Braintree Table Tennis League: The 16-14 Fifth Game and Two Win Rates Shaping the Promotion Race

In division three of the Braintree Table Tennis League last season, Lucien Nolan-Bradford went through the entire campaign with exactly one defeat. The man who beat him was Ben Southgate, and the match ended 16-14 in the fifth game. Two points at the very end split a near-perfect season in half. In the data sheet I keep for this season, that is the most valuable fact available, because it shows how narrow the margin of error is in a small division. When the gap between an unbeaten player and the man who beat him is two points in a single game, every attractive win percentage deserves a warning label.

The first data sheet I built by hand contained hundreds of errors, but it taught me more about clean data than any course I have taken. Since then, whenever I read a league preview, I separate the numbers from the predictions before I believe anything. Table Tennis England's preview of Braintree's second and third divisions is a good exercise in that discipline.

Context: a local league with unusually clean data

The Braintree Table Tennis League is a club-level competition in England, run on a promotion-and-relegation model. There are no ITTF ranking points, no prize money, no international qualification places. Its value sits elsewhere: every match is recorded, and each player's win percentage is published by division. For anyone working with data, this is a rare clean sample.

In division two this season, Black Notley B are the highest-rated side. Their line-up includes Neil Freeman, who scored 60 per cent in division one last season, Rev Matthews with 86 per cent in division two, and Steve Kerns, a former men's singles champion, who will appear in roughly half of their matches. Sudbury Strollers, second last year, are the most frequently mentioned challengers, with Dave Fiddeman at 92 per cent and John Colvin at 75 per cent.

In division three, Finchingfield B, runners-up last season, lose Nolan-Bradford but gain Dave Punt, who is moving down from division two. Rayne D enter the season with two players aged 14 and 13. Black Notley F are a new team, and their players were reported to have impressed on debut.

What the data actually says

Win percentages do not travel with a player when that player changes division. This is the point I want to stress most, because it is the source of almost every error in league previews. Dave Fiddeman's 92 per cent was recorded in division three. When he plays in division two, the denominator changes: stronger opponents, more extended games, and a higher overall baseline across the division. A player holding the same form can lose 10 to 15 percentage points simply by moving up. So when I read that Sudbury Strollers have two players at 92 and 75 per cent, I record attacking capability, not championship capability.

By contrast, Neil Freeman's 60 per cent in division one is worth far more than it appears. A player who scores 60 per cent in the hardest of the three divisions, on dropping to division two, typically holds a level around 75 to 80 per cent. If Freeman sustains that, Black Notley B have a stable anchor in the number one slot. But the second anchor depends on a different variable.

Steve Kerns appears in roughly half the matches. That is the single most important variable in division two, and it does not appear in any win-percentage table. A team with three good players but only two regular attendees has a far lower floor than its ceiling. In local leagues, availability decides final standings more often than the quality of the best player does. I have re-checked several seasons at club level and the pattern repeats fairly consistently: a team with three reliable players beats a team with two excellent players on rotation.

Sudbury Strollers sit on the opposite side. The preview states plainly that their fate depends on who backs up Fiddeman and Colvin, and how often they play. That conditional is written as a remark, but it functions as a forecast: if Sudbury cannot find a steady third player, they will drop points in the middle rounds, exactly the phase when thin squads tend to collapse.

Division three is messier for two reasons. Finchingfield B lose Nolan-Bradford, who barely lost a match last season. In his place they have Dave Punt dropping from division two, a player whose opponent baseline is higher than the division-three average. Alongside that, Black Notley F arrive with new players, and a new team in division three is an unpriced variable. There is no historical data, no reference win rate, only debut matches.

The counter-intuitive angle

Data does not need my belief. Data needs my verification. The 2026 World Cup taught me one thing: the model did not collapse, I was the one who believed it absolutely. A few years ago I built a regression model over 500 international matches and it returned a 78 per cent probability that a certain team would reach the semi-finals. That team went out in the group stage. The lesson sits in the fact that a model only answers the question it was built to answer.

When the Bundesliga played behind closed doors, I realised home advantage is just a variable waiting to be deleted. The same applies to the claim that Black Notley B are the team to beat in division two. That phrasing sounds like a conclusion, but it is only a comparison of win rates at a single point in time. It describes the players' past, not the team's upcoming fixture list. In a league with limited availability and mid-season division moves, every forecast has an expiry date.

That expiry date is visible. JJ Calisin, 18, is scheduled to move up to division one at Christmas, after progress described as impressive. The detail matters: it shows the league operates on mid-season movement windows, meaning a September line-up is not a January line-up. Anyone who builds a prediction table in early season and freezes it is analysing a league that does not exist.

There is one more cluster of variables: the juniors. Ethan Collins, aged 12, already holds three cadet titles and one junior boys' title. Sai Suresh, 14, and Aryaman Singh, 13, are placed in Rayne D, under the watchful eye of league coach Keith Martin. For players at this age, the second season's win rate matters far more than the first, because opponents now have data on them. A 12-year-old winning titles in his own age band says nothing yet about handling pressure against a 40-year-old who plays safe and reads serve well.

What to watch

The first signal is Steve Kerns' actual appearance count. If he plays 12 matches or more, Black Notley B's floor stabilises and Freeman's win rate becomes the decisive factor. If the number is lower, division two becomes an open race.

The second signal is Ben Southgate. He cleared division three at 87 per cent, and his step up will test whether that record reflected real level or merely the standard of his opponents. I read a team through thirty variables before I listen to a commentator, and with Southgate the variable to track is his win rate over the first six rounds, not across the whole season.

The third signal is the junior group at Rayne D and Black Notley. Their first-month results will show whether the league is developing successors or simply filling gaps.

The 86 per cent and 92 per cent figures are beautiful numbers. They are also records of a finished season, in different divisions. Their real value lies in forcing us to ask again after every round: what is the denominator behind these rates, and has it changed yet?