How a Hastings junior chess club took puzzle homework from 34% to 81%

Hastings Rooks Junior Chess Club runs on Friday evenings in a community centre hall, 58 players aged 7 to 16 split across three strength groups, two coaches and six parent helpers. Between one Friday and the next, Toby Renshaw had no way of knowing which of those 58 juniors had touched a chessboard. In September 2024 he counted: 20 players out of 58 handed back the week's puzzle sheet.

Where the club started

Toby is a secondary school maths teacher who has coached at the Rooks for nine years, the last three as head coach. The club charges £4 a session, runs 34 Fridays a year, and sends players to county junior events and the occasional weekend congress.

Chess is unusually demanding as a club sport because almost all the improvement happens between sessions. Two hours of Friday club play is the smallest part of a junior's week. Tactics puzzles, opening review and game analysis are the part that moves a rating on the English Chess Federation's four-digit scale, and none of it happens in the hall.

The club's tracking was normal for a volunteer-run junior section, which is to say it was mostly in one person's head.

• Puzzle sheets were photocopied on Thursday and handed out at the door on the way home.
• Ratings lived in a spreadsheet Toby updated after congresses, roughly four times a year.
• Attendance was a tick list on a clipboard when a parent helper remembered the clipboard.
• Event entries were announced in the parent WhatsApp group and chased individually in the final 48 hours.
• What a player was actually working on lived in Toby's memory, and split badly between him and the second coach.

The September 2024 count came out of a complaint rather than a plan. A parent asked why her son had not improved in a year. Toby could not tell her whether the boy had done any work outside the hall, which he found a harder thing to admit than the lack of improvement.

What the club found underneath the completion rate

Three things mattered more than the 34%.

First, the same sheet went to everyone. A player rated around 700 and a player rated around 1,500 got identical puzzles, which meant the sheet was too hard for a third of the club and pointless for the strongest six. Both groups stopped doing it for opposite reasons.

Second, nobody was accountable for the middle. The top group did the work because their parents pushed. The bottom group was visible because they struggled in the hall. The 24 players in the middle group were invisible in both directions and drifted out at a rate the club had never measured.

Third, congress entries were failing on admin, not appetite. Across the 2024/25 season nine juniors missed an entry deadline for an event they had said they wanted to play. For a 12-year-old, missing a county congress means waiting five months, and two of the nine did not come back after Easter.

"The bit that stung was realising I had 24 players I could not describe," Toby says. "I could tell you everything about the six who win things and the eight who need help, and for the middle group I had a name and a rough rating and nothing else. They are most of the club."

"I had 24 players I could not describe. They are most of the club."

What the Rooks set up in Bitir

Toby moved the middle group across in January 2025 and had all three groups running by the summer. The structure follows the approach in our guide to homework tracking for teachers.

1. One private group per strength squad, plus a parent group. Parents get logistics, hall changes and event announcements. Coaching content stays in the squad groups.
2. Each week's puzzles are a per-player assignment. Three difficulty bands, set on Friday night with a due date of the following Thursday, marked by the player and checked by a coach.
3. Every junior has one written goal. In their own words where possible. "Stop hanging pieces on move 12" is a better goal than "improve tactics", and 41 of the 58 wrote something specific enough to work with.
4. Ratings are a private per-member record. Visible to the player, their parent and the two coaches. Not posted, not ranked, not on the wall.
5. Congress entries are a poll, posted the week the event opens. Intent is collected six to eight weeks out instead of two days out.
6. Attendance every Friday, taken by whichever parent helper is on the door. Under a minute, and it finally answers who has stopped coming before the club loses them.

The private-rating decision was deliberate and Toby argued about it with two other coaches. He had run a wall ladder in 2023 and pulled it down in eight weeks, because the four juniors at the bottom stopped playing club games and started sitting out with a phone. The reasoning matches our guide to progress tracking for groups: private progress, public celebration at the moment of a real result.

What changed in the numbers

Toby compared the spring term of 2026 against the autumn term of 2024, using the same completion measure he ran on the day of the original count.

Toby is careful about the rating question, and it is worth repeating how he puts it. Across the 31 players rated at both ends of the 14-month window, the club's mean rating rose by 118 points. He will not claim an app did that. What he will claim is that the players who were doing no work between sessions are now doing some, and that a coach who can see who did what sets better puzzles the following week.

The entry-deadline number is the one he expected least and values most. Nine missed deadlines in a season is not a scheduling problem, it is 27 lost weekends of chess spread across nine families, and it was fixed by asking the question two months earlier in a format that produced a list rather than 40 messages.

Attendance mattered for a duller reason. The club now knows that four juniors stopped coming in the six weeks after the autumn half term, which it had previously discovered in January. Two were persuaded back. The same argument runs through our guide to attendance tracking for classes and clubs.

What Toby does in Bitir

• Assignments for the weekly puzzle set, in three difficulty bands, due the Thursday before club night.
• Goals written by the player in their own words, one per junior, reviewed each half term.
• Per-member records for ratings and coaching notes, private to the player, their parent and the two coaches.
• Polls for congress and county event entries, posted the week an event opens.
• Attendance every Friday, taken at the door by a parent helper with limited permissions.
• Parent group for hall changes, transport to away events and results announcements.

What he would tell another junior club

• Count your completion rate before you change anything. It takes one Friday and it is the number that ends the argument.
• Band the homework. One sheet for a 700 and a 1,500 is the reason half your club ignores it.
• Ask about events the week they open, not the week they close.
• Take the ladder off the wall. You will lose the players at the bottom of it, and they are the ones who need the club most.
• Watch the middle group. The strong and the struggling both get attention by default.

"We have not added a coaching hour. We have stopped guessing about 24 of our players."

What's next

The Rooks are trialling per-player game analysis, where a junior uploads a game they lost and the assignment is one written sentence about the move where it turned. Toby also wants the two sixth-formers who help on Fridays to run the bottom group's puzzle marking themselves, which needs a role that can mark work without seeing coaching notes or family contact details.

Related reading: how to track homework without chasing it, and how to run a volunteer rota without the group chat doing the work.