# Tennis set-score betting: how to assess a price

> Understand tennis set-score betting with formulas, a p = 0.6 example, margin calculations, break-even prices and rules for player retirement.

- Source: https://toolsgambling.com/blog/tennis-set-betting
- Publisher: ToolsGambling
- Published: 2026-10-26
- Updated: 2026-10-26
- Cite as: "Tennis set-score betting: probabilities and reference prices", ToolsGambling, https://toolsgambling.com/blog/tennis-set-betting

A tennis set-score bet pays when the final set score of a match matches the score you picked, such as 2-0 or 2-1 in a best-of-three match. At p = 0.6, 36 out of 100 matches end 2-0 for the favourite under the model. Here, p is the favourite’s chance of winning each set: P(favourite 2-0) = p². This is a calculation, not a forecast.

## What is tennis set-score betting?

It is a correct score bet on the final score in sets, rather than simply on the match winner. A favourite winning 2-0 and the same player winning 2-1 are separate outcomes.

A best-of-three match has four possible final set scores; a best-of-five match has six. A Russian guide at bookmaker-ratings.ru gives the same counts. The calculations below use a simplified model, and the example prices are illustrative rather than operator odds. Competitor price examples and claims of profit are not evidence that a score is worth backing.

## What is the calculation for?

The calculation lets you compare a price with the probability of a particular score—not just with a player’s chance of winning the match.

A reference price is the inverse of the model probability, without margin. See [how betting odds work](https://toolsgambling.com/blog/how-betting-odds-work) for that relationship. Assessing players and match conditions is a separate task, covered in [tennis betting before the match and live](https://toolsgambling.com/blog/tennis-betting-strategy).

## How does the model work?

The model assumes independent sets and a constant set win probability throughout the match. Write the favourite’s probability as p and the underdog’s as 1−p.

Multiply the probabilities of individual set results to find the probability of a sequence. Where different sequences lead to the same final score, add their probabilities.

The value that most affects the reference price is p. At p = 0.60, the favourite’s 2-0 probability is 0.36 and its reference price is 2.78. Changing p to 0.55 raises that price to 3.31; changing it to 0.65 lowers it to 2.37.

## How to calculate: example at p = 0.6

For a best-of-three match, calculate the four score probabilities using these formulas:

- P(favourite 2-0) = p²
- P(favourite 2-1) = 2p²(1−p)
- P(underdog 2-0) = (1−p)²
- P(underdog 2-1) = 2(1−p)²p

The factor in each 2-1 formula accounts for the possible orders of the sets before the deciding set. The four probabilities sum to 1.

![Diagram: four best-of-three set scores at a 0.6 set probability](https://toolsgambling.com/media/article/2026/10/399c097a240fdb45654f58d10e03eb05.webp)

*Teaching diagram: score probabilities from the model, not operator odds*

Substituting p = 0.6 under the model assumptions gives:

| Final set score | Probability | Reference price |
|---|---:|---:|
| Favourite 2-0 | 0.36 | 2.78 |
| Favourite 2-1 | 0.288 | 3.47 |
| Underdog 2-0 | 0.16 | 6.25 |
| Underdog 2-1 | 0.192 | 5.21 |
| Total | 1.0 | — |

Reference prices are rounded to two decimal places, so converting them back gives approximate probabilities: a price of 2.78 implies approximately 0.36. To check this with the [implied probability calculator](https://toolsgambling.com/betting/implied-probability), enter 2.78 in “Enter your odds” and read “Result”. This checks the price conversion; it does not validate the assumed p.

## What changes in a best-of-five match?

A best-of-five match requires probabilities for 3-0, 3-1 and 3-2 for each player. The best-of-three score probabilities cannot be carried over.

At p = 0.6, the favourite’s probabilities are:

| Final set score | Probability |
|---|---:|
| Favourite 3-0 | 0.216 |
| Favourite 3-1 | 0.2592 |
| Favourite 3-2 | 0.20736 |

Their sum is 0.68256, or 0.683 rounded. That is the favourite’s probability of winning the match, not of achieving one correct score. In the best-of-three model, its match win probability at the same p is 0.648.

## How sensitive is the result to p?

A shift of 0.05 in p changes the favourite’s 2-0 reference price by about 0.41–0.53.

| Set win probability p | Probability of favourite 2-0 | Reference price |
|---|---:|---:|
| 0.55 | 0.3025 | 3.31 |
| 0.60 | 0.36 | 2.78 |
| 0.65 | 0.4225 | 2.37 |

A well-supported estimate of p matters more than extra decimal places in the output.

## Margin and a conditional conclusion from the price

Under these assumptions, a price of 2.60 on the favourite winning 2-0 reaches break-even only at p ≥ 0.620. The corresponding score probability is 0.3846; both thresholds are rounded. At p = 0.60, the price does not break even: expected net profit is −0.064 per unit staked.

Consider this complete illustrative market:

| Final set score | Illustrative price |
|---|---:|
| Favourite 2-0 | 2.60 |
| Favourite 2-1 | 3.50 |
| Underdog 2-0 | 6.00 |
| Underdog 2-1 | 5.00 |

The sum of 1/odds across these four scores is 1.037, giving a margin of 3.7%. The [margin calculator](https://toolsgambling.com/betting/margin-calculator) helps check that sum.

Expected net profit per unit staked is:

**Outcome probability × price − 1**

A negative result means an expected loss, not a negative payout.

- At p = 0.65, the same 2.60 price on favourite 2-0 gives expected net profit of +0.0985 per unit (about +0.10).
- For a price of 3.50 on favourite 2-1, expected net profit is positive only for p from about 0.591 to 0.737: the probability of 2-1 peaks at 0.296 when p = 2/3, and a stronger favourite wins more often 2-0. At p = 0.60 it is +0.008 per unit—a very narrow modelled edge.

These are conditional comparisons, not advice to bet or a guarantee of profit.

## What are the model’s limits?

The model does not estimate p, account for changes in that probability during a match, or adjust for dependence between set results.

Pinnacle’s “Tennis set betting odds and statistics: Part one”, published 30 September 2018, states:

> “We can expect some correlation within matches for set one and set two results.”

The size of that correlation is not estimated here, so no numerical adjustment has been applied to the tables.

## What if a player retires?

Settlement depends on the operator’s rules; a retirement does not automatically establish whether a correct score bet wins or becomes void.

Betbazar (B1)’s “Tennis Settlement Rules”, Correct Score section, updated 19 August 2026, states:

> “If one of the players retires or is disqualified, all unsettled bets on markets whose outcomes were not determined prior to the interruption will be void.”

The same section states:

> “If the match format is changed (e.g. a best-of-5 sets match is played instead of a best-of-3, or vice versa), bets will be void.”

These are Betbazar’s rules, not universal settlement rules. Check your operator’s terms for correct score bets and interrupted matches.

## Which tools fit this task?

Implied probability and margin tools can check the arithmetic behind prices, but they do not build a tennis model or estimate a player’s chance of winning a set.

The ToolsGambling correct score calculator uses a Poisson distribution for football; it should not be used to calculate tennis set scores. See [the football correct score guide](https://toolsgambling.com/blog/correct-score-odds-explained) for that different modelling approach.

## FAQ

### Where does the probability of winning a set come from?

In this example, p is an assumption. The formulas convert it into score probabilities; they do not estimate it from player statistics. Without a justified estimate of p, the result remains conditional.

### Do I need to account for the order of sets separately for a 2-1 score?

No. The model already combines both possible sequences: the favourite wins the first and third sets, or the second and third sets. Both produce the same final correct score.

### Can I use the probability of winning the match directly as p?

No. Here, p means the probability of winning an individual set, not the match. Substituting a match win probability changes the meaning of the calculation.