Last week, I set out to explain what the Mean Reversion Machine project is but, having spent the ensuing days suspecting I could have been clearer, perhaps it would be helpful to begin today by emphasising what it is not. First – and pretty much as a consequence of how it is created – this ‘benchmark portfolio’ of the most-owned ‘assets’ in the FPL fantasy football game is in no way an attempt to win or succeed, do well or outperform.
By the same token – though more gratifyingly for a column that will unashamedly seek every opportunity to work in tenuous investment references – none of what you read is ever a recommendation to buy or sell. There are plenty of proper fantasy football ‘content creators’, both human and AI, you can turn to for tips or, as some poor souls apparently do, full-on cloning. Act on this one and you really do have only yourself to blame.
Where I am less certain is whether the benchmark should be any sort of aspiration – after all, as any worshipper at the altar of the Magnificent Seven will testify, a stock does not become bad simply by virtue of being widely owned. That said, when I discovered my own team’s – deeply-researched – midfield of Salah, Palmer, Wirtz, Kudus and Reijnders was exactly the same as our index’s line-up, I felt deeply, deeply ashamed.
For now, then, let’s just see Mean Reversion Machine as a meditation on the wisdom – or otherwise – of crowds; on the dangers of being too index-aware – or perhaps not index-aware enough; and the question of whether running with the herd risks coming to the sort of knee-jerk decisions one tends to regret bitterly just a week or two later.
Which brings us neatly to our MeanReversionMachine squad, which comprises as many of the game’s most-owned players the £100m budget allows me to cram in – ‘enabled’ by the most-owned cut-price player in each of the four positions, for which I am particularly grateful to Dubravka and, already a hero to this column, ‘Baby’ Reijnders. The index ‘rebalances’ every month – more on which in a moment – but this was the GW02 squad:
“There are plenty of proper fantasy football ‘content creators’, both human and AI, you can turn to for tips. Act on this one and you really do have only yourself to blame.
As you can see, the vagaries of the market – not to mention real life – quickly dented the integrity of our index. Frimpong’s injury and Konsa’s red card in GW01 meant the likes of Virgil, Ait-Nouri and Pedro Porro quickly turned the market’s head. And while our goalkeepers and midfield still perfectly mirror the market, Haaland’s star is on the rise. Even when I do rebalance, we may have to accept owning him and Salah will be tricky.
For now, though, I am holding firm on rebalancing – the aim being to have three free transfers to bring the index more into line with the market at the start of September. As an aside, this enforced discipline seems to be seeping into my own team as, for the first time in 20-odd years of playing, I may actually not make a transfer in the first three weeks of the season – though let’s see if my resolve holds over the next 24 hours …
So, for what it may or may not be worth, how did our team fare in GW02? Bluntly, better than it deserved. Palmer’s warm-up injury saw Salah bumped up to captain, which meant we benefitted doubly from his last-gasp assist against Newcastle. The enforced double-Burnley defence of Dubravka (cheap and playing, thus highly-owned) and Esteve (Frimpong and Konsa not playing, as discussed) also paid off nicely.
Best of all, of course, was the 15 points from Joao Pedro. Most interesting, too, for the second part of Mean Reversion Machine’s dual mandate: keep a close eye on the footballers who are most transferred in and out of squads by Fantasy Premier League’s millions of players. For, as you will see below, the Chelsea striker was the fifth most transferred out player in the preceding week, with 370,000-odd therefore missing out on his points.
To be fair, with a number of the other popular sales being injury and suspension-related, there was no other opportunity for the schadenfreude that I hope will become a regular feature of this column. For their part, the bandwagon-jumpers – and, yes, I am sneering as I type that – had a mixed week, striking gold with Ekitike and, less so, Semenyo; striking out with Reijnders, Wood, Haaland and especially, I would suggest, Ballard.
To aid with the aforementioned schadenfreude and sneering, we will keep an eye on hypothetical missed and gained points for three weeks after all our highlighted transfers, which this week sees continuing interest in Semenyo and Wood, growing optimism for Spurs attackers and not a lot of love at all for Frimpong, Wirtz and Watkins, three founding members of our benchmark:. They may struggle to make it into the September imprint – unless of course we see what I am banking on being another regular feature here – reversion to the mean.
