Ben's team ended the season on a high with 93 points in gameweek

As the Premier League football season comes to an exciting end, I reflected on my fantasy football season. My team, Dumb and Dummett, finished in rank 290314 out of 9167407, so in the top 4% of "FPLers". Fairly impressive, although less so considering the amount of teams that end up inactive; the best estimate that I could find was that nearly half of teams are inactive by February time, based on not having made a transfer in five weeks. I did win the "mini-league" between my school friends, though, and bagged a small amount of money in the process, so at least I have something to show for it.

Ben's big

I have played for the best part of a decade now, and I have attempted to figure out what works, and what doesn’t. As a self-confessed and embarrassingly proud football nerd, I forsake sometimes upwards of an hour of work on my degree per week to debilitating deliberation over my team. I check statistics, read match reports and upcoming team news, and examine my rivals’ teams. I have watched many an interview detailing the techniques used by the best of the best, and I will share my insight here. How generous of me.

"Knowing which players to go for - and more importantly, when to go for them - is a deceptively multifaceted task"

There are some classic tactics, that have always been and will always be used. One thing that tends to prove successful is identifying players who play in more attacking roles than their position on the game indicates. With only three (quite restrictive) categories of outfield player (defender, midfielder, forward), wingers classified as midfielders - think Salah, Mané, Jota, Diaz for Liverpool alone - are always popular choices. They get more points than an “out-and-out” striker classified as a forward, despite performing vastly the same role. 

Then there are what are called “differentials” in the FPL jargon. Essentially, they are scarcely selected players that provide that little bit of difference between teams with seven or eight of the same 11 players. This always ends up being the case about ten weeks in; it feels so uncomfortable not to pick those immensely popular players because of the large potential losses on the majority of your competitors when they inevitably do haul. This phenomenon has formed the basis for new metrics like “effective ownership”. Imagine that 70% of FPLers own Vardy and 40% (total) have captained him, doubling his points. His effective ownership becomes 110%, so those 30% of total owners who own a non-captained Vardy actually stand to lose out in terms of the average points gained per team as a result of Vardy’s score. 

Effective Ownership statistics for gameweek 38, for the top 10k and for

In general, knowing which players to go for - and more importantly, when to go for them - is a deceptively multifaceted task. FPLers might consider the difficulty of their team's next few fixtures; the manager’s propensity to rotate his starting XI, especially when the team is playing in European competitions midweek between league games; the player’s susceptibility to injury, and likeliness of substitution.

"This technique of intensive statistical analysis is clearly working"

The special “chips” provide another dimension to an already complex game. Shocks like several injuries at the same time can be mitigated using the wildcard chip, which gives unlimited transfers for a week. Or, “double gameweeks” (two games in the same FPL week) can be exploited using the free hit chip - choose a brand new team just for the week; the triple captain chip - triple rather than double points on the captain; or the bench boost chip - get the points from your four bench players as well as your starting XI. But it is critical to make the most of these, and they can make or break a season.

And success also requires being a realist, rather than an optimist - something that I eternally struggle with myself. As a Spurs fan, I can’t resist owning Tottenham players even when we are on a bad run of form, and I shudder when I force myself to transfer in an Arsenal or West Ham player. 

Jarrod Bowen was a good differential for much of the season, a lot of which saw him selected by less than 10% of

This technique of intensive statistical analysis is clearly working. Oscar, AKA FPL Focal, who spent time earlier in the season at the very top of the nine million plus, has a YouTube channel where he posts several videos for each gameweek. He provides data on predicted points for the next four weeks, the most bought and captained players, his own team, and more. And in 2019/20, Magnus Carlsen spent some time at the top of the world too, finishing 10th overall, and you’d imagine he’s got quite a high-functioning analytical brain (as I’ve heard he’s quite good at chess as well). 


Mountain View

Fitzwilliam win men’s football Cuppers for sixth year running

I’ve thought about how this translates over into Cambridge terms, as I like these Varsity Sport articles to be relevant to the university in some way. In my opinion, these abilities listed most correlate with the skills possessed by an Economics student. So many of them are working towards careers in investment banking, quantitative analysis, or strategy consultancy. Key skills for these include (though aren't limited to) statistical competence, decision making, commitment and persistence, following the right trends and ignoring misleading ones, identifying one’s own hidden gems, predicting the future, and remaining realistic. 

To conclude (this feels like a dissertation - I only hope I put the same amount of effort into my actual one next year), I am going to be testing this hypothesis. I will sort a Cambridge-wide fantasy football mini-league for next season, the details for which I will release when the season returns. And I am very much looking forward to reviewing this piece in a year’s time, with the hope that an Econ student wins the league, and vindicates my hypothesis. Until this time next year!