Table tennis: sets won %, comebacks and points per match
TL;DR. In table tennis the useful statistics are not matches won but sets won % (real level, with far more sample than the final score), comeback % (the ability to survive a bad start) and average points per match (the number that governs total lines).
Note. Informational and statistical content. It is not betting advice nor a promise of profit. No statistic predicts a single match: they are there to estimate probabilities over the long run. 18+ only. Play responsibly.
Why sets won beats matches won
A player can win 50% of their matches and 55% of their sets, or the other way round. The gap says a lot: whoever wins more sets than their match record suggests tends to lose the tight ones, and whoever wins fewer sets than expected tends to close finishes well.
The practical advantage is sample size. A best-of-five match leaves one result but three to five sets. In a sport with dense calendars and continuous table events, that multiplies the available data by four for the same number of matches.
The metrics we publish
| Metric | What it measures |
|---|---|
| Sets won % | Sustained level with the largest sample available |
| Comeback % | How often they win after falling behind on sets |
| 1st set win % | How well they start, before reading the opponent |
| 1st set to match % | How much that start is actually worth |
| Points for and against | Scoring rate; feeds the total points line |
The first set decides less than it seems
The pair «1st set win %» and «1st set converted into match %» is the most interesting read in the block, because it puts an intuition against a number. Winning the opening set is a real edge, but in table tennis conversion is far from automatic: best-of-five or best-of-seven leaves plenty of room to correct, and tactical adjustments between sets (service, tempo, side of the table) happen very fast.
A player with a high first-set win rate but low conversion is exactly the profile that slips past anyone watching only the live scoreboard.
What these statistics do NOT tell you
- They are not opponent-adjusted. 60% of sets against a weak draw is not the same as 55% against hard opposition.
- Comebacks have the smallest sample of the block. They require falling behind first, so they are computed over a fraction of matches: two or three comebacks do not describe a character.
- Average points depend on the format. A best-of-seven event produces larger totals than a best-of-five; comparing across formats without correcting is a mistake.
- We do not model this sport. In table tennis we publish statistics, not forecasts.
Where to see them
On the table tennis statistics page, with average points as the highlighted column and range filters for sets won % and comeback %. If you want the concept behind total lines, see the Over/Under glossary entry and the definition of a line.
Frequently asked questions
Why look at sets won rather than matches won?
Because it carries three to five times more sample for the same number of matches, and because it separates sustained level from the ability to close tight finishes. The gap between the two percentages is itself informative.
Does winning the first set decide a table tennis match?
It is a real edge, but conversion is far from automatic: best-of-five or best-of-seven leaves room to correct, and tactical adjustments between sets are very quick. That is why we publish the conversion rate separately from the first-set win rate.
Can total points be compared across different events?
Not directly. A best-of-seven format produces larger totals than a best-of-five. You have to compare within the same format or correct for the number of sets.