a total of exactly 44 becomes a third outcome: the Over pushes and the Under wins, for +$90.91. That row is why the single break-even figure disappears.

Common questions

How often does a middle need to hit?
At two −110 legs with equal stakes, about 4.76% — you lose $9.09 on a miss and win $181.82 on a hit. Wider windows and better prices lower the bar further. The calculator gives the exact figure whenever it is defined.
Should I take both sides at once?
Usually not. Betting both sides simultaneously means paying both margins with no line movement in your favour, and the window is rarely wide enough to cover that. Middles are normally created by a line moving after you have already bet one side.
Why is the break-even rate sometimes missing?
Because it is only defined when every non-middle result pays the same. Different prices on the two sides, or a whole-number line that can push, create outside results that pay different amounts — and quoting one number would mean inventing a probability split the calculator has not been given.
What is a "middle" on a spread?
The same idea. Take +7.5 with the underdog and later −3.5 with the favourite, and any margin of 4 through 7 wins both. Enter the underdog line as the lower number and the favourite line as the higher one.
Does this tell me the probability of hitting?
No, and deliberately. That depends on the sport, the pace and the scoring distribution around your specific numbers — key numbers in football, for instance, are far more likely than the ones beside them. The calculator gives you the rate you need and leaves the estimate to you.

The guide behind this calculator

Related calculators

For informational and analytical purposes only. These tools do not predict outcomes and do not recommend wagers.