Over/under 2.5 goals is a bet on the total number of goals both teams score in normal time: three or more wins the over, and two or fewer wins the under. No match can end on two and a half goals, so every bet on the line wins or loses in full, and a 2-1 or a 3-0 is enough for the over.
Over under goals betting asks the same question at other lines, from 0.5 to 4.5 goals and beyond, and Asian totals such as 2.25 and 2.75 split your stake across two lines. Over under betting covers totals in every sport, and soccer betting the other match markets.
How over/under 2.5 goals settles
| Score after normal time | Total goals | Over 2.5 | Under 2.5 |
|---|---|---|---|
| 0-0 | 0 | Loses | Wins |
| 1-1 or 2-0 | 2 | Loses | Wins |
| 2-1 or 3-0 | 3 | Wins | Loses |
| 4-2 | 6 | Wins | Loses |
| 2-0 at 90 minutes, 3-1 after extra time | 2 | Loses | Wins |
Example: Illustrative prices. Over 2.5 is $2.02 and under 2.5 is $1.82. $60 on the over returns $60 x 2.02 = $121.20 if a third goal comes, a $61.20 profit; $60 on the under returns $60 x 1.82 = $109.20 if it does not.
Every goal that stands when normal time ends counts, including goals in stoppage time, penalties scored in normal time and own goals, which count for the other side. Goals in extra time, kicks in a penalty shootout and goals the referee disallows do not. Bookmakers authorised in NSW work to the Bookmaker Declared Betting Events Betting Rules, under which a soccer bet is decided on the score when normal time ends, unless agreed otherwise (clause 5.2.7.3).
A match called off is the main way a 2.5 bet ends in a refund. If the governing body declares an official result, those rules refund every bet that depends on the number of goals (clause 5.2.7.2).
If it declares none, bets already decided when play stopped stand and the rest are refunded (clause 5.2.7.1). An over is decided once the total passes the line, an under only at the end. Abandoned and postponed matches compares other rule books.
Goal lines from 0.5 to 4.5
Each line asks a different question about the same goal count. A Poisson model turns an expected total into a chance for each count. The formula is P(k goals) = e^-m x m^k / k!, where m is the expected total, e is the constant 2.718 and k! is 1 x 2 x ... x k. With an illustrative expected total of 2.60 goals:
| Total goals | 0 | 1 | 2 | 3 | 4 | 5 | 6 or more |
|---|---|---|---|---|---|---|---|
| Chance | 7.4% | 19.3% | 25.1% | 21.8% | 14.1% | 7.4% | 4.9% |
Working for two goals: e^-2.6 x 2.6^2 / 2 = 0.0743 x 3.38 = 25.1%. Adding the counts gives every line, with each fair price worked from the unrounded chance:
| Line | Over wins on | Under wins on | Over chance | Fair over price | Fair under price |
|---|---|---|---|---|---|
| 0.5 | 1 goal or more | 0-0 | 92.6% | $1.08 | $13.46 |
| 1.5 | 2 or more | 0 or 1 | 73.3% | $1.36 | $3.74 |
| 2.5 | 3 or more | 0 to 2 | 48.2% | $2.08 | $1.93 |
| 3.5 | 4 or more | 0 to 3 | 26.4% | $3.79 | $1.36 |
| 4.5 | 5 or more | 0 to 4 | 12.3% | $8.16 | $1.14 |
For 2.5, under = 7.4% + 19.3% + 25.1% = 51.8%, so over = 48.2% and the fair over price is 1 / 0.4816 = $2.08. At this total the 2.5 line sits nearest even money; further out, one side becomes a short favourite and the other a long shot.
Asian totals: 2.25 and 2.75 split your stake
A whole-number total, such as 2.0 or 3.0, usually refunds your stake when the goals land exactly on it. A quarter total normally divides your stake into two equal halves, one on each neighbouring line. Asian handicap quarter lines follow the same convention, and your bookmaker's terms are where to confirm it. Some betting screens show a quarter total as the two lines it covers, so 2.25 appears as 2 and 2.5, and 2.75 as 2.5 and 3.
Take illustrative prices of $1.76 for over 2.25 and $2.32 for over 2.75, with $80 on each. Over 2.25 is $40 on over 2.0 and $40 on over 2.5:
| Total goals | $40 on over 2.0 | $40 on over 2.5 | Return | Profit or loss |
|---|---|---|---|---|
| 0 or 1 | lost | lost | $0 | -$80.00 |
| 2 | $40 back (push) | lost | $40.00 | -$40.00 |
| 3 or more | $40 x 1.76 = $70.40 | $70.40 | $140.80 | +$60.80 |
Over 2.75 is $40 on over 2.5 and $40 on over 3.0:
| Total goals | $40 on over 2.5 | $40 on over 3.0 | Return | Profit or loss |
|---|---|---|---|---|
| 0 to 2 | lost | lost | $0 | -$80.00 |
| 3 | $40 x 2.32 = $92.80 | $40 back (push) | $132.80 | +$52.80 |
| 4 or more | $92.80 | $92.80 | $185.60 | +$105.60 |
The unders are mirror images: under 2.25 half-wins on exactly two goals, and under 2.75 half-loses on exactly three. Against a plain over 2.5 at $2.02, over 2.25 takes a lower price for a softer result on two goals, and over 2.75 a higher price in exchange for winning only half on three.
Turning a total into scoreline chances
A total says how many goals, not who scores them. For scorelines, split the expected total between the sides, for example from the handicap or 1X2 prices, and multiply each side's Poisson chance. With the 2.60 split as 1.45 goals for Team A and 1.15 for Team B (illustrative):
| Total goals | Scorelines and their chances | Together |
|---|---|---|
| 0 | 0-0 7.4% | 7.4% |
| 1 | 1-0 10.8%, 0-1 8.5% | 19.3% |
| 2 | 2-0 7.8%, 1-1 12.4%, 0-2 4.9% | 25.1% |
| 3 | 3-0 3.8%, 2-1 9.0%, 1-2 7.1%, 0-3 1.9% | 21.8% |
Working for 2-1: (e^-1.45 x 1.45^2 / 2) x (e^-1.15 x 1.15) = 0.2466 x 0.3641 = 9.0%.
You can also run it backwards to read the market's own expected total. The illustrative $2.02 and $1.82 add to 1 / 2.02 + 1 / 1.82 = 104.5%, so scaling the over's 49.5% down gives 49.5 / 104.5 = 47.4%; the fair odds calculator runs the same sum on any market's prices. A Poisson total with 52.6% on two goals or fewer is about 2.57 goals, a shade under the 2.60 above.
Poisson distribution betting builds the full model and sets out its limits, such as draws and goals that do not arrive independently.
What moves a goal line before kick-off
- Team news. Losing the striker who takes most of a side's shots cuts its expected goals, and losing its first-choice goalkeeper can raise the other side's. Confirmed line-ups come out shortly before kick-off, which for European leagues falls in the Australian night or early morning. A price taken the evening before can look different by then; EPL betting covers the timing.
- Weather and the pitch. Heavy rain, a waterlogged or heavy pitch, strong wind or the heat of a summer A-League match can slow a game. Measure how much from past matches rather than assume it, and remember a forecast is public, so the line may already include it.
- What each side needs. A side that only needs a draw may sit deep, a dead rubber can open up, and the second leg of a two-legged tie is still settled on this match's normal time, not the aggregate.
- Where the money goes. A bookmaker can answer weight of money by moving the line, from 2.5 to 2.75, rather than the price, so compare lines before you compare prices.
Example: Illustrative numbers. Team B's main striker is ruled out shortly before kick-off, and your Team B figure falls from 1.15 to 0.90 goals, taking the total from 2.60 to 2.35. Under 2.5 becomes e^-2.35 x (1 + 2.35 + 2.35^2 / 2) = 0.0954 x 6.111 = 58.3%, so the fair over is 41.7%, or 1 / 0.4172 = $2.397, about $2.40. An over taken at $2.02 the night before now sits 2.02 / 2.397 - 1 = 15.7% below that fair price.
All of this has to be weighed before the game starts, because in Australia online betting on a sporting event is lawful only before it begins and only with a provider that holds an Australian state or territory licence (Interactive Gambling Act 2001, s 8A(3), s 10B and s 15AA).
Totals and BTTS together
A total and both teams to score ask different questions about the same goals. Over 2.5 can come from one side alone, as in 3-0, and the only result that is both under 2.5 and BTTS Yes is 1-1.
So the total alone cannot tell you BTTS; the split between the sides does. Two illustrative matches with the same expected total of 2.60 have the same 48.2% chance of over 2.5:
- Split 1.30 and 1.30, both teams score (1 - e^-1.3) x (1 - e^-1.3) = 0.727 x 0.727 = 52.9% of the time.
- Split 1.90 and 0.70, both teams score (1 - e^-1.9) x (1 - e^-0.7) = 0.850 x 0.503 = 42.8% of the time.
And because BTTS Yes and over 2.5 tend to land together, a same game multi that combines them is priced well below the two prices multiplied.
Mistakes that cost money on goal totals
| Mistake | What it costs, on the illustrative prices |
|---|---|
| Treating over 2.75 as over 2.5 | On exactly three goals, $80 on over 2.5 at $2.02 returns $161.60, while over 2.75 at $2.32 returns $132.80 |
| Treating over 2.25 as a cheaper over 2.5 | Three or more goals return $140.80 per $80 instead of $161.60, $20.80 less, in exchange for $40 back on exactly two goals |
| Reading the over's implied chance as its true chance | $2.02 implies 49.5%, but with the margin taken out it is 47.4% |
| Counting extra time | A cup tie at 2-0 after 90 minutes has two goals for the bet, so over 2.5 loses whatever extra time adds |
| Keeping a night-before view after team news | In the example, the $2.02 over ends 15.7% below the new fair price |
When a goal total is the wrong bet
- Your view is about one side. If you expect Team A to score freely whatever Team B does, a Team A total or a handicap fits better than the match total.
- Your number does not beat the price. With a fair over of $2.08 and a price of $2.02, each $1 returns 0.4816 x 2.02 = 97.3 cents on average, an expected loss of 2.7 cents per dollar.
- Your view is about pressure rather than goals. Corners and cards have totals of their own, with different settlement rules, covered in corners and cards betting.
Goal lines on the Terminal
The main goal line and its price can both move before kick-off, and bookmakers do not always hang the same line.
On B337's Terminal, soccer total prices sit bookmaker by bookmaker beside head to head and line markets, with the best available price picked out, and the price history (flucs) shows how each one has moved. Prices are read on a repeating cycle, so check both the line and the price with your bookmaker before betting. A free account opens a limited view of the Terminal with live odds.
Risk: Betting involves risk. A total can turn on one goal in stoppage time, a fair price from your own model is an estimate, and there is no guarantee of profit. See responsible gambling for limits and support.