How I try to predict the future in FPL

It’s a gift that every Fantasy Premier League (FPL) manager wishes they had: the ability to see into the future and know what is going to happen in the upcoming Gameweek.

Short of 1) being in possession of a working crystal ball, 2) getting our hands on a copy of Biff’s Back To The Future Grays Sports Almanac book or 3) having Nicholas Cage-esque powers from the 2007 film Next, we have to look for the next best thing. 

When Tom from the excellent and long-running FPL podcast Who Got The Assist tweeted that he was going to captain Newcastle’s Chris Wood in Double Gameweek 33 last season, I feared the worst and suspected that his Dostoevsky-like trials and tribulations with FPL would continue.

And sure enough, BigManBakar’s favourite player of all time duly went on to score 2 x 2 points in the double. Mohamed Salah, in contrast, scored 19 points in the same Gameweek with a triple return (a brace and an assist) in a single fixture, which was a 4-0 perch-knocking home win for Liverpool against Manchester United. But was there any way to see this coming? Could we have predicted what was going to happen in Gameweek 33 to a certain extent?

Well, I remembered listening to a fireside chat episode of the FML FPL Podcast with Michael Caley, the inventor of expected goals (xG) maps. Michael had been talking about Harry Kane and saying how important “shot volume” was to him and to any other attacking player. If a player is getting lots of shots off in a game, then regardless of whether they are blocked or off-target, the fact they are shooting all the time is very positive and half the battle.

With this in mind, from Gameweeks 26 to 32, Wood had a minutes-per-shot average of 77.7. This means he was likely to have roughly one shot in a 90-minute football match. It is asking a lot for him to score with that one effort. Put simply, he’s not getting the volume of goal attempts to do well. 

Compare this with Salah, who was having a shot every 15.3 minutes between Gameweeks 26 to 32 – around six attempts per 90-minute game. The odds are in his favour: he is much more likely to score with six shots compared to Wood’s one effort. Of course, events could conspire to get Wood a penalty and a nine-pointer with his one attempt, but then you are playing the game in hope rather than in expectation. 


It was something of a Eureka moment for me halfway through last season when I realised I should be paying a lot more attention to shot volume. I duly created a custom stats table in the FFScout Members Area that was set to the last six Gameweeks and looked at things like minutes per shot, minutes per xGI and minutes per chance created.

The concept was simple: if a player isn’t going into a given Gameweek taking plenty of shots and/or creating at least a good three or four chances per game, then goals and assists and therefore points are unlikely to be forthcoming. Accordingly, the player should not be captained or transferred in.

Similarly, if a player I own has been blanking for a number of Gameweeks, but he is still taking three or four shots per game, then I should think twice before selling. I was seeing FPL pundits and content creators talking up players who in the last six Gameweeks had triple-figure minutes-per-shot averages (i.e. they were taking a shot on average every 100 minutes or more and couldn’t even be relied on to shoot once – let alone score – in a single match!) and I felt that I had a secret weapon to play FPL with. By paying close attention to the crucial stat of minutes per shot, I could ignore players who just weren’t getting the shot volume and predict with some degree of certainty what was going to happen when Saturday comes.