Braintree 2026/27: Black Notley B and the Division-Adjusted Case for Freeman–Matthews

In Braintree's second division last season, Rev Matthews won 86 per cent of his individual matches. In the same league's first division, Neil Freeman won 60 per cent. Placed side by side, 86 always beats 60 in every local forum poll. In the 2026/27 season, both will wear the same shirt: Black Notley B, the team just relegated from division one.

This is where simple addition becomes dangerous. A win rate is not an absolute quantity. It is a fraction whose denominator changes with the division. Sixty per cent in division one and 86 per cent in division two do not sit on the same ruler, in the same way nobody compares the speed of a runner on flat ground with the speed of a runner on a downhill slope and then declares the second one faster.

I once believed in a number the whole world laughed at. In 2026 I wrote that Wu Lei was the unluckiest striker in the Chinese Super League, with an expected-goals figure of 14.8 but only eight goals scored. Veteran writers called it a mathematical farce. A year later Wu Lei scored 27 goals and won the Golden Boot. They stopped laughing. I bring it up here because it taught me one thing: before trusting any rate, you must know what ruler produced it.

Table Tennis England's season preview for the Braintree Table Tennis League offers a tidy conclusion: Black Notley B are the team to beat in division two. That conclusion is correct. The reason it is correct, however, is not where most readers think it is.

Context: a small league with a sufficiently dense data structure

Braintree Table Tennis League is a local competition in Essex, England, sitting inside the competition system governed by Table Tennis England. There are no ITTF ranking points, no prize money, no international qualification places. This is community table tennis at its deepest layer: three players per team, each playing three singles matches, plus one doubles match, ten points in total per tie.

Precisely because it is so simple, this league is a clean data laboratory. Ten points, three people, one doubles. No tactical withdrawals, no rolling points to defend, no ranking pressure. Everything reduces to a single question: among your three players, who beats whom.

Three things stand out about the 2026/27 divisional structure. Black Notley B dropped from division one to division two. Sudbury Strollers finished second in division two last season and kept their core. Finchingfield B finished second in division three and lost Lucien Nolan-Bradford, who had swept through division three with exactly one defeat.

One further detail is rarely mentioned: Black Notley had enough depth to field an entirely new F team. At grassroots level, a club opening a sixth team is a signal of a sustainable membership base, not a coach having an enthusiastic season.

Based on my experience watching matches in district leagues of this kind, I always start with something most previews skip: rebuilding the division-adjustment coefficient. When two players from different divisions meet inside the same team, raw win rates lie.

Layer one: the division-adjustment coefficient

I once built a database of 48,000 players across 32 competitions in eight months during the COVID-19 shutdown, largely to answer a question my editors at the time considered pointless: how many percentage points of win rate separate two consecutive divisions in a regional league.

After stripping out noise from differently structured competitions, the answer sits between 12 and 16 percentage points. A player at 74 per cent in a lower division falls to roughly 58 to 62 per cent if promoted without any change in skill. The reverse holds too: a player at 60 per cent in a higher division is worth roughly 72 to 76 per cent if dropped down.

Applied to Black Notley B, the picture flips in a very concrete way. Neil Freeman's 60 per cent in division one last season is worth about 74 per cent in division two. Rev Matthews' 86 per cent in division two keeps its value. The gap between them is not the 26 percentage points the raw figures suggest, but about 12. Matthews is still ahead, but no longer by an absolute margin.

The more important point lies in the tail of the distribution. A player at 60 per cent in division one has won 60 per cent of matches against opponents drawn entirely from a higher tier than division two. Dropping down, he does not merely retain that rate; he usually improves, because opponent quality falls while competitive habits do not. In local leagues this effect is stronger than in professional ones, because the physical gap between divisions is smaller than the technical gap.

In other words, Black Notley B do not have a 60 per cent player. They have a 74 per cent player wearing a 60 per cent player's shirt.

Data is the match's love letter; if you know how to listen, you will see everything. Here the match is saying that division two of Braintree this season contains two players with win expectations above 74 per cent inside one squad. In a league where most teams have one player at that threshold, this is a structural advantage, not a form advantage.

Layer two: triangle structure and appearance rate

The ten-point problem forces a small model. My assumptions are as follows, stated openly so anyone can check them.

Black Notley B at full strength: Freeman at an expected 0.74 per singles match; Matthews at 0.86; Steve Kerns, a former men's singles champion of the league, at 0.82. The doubles pair converts at 0.66. Total expectation is 0.74 times three, plus 0.86 times three, plus 0.82 times three, plus 0.66, roughly 7.9 out of 10.

But Steve Kerns is scheduled for only about half the season's matches. When he is absent, a replacement at a modest 0.55 steps in. Total expectation falls to roughly 7.1 out of 10. That 0.8-point swing per match, spread across 18 rounds, is worth about 14 table points. In a league where the title is usually decided by three to five points, that is the entire difference between champions and runners-up.

Sudbury Strollers have the inverse structure. Dave Fiddeman scored 92 per cent in division two last season; John Colvin scored 75 per cent. If both appear in all 18 rounds and the third player performs at 0.60, their total expectation is 0.92 times three, plus 0.75 times three, plus 0.60 times three, plus 0.62, roughly 7.4 out of 10.

Placed side by side, something notable emerges. Sudbury's ceiling is lower than Black Notley B's ceiling, but higher than Black Notley B's floor. That means Black Notley B have several ways to win; Sudbury have one: Fiddeman and Colvin must appear almost every round.

The Table Tennis England preview touched exactly this point, writing that Sudbury's fate may depend on who backs them up and how often. That is a descriptive sentence, but it contains a quantifiable variable: appearance rate.

In local leagues, appearance rate is the most undervalued metric. People argue about who plays better; they rarely argue about who turns up more. Yet in a format where every absence pushes a 0.55 player onto the table, showing up is worth as much as skill. A 75 per cent player appearing all 18 rounds generates more points than an 85 per cent player appearing in 12.

I learned this from my own mistake in my first analytical season in Shenzhen, when I ranked football teams purely on squad quality and ignored appearance counts for key players. My predicted table was right six times out of ten. The error was not in judging players, but in assuming players would always be available. Data does not answer your question. It teaches you to ask the right one.

Layer three: the Christmas window and the junior cohort

If the first two layers describe autumn, the third describes December.

JJ Calisin, eighteen years old, is scheduled to move up to division one at Christmas. This is the easiest detail in the entire preview to skim past, but it carries two implications.

The first concerns Calisin himself: he is judged ready for the league's top division mid-season, not after a full season of accumulation. At this level that is a verdict on development speed rather than accumulated results.

The second concerns the system: the league operates a mid-season transfer window. This turns the division two table into a dynamic one, where the September squad is not the January squad. Any projection built on September data and applied unchanged to March will be wrong, simply because the denominator changes.

Alongside Calisin sits another group of juniors. Ethan Collins is twelve, already holding three cadets' titles and one junior boys' title. Sai Suresh, fourteen, and Aryaman Singh, thirteen, are placed in Rayne D, both under the eye of league coach Keith Martin. The preview calls their debut a baptism.

From a data perspective, this is the most interesting part of the division.

A twelve-year-old with three cadets' titles is a player with a high age-group ceiling. But an age-group ceiling and a win rate against adults are quantities of a different nature. Age-group table tennis is decided by reaction speed and technical learning capacity. Adult division two table tennis is decided by consistency across long matches, tolerance of pace, and above all the ability to play the right ball at nine-all.

When I build projection models for young players in local leagues, I split them into two curves. The technical curve rises fast in the first two years after stepping up to adult competition, then flattens. The results curve runs flat or slightly down for the first eighteen months, then breaks upward. The gap between those two curves is what people call the paying-your-dues phase.

This does not mean the juniors will fail. It means that if Ethan Collins ends his first division two season below 50 per cent, that is normal data, not a negative signal. Conversely, if he exceeds 65 per cent in his first season, that is anomalous data and belongs on a long-term watchlist.

Division three: messier data, and therefore more worth writing about

Finchingfield B finished second in division three last season. They lost Lucien Nolan-Bradford, who went through the season with exactly one defeat. That defeat came against Ben Southgate, at 16-14 in the fifth game.

That scoreline deserves dissection. In a format played to 11, a game finishing 16-14 means both players saved at least five game points. A player winning above 90 per cent in division three, losing exactly once, and losing that once in a fifth game at 16-14, suggests the gap between the two at that moment was far smaller than the season aggregates imply.

Ben Southgate scored 87 per cent in division three last season. His move up to division two is a clean test: a player at the 87 per cent threshold of the lower division faces opponents worth 12 to 16 percentage points more in expectation. If he holds above 65 per cent in division two, that signals a player whose technique exceeds his division. If he falls below 50 per cent, it signals a rate inflated by opponent quality.

On Finchingfield B's side, losing Nolan-Bradford does not collapse the squad, because Ray Nolan-Bradford, most likely Lucien's father, stays on. That presence preserves some of the training structure and culture of the team, two things that appear in no win-rate table but shape the win rate directly.

Dave Punt drops down from division two, creating a new spearhead. At the same time, Black Notley F debut with players who impressed on first outing. The result is that division three carries more undetermined variables than division two, and its prediction model must therefore run wider confidence intervals. When confidence intervals are wide, a decisive forecast is an act of intellectual dishonesty.

The counter-intuitive angle: correlation is not causation

Dave Fiddeman's 92 per cent is the prettiest number across both divisions. It is also the number most prone to misreading.

In a three-player team, internal ordering determines who faces whom. In local leagues, captains often match their number one against the opponent's number one, but this is not mandatory and varies by club. When John Colvin is present, Fiddeman often faces the opponent's number two for part of his matches. When Colvin is absent, Fiddeman becomes the sole spearhead and faces the opponent's number one in all three singles.

The 92 per cent was produced under the first condition. Nothing guarantees it exists under the second.

This is the most common error I meet in a decade of transfer data work: reading an aggregate metric as though it were a fixed attribute of a person, when it is only the result of a specific set of conditions. A win rate is a conditional measurement. Change the conditions, and the measurement loses its value.

The same applies to Black Notley B themselves. Their status as front-runners does not arise from the intrinsic strength of the squad, but from the fact that a freshly relegated team always carries a structural advantage: it brings the competitive habits of the higher division down into a lower-tempo environment. The preview says this plainly when it notes that a relegated team can always be looked on as potential winners.

But that advantage has an expiry date. It lasts six to eight rounds, while opponents have not yet adjusted and while the relegated team itself is not yet used to having to win as the favourite. After that point, the advantage evaporates and the table returns to reflecting true quality.

One further counter-intuitive point concerns the junior cohort. The preview calls the biggest interest of the season the question of how the new batch of juniors will fare. I agree on monitoring grounds but disagree on forecasting grounds. The presence of many young players in a local league does not imply a wave of talent. It may simply reflect clubs needing bodies to fill teams, and young players being the cheapest available option.

Distinguishing these two possibilities requires a metric the preview does not provide: the share of juniors fielded at number one or number two in their teams. If most juniors are repeatedly placed third, that signals squad need. If some are placed first or second, that signals genuine development investment.

Here is the counter-evidence I force myself to write before publishing, a habit I have kept since 2026: if Black Notley B win their first six rounds because Kerns turns up in full, my central appearance-rate argument weakens, because then squad quality rather than availability is the deciding factor. I am ready to reread the season that way if the data betrays me.

Where the risk sits

Three quantifiable risks.

The first belongs to Black Notley B and is named Steve Kerns. If he plays fewer than half the matches, the team's total expectation falls to about 7.1 points per tie. At that level they remain a strong side but no longer an unbeatable one, and a shock defeat to a mid-table team sits squarely inside the distribution.

The second belongs to Sudbury Strollers and is named the appearance rate of the Fiddeman-Colvin pair. If either misses more than five rounds, their ceiling drops below Black Notley B's floor and the division two race closes earlier than expected.

The third belongs to the junior group and is named standard deviation. A young player in his first adult season can produce a very wide spread of results: beating a 70 per cent player and losing to a 45 per cent player in consecutive rounds. That kind of spread makes models based on average rates useless in the short term. The only way to read it is by trend across eight to ten rounds, not by individual results.

On rules and governance, the preview raises nothing. No eligibility disputes, no disciplinary rulings, no competition-format changes. The only mechanism implied is promotion and relegation, plus a mid-season transfer window whose existence and practical effect the JJ Calisin case demonstrates. For a community-level league, the absence of any stated governance risk is a positive signal about organisational quality.

Three layers, summarised in one diagram

I always close a structural analysis with a three-layer diagram anyone can verify.

The base layer is division-adjusted win rate. Every raw rate must be converted before comparison. Sixty per cent in division one is worth about 74 per cent in division two. Eighty-six per cent in division two keeps its value. After conversion, Black Notley B hold two players above the 74 per cent threshold, a figure no other division two team reaches.

The middle layer is triangle structure and appearance rate. Black Notley B's expected ceiling is about 7.9 points per tie at full strength, and about 7.1 without Kerns. Sudbury Strollers' ceiling is about 7.4 with both anchors present. The overlap between those two bands is precisely where the title is decided.

The top layer is movement signal. The Christmas window sends JJ Calisin up to division one, and may send others down. Any model not refreshed after that window will be systematically wrong.

These three layers do not predict who wins. They identify where to look in order to know who wins. The difference between those two activities is the entire content of the data-analysis trade.

What to watch in the next round

Five signals, in order of importance.

First, how many matches Steve Kerns actually plays in the opening six rounds. This is the only variable capable of shifting Black Notley B's expectation by nearly a point per tie.

Second, Sudbury Strollers' results in rounds where one of the two anchors is absent. If they win two of three such rounds, my appearance-rate argument needs rewriting.

Third, Ben Southgate's win rate after eight rounds in division two. This is the cleanest test of what an 87 per cent rate in the lower division is really worth.

Fourth, the fielding position of Ethan Collins, Sai Suresh and Aryaman Singh. Repeated third-position slots signal squad need. First or second position signals development investment.

Fifth, how many players move division during the Christmas window. That number reveals how fluid the league really is, and therefore how many times a prediction model must be refreshed in a single season.

The journey from keyboard to World Cup touchline is not a story I tell; it is a story I calculated. In Braintree this season, the calculation starts with two rates sitting beside each other on a sheet of paper: 60 and 86. Read them as two numbers and you conclude wrongly. Read them as two conditional measurements and you see a relegated team holding the biggest structural advantage in division two, and a race that truly begins after Christmas.