Since yesterday you can click any team name in our mini league table and see its squad. We used that differently from a normal reader: we pulled the squads of all 60 teams and counted who our league actually owns. Then we set it against ownership across the whole of FPL and against our own AI Rating.
The result is more interesting than we expected. Our league plays noticeably differently from the rest of the world — and in a few places noticeably differently from our own model.
Our league vs the rest of the world
Sixty squads contained 139 different players. These are the ones where the gap between our league and the wider game is biggest:
| Player | Price | Our league | All FPL | Gap | AI Rating |
|---|---|---|---|---|---|
| Mbeumo (MUN) | £8.0 | 71.7% | 38.9% | +32.8 | 88 |
| Calafiori (ARS) | £5.5 | 66.7% | 39.8% | +26.9 | 72 |
| Maguire (MUN) | £5.0 | 45.0% | 18.4% | +26.6 | 68 |
| Tzolis (ARS) | £6.5 | 51.7% | 25.2% | +26.5 | 19 |
| Verbruggen (BHA) | £4.5 | 48.3% | 22.0% | +26.3 | 72 |
| Calvert-Lewin (LEE) | £6.0 | 55.0% | 31.3% | +23.7 | 87 |
| Semenyo (MCI) | £8.5 | 46.7% | 26.3% | +20.4 | 90 |
| Kinský (TOT) | £4.5 | 41.7% | 24.5% | +17.2 | 42 |
One thing stands out: our league is very Manchester United and very Arsenal. The top twelve most-owned players in our league include four from United (Mbeumo, Bruno, Maguire, Shaw) and three from Arsenal (Calafiori, Tzolis, Gabriel). Club loyalty shows up in the data as clearly as it gets — and that is not a criticism, because that is how most people play.
Where the league and the model agree
In four cases our league landed exactly where the rating points:
- Bruno Fernandes — 66.7% here, rated 90, Strong Buy.
- Semenyo — 46.7% here against 26.3% in the wider game, rated 90.
- Szoboszlai — 48.3%, rated 90.
- Calvert-Lewin — 55% here against 31.3% in the game, rated 87 at £6.0m.
The last two are worth underlining: these are cases where our league got ahead of the crowd and the model agrees with it.
Where the league goes against the model
Now the less comfortable part. Three players near the top of our list carry low ratings:
- Tzolis — 51.7% in our league, rated 19. Every second manager here owns a player the model tells you to sell.
- Kinský — 41.7%, rated 42. The Tottenham goalkeeper puzzle we wrote about before the deadline.
- Maguire — 45%, rated 68. A cheap United defender, but the model sees rotation risk.
On Tzolis we have to be honest: here it is probably our managers who are right, not the model. The rating of 19 comes purely from the fact that the engine counts proven Premier League minutes, and Tzolis has not played a single one — not from any lack of quality. He registered six assists in pre-season, more than any other player in the league. Our model cannot see friendlies and cannot reward that; we said so plainly before the season started and we stand by it. How the rating works, and what it does not cover, is set out in our methodology.
Who our league missed
Often the most telling part is what is absent. Four players from our own top ten barely exist in our league:
| Player | Price | AI Rating | Our league | All FPL |
|---|---|---|---|---|
| Harry Wilson (LEE) | £6.5 | 89 | 1.7% | 6.4% |
| Guéhi (MCI) | £6.0 | 88 | 3.3% | 17.1% |
| Gakpo (LIV) | £7.0 | 88 | 3.3% | 3.5% |
| Anderson (MCI) | £6.5 | 89 | 6.7% | 7.9% |
Harry Wilson is the extreme case: one team out of sixty, at a rating level with Haaland and a price of £6.5m. Guéhi looks stranger still — one in six managers owns him across FPL, one in thirty here.
Captaincy: our league split down the middle
Across FPL the opening captaincy is a formality — almost seven in ten managers own Haaland. Here it went differently:
- Bruno Fernandes — 24 armbands (40%)
- Haaland — 23 armbands (38.3%)
- Mbeumo — 4 armbands
One armband in it. That happens to be the direction our model pointed in Friday’s captaincy piece — though in fairness, we have no way of knowing whether anyone decided after reading it, or whether it is simply a coincidence.
The number that surprised us
Sixteen of the sixty teams (27%) played their Bench Boost in the opening gameweek. The rules allow it — that chip is available from GW1. But this is the week when nobody knows the starting elevens, plenty of players are still finding fitness, and benches are often filled with £4.0m squad-fillers.
It is not necessarily a mistake, but statistically it is the hardest possible moment to use it. When the chip is genuinely worth playing is covered in our chips guide, and the best windows for a first wildcard in a separate fixture analysis.
How to read these numbers
The data comes from the game’s official API and covers all 60 teams registered in our league, as of the opening deadline. Three caveats, so there is no misunderstanding:
- Sixty teams is a small sample. This is a photograph of our community, not a cross-section of FPL.
- Our league largely arrived through Polish social media, so club loyalties show up more strongly here than in the global game.
- The AI Ratings are today’s state. The opening gameweek is still running and the numbers will move with every match played.
You can look through every squad yourself — open the mini league page and click any team name in the table. If you are not with us yet, the league code is 3ix1cw; you can join at any point in the season and your full points total goes straight into the table.
This article was produced with the help of artificial intelligence, using data from the official Fantasy Premier League API and our own rating engine. Fantasy PL Edge is not affiliated with the Premier League or the Fantasy Premier League game.