Braintree Preview: The Forgotten Denominator and Why 60% in Division One Outweighs 92% in Division Three

The Braintree Table Tennis League win-percentage table from last season contains two numbers sitting side by side on the same page: Neil Freeman won 60 per cent of his matches in division one, Dave Fiddeman won 92 per cent in division three.

Read along the axis of who wins more, Fiddeman is clearly superior. Yet when the new-season preview was published, the team placed as the favourite for the division two title was Black Notley B, Freeman's team. Sudbury Strollers, Fiddeman's team, were placed only among the challengers.

That mismatch is not the preview writer's error. It is an error of how we read percentages. A win rate is a fraction. People argue endlessly about the numerator. Almost nobody asks about the denominator.

Context: a local league with clean data

Braintree Table Tennis League is a club-level competition inside the system of Table Tennis England, the national governing body for table tennis in England. The new-season preview was released through that organisation's official media channel, which means its numbers come from internal league records rather than sentiment or dressing-room rumour.

Based on my experience following matches and club-level competitions in England over several years, one structural feature stands out: the smaller the league, the cleaner the data. There is no coach's press conference, no medical bulletin written by a communications department, no advanced metric recalculated three times before publication. There are results, win rates and appearances. Three variables. But those three variables operate under a very tight logic.

The league runs on promotion and relegation. A team that has just been relegated is always treated as a potential winner of the division below, because squad quality does not drop with the division, only the opposition weakens. That is not a gut prediction, it is the mechanical consequence of a promotion system. Black Notley B sit exactly in that case. That is why this season's data trail starts with them.

The preview also notes that tight races can be anticipated at the top of divisions two and three. In a league where line-ups change week by week, tight is not a description. It is a forecast that can be verified by actual appearances.

The evidence chain: three teams, three different denominators

Black Notley B, the hardest denominator

Neil Freeman scored 60 per cent in division one last season. Beside Fiddeman's 92 per cent, it looks modest. But division one is where the denominator is hardest: stronger opponents compress every player's win rate. A 60 per cent record one tier up is not equivalent to 60 per cent one tier down. It is equivalent to a considerably higher rate if re-scaled into division two.

Freeman does not stand alone. Rev Matthews scored 86 per cent in division two last season. That figure was built on the very denominator Freeman will face again, and it sits in the highest bracket of the division. Together they form a high-stability axis, exactly the kind of core a team-format competition needs.

The weakness of Black Notley B lies in a third denominator: Steve Kerns. The league's former men's singles champion will be included for around half of the team's matches. In a three-player line-up, a 50 per cent availability slot is not an added asset, it is a risk variable. Every time Kerns is absent, Black Notley B must field a player who has not been validated at the 60 to 86 per cent level, and the team's average win rate dilutes in exact proportion to the matches he misses.

Sudbury Strollers, a high numerator with a thin denominator

Dave Fiddeman scored 92 per cent, John Colvin scored 75 per cent. This is the finest numerator pair in the entire published dataset. On paper, no division two team has a more consistent attack.

But the preview itself states that Sudbury Strollers' fate may depend on who backs them up and how often. That is a question about the denominator, not about quality. A team with two players at 92 and 75 per cent who only play together ten times will lose to a team with three players at 60 to 86 per cent who play all twenty-two matches. In division two of Braintree, that margin is wider than the technical gap between the two groups of players.

One detail is routinely ignored: a 92 per cent record does not reveal its sample size. Ninety-two per cent across eleven matches and 92 per cent across three matches are entirely different pieces of information. The preview does not publish detailed sample sizes. That is the first data gap I will track once the season begins.

Division three, where old data depreciates fastest

Lucien Nolan-Bradford strolled through division three last season with exactly one defeat. That defeat came against Ben Southgate, finishing 16-14 in the fifth game. One loss all season, decided by two points in the final game. This is the type of data point I always flag in red: too small a margin to conclude anything about nerve, large enough to prove that the gap between the division's two best players is negligible.

Nolan-Bradford has left Finchingfield B. Finchingfield B, runners-up in division three last season, instead received Dave Punt moving down from division two. Structurally this is a favourable swap: a player arriving from a higher tier brings a harder denominator, while Nolan-Bradford moves into a different environment with a softer one. Even so, Finchingfield B are described as having a strong line-up.

The unanswered question lies elsewhere: whether Finchingfield B could be stretched by Black Notley's new F team. A club with the resources to field an additional team is a club with genuine membership depth, and that depth shows up most clearly in mid-season fixtures, when other teams begin to run short of players. The new Black Notley F players impressed on their debuts.

Ben Southgate, the only man to beat Nolan-Bradford, steps up from division three after an 87 per cent season. He is the single most interesting player to watch over the first six rounds: he is moving from a soft denominator to a hard one, and his win rate will be repriced faster than anyone else in the dataset.

The contrarian angle: correlation is not causation

There is a strong temptation when reading this data: rank the players by win rate and conclude which team is strongest. That is a textbook causal error.

A high win rate does not create a title. A high win rate is the output of three other things: technical quality, opponent denominator, and availability. Only the first two are published. The third, which decides most of the outcome in a team-format league, sits outside the spreadsheet.

Put differently, the strongest variable at Braintree this season is not the win rate. It is the attendance schedule.

That leads to a second counter-intuitive consequence: a team with weaker players but full availability can beat a team with stronger players but missing personnel. Steve Kerns is the clearest example. A former men's singles champion playing half a season is not half a former champion in the spreadsheet. He is an empty slot across eleven matches, and that slot gets filled by a lower-tier player.

And I have to state plainly what this dataset admits about itself: it contains no technical information whatsoever. No serve data, no receive data, no description of playing style, no equipment information. Any conclusion about form, nerve or class drawn from a preview like this is inference. The data lets me speak about squad structure. It does not let me speak about technique. Numbers never lie; only the reading of them does.

The blind spot around the junior cohort

The most interesting part of this preview is its junior section, and it is also the least analysed.

Ethan Collins is twelve years old and already holds three cadets' titles and one junior boys' title. At that age, a competitive record that dense means he is not merely coached during lessons, he competes regularly. This is the most important development variable in the entire dataset, and it appears in no percentage column at all.

Rayne D field Sai Suresh, fourteen, and Aryaman Singh, thirteen. Both have been under the watchful eye of league coach Keith Martin. The word used for this debut is baptism, which is data-accurate: a first sustained encounter with adult opposition creates an entirely new denominator for both.

JJ Calisin, eighteen, is scheduled to move up to division one at Christmas. This is a progressive-challenge model: pushing a young player into the hardest tier mid-season, once the denominator is large enough to evaluate. If Calisin holds a positive win rate in division one, that is a stronger signal than any age-group title.

A notable blind spot: none of these players has a projected appearance count published. For a twelve-year-old, playing frequency matters more than win rate. Pushing a child into sixteen matches in four months and pushing him into six are two different development programmes. The dataset does not distinguish between them.

Three signals to track

The Braintree season has just begun, and the only certainty is that the pre-season numbers will be rewritten. The data ocean is no place for those afraid of getting wet.

First, the actual appearance counts of Steve Kerns, Dave Fiddeman and John Colvin. If Black Notley B keep Kerns above seventy per cent of matches, the favourite tag holds. If not, Sudbury Strollers will close the gap with something that appears in no win-rate table.

Second, Ben Southgate's division two win rate after the first month. That is a direct test of the hypothesis that win rates do not convert linearly between divisions.

Third, the real match counts of Ethan Collins, Sai Suresh and Aryaman Singh before winter arrives.

And one final question for the end of the season: if Sudbury Strollers win division two with two players at 92 and 75 per cent while Black Notley B lose with three players at 60 to 86 per cent, will we call it nerve, or will we admit we forgot the denominator?