The Ball Grid
Which Premier League seasons players remember best
September 2026
Recent seasons are remembered best, but 2023/24 lags behind an older season and 2019/20 rests on thin evidence.
Every answer on The Ball Grid is a small test of memory. A player is named for a square, and that name is either right or wrong. Group those answers by the Premier League season they relate to and the results show which seasons players still know well and which ones they are starting to forget. Over the last 30 days, 15,230 answers were logged across 5 seasons, and accuracy was far from even.
The table
Here is accuracy by season over the last 30 days, from best remembered to least:
- 2025/26: 65.2% correct, from 7,033 answers
- 2024/25: 61.8% correct, from 3,602 answers
- 2022/23: 60.6% correct, from 1,654 answers
- 2023/24: 53.0% correct, from 2,602 answers
- 2019/20: 52.2% correct, from 339 answers
The gap between the best remembered season, 2025/26, and the worst, 2019/20, is 13.0 percentage points.
The current season is the easiest ground
The season with the highest accuracy is also the most recent. At 65.2%, answers about 2025/26 were correct more often than those for any other season, and this was by far the largest group, with 7,033 answers. That size matters. With so many answers behind it, the 2025/26 figure is the most dependable number in the set, and it is reasonable to treat it as a fair picture of how well players know the current campaign.
This fits a simple idea: the season that is still being played, and still being talked about each week, is the one people have to hand. Squads are fresh in the mind and names do not need to be dug out of memory. The data does not prove why accuracy is highest here, but the pattern is consistent with recency doing much of the work.
2024/25 follows at 61.8% from 3,602 answers. It is the previous season, and it sits just behind the current one. Again, the sample is large enough to take seriously.
The awkward case of 2023/24
If recency explained everything, 2023/24 would sit ahead of 2022/23. It does not. Answers about 2022/23 were correct 60.6% of the time, from 1,654 answers, while 2023/24 managed only 53.0%, from 2,602. The more recent season is remembered noticeably worse, and the 2023/24 sample is the larger of the pair, so this is not simply a small group producing an odd result.
The figures do not explain the dip. It may reflect which players and clubs the grids happened to ask about for that season, or it may reflect something about how that season sits in players' memories. Nothing in the data separates those possibilities, and it would be wrong to present either as the answer. What can be said plainly is that 2023/24 behaves like an older season than it is, and on these figures it sits much closer to 2019/20 than to the seasons either side of it.
2019/20: the least remembered, and the least certain
2019/20 is bottom of the table at 52.2%. On the face of it, that is the expected result for the oldest season in the set. But it rests on just 339 answers, a small fraction of the 15,230 total and far fewer than any other season. A group that size can move noticeably with only a handful of different grids, so the figure should be read as a rough indication rather than a firm measure.
This also means the 13.0 percentage point gap between best and worst should be handled with care. The top end of that range is built on a large sample; the bottom end is not. The gap is real in the data collected, but it is not strong evidence that 2019/20 is uniquely forgotten. Notably, 2023/24, with a much larger sample, is only just above it.
What this means for players
For anyone working through a grid, the practical lessons are modest but useful:
- Current and recent seasons are the safer ground. Squares tied to 2025/26 and 2024/25 are where answers were most often correct, so these are sensible places to start and to build confidence.
- Do not assume a recent season is easy. 2023/24 caught players out more often than 2022/23. A square linked to it deserves a moment's extra thought before committing.
- Older seasons are harder, but the evidence is limited. 2019/20 had the lowest accuracy, though with only 339 answers it is too early to say how much harder it really is.
Put simply, if you are unsure where to spend a guess, the recent seasons have offered better odds over this period.
What the figures cannot show
These numbers cover only the last 30 days and only 5 seasons. They measure how often answers were correct, not how difficult each grid was, and the mix of questions asked for each season may differ. A season can look poorly remembered because its squares were unusually hard, not because players have forgotten it. The seasons between 2019/20 and 2022/23 do not appear in the data at all, so nothing can be said about them.
The clearest conclusion is also the narrowest: across 15,230 answers, the current season was remembered best at 65.2%, the previous one followed at 61.8%, and 2023/24 stands out as the recent season players found hardest. Everything beyond that needs more answers before it can be stated with confidence.
Figures drawn from 28,901 recorded answers on The Ball Grid. Play today's grid.