Double or triple-up? How much trust should FPL managers put in one club?

It could be argued that the question of whether it is better to centralise or decentralise control or resources has caused more conflict than any other throughout history as it does, after all, underpin almost all political debate.

At the core of these discussions exists the balancing of some very disparate beliefs, priorities and ideologies such as flexibility versus efficiency and individualism versus collectivism. Since, as we already know, Fantasy Premier League (FPL) is a powerful metaphor for life in general (or perhaps vice-versa), it is not, itself, absent of these tensions.

Within an FPL context, the clearest manifestation of this question is whether it is better to focus your squad selection on players from just a few teams or to spread that selection across many teams. This is the question that this article will seek to address.

The Case for Centralisation

Centralisation in an FPL context begins to occur when more than one player is selected from an individual Premier League team. Given that three players per team is the maximum in this respect, the highest level of centralisation would involve 15 players being selected from just 5 teams.

What are the advantages of this method?

First, by limiting your player selection to fewer teams, it reduces the number of teams you have to research. This permits you more time for analysis and, thus, an opportunity to gain a deeper understanding of individual teams. Typically in FPL, the more you understand a team, their line-ups, patterns of play, set-piece takers, general strengths and weaknesses, the better-informed your decisions relating to their players can be.

It also means you can assess their fixtures and overall team form more easily. As with engine mechanics, the fewer the moving parts in FPL planning, the greater the overall efficiency.

There is also a ‘multiplier effect’ associated with picking several players from an individual Premier League team. For example, if you owned three Manchester City defenders, one clean sheet could be worth at least 12 points whereas, if your defenders came from three different teams, you would need each of these teams to keep a clean sheet in order to score the same amount, an outcome that is objectively less likely to occur.

Equally, if an individual team is scoring a lot of goals, it may well pay to maximise your coverage of their attacking assets. Not only will this give you access to a potentially strong supply of attacking points, it will increase the likelihood that you own at least one of the players generating them.

A key problem with the centralised strategy, however, is that it reduces flexibility. If, for example, you had three Man City defenders and wanted to bring Raheem Sterling (£11.6m) in, achieving this would involve at least two transfers.

The spend balance throughout your team will also be dictated by the relative pricing of individual teams. Three Man City defenders would involve more money spent in defence than, say, three Burnley defenders. This will have a knock-on effect on spend elsewhere and potentially reduce options for alternatives should you want to swap one out.

The Case for Decentralisation

Decentralisation, in an FPL context, would mean spreading out your player selection across a wider range of Premier League teams, the maximum, of course, being 15.

One key aspect of the decentralised approach is that it is less likely to produce extreme outcomes. Similar to how, in probability equations, the introduction of more variables typically reduces the odds of a particular outcome, utilising players from a wider range of teams is both less likely to produce a huge score and less likely to produce an absolute stinker.

You can observe this effect in practice by checking out the highest points scorer on the Official FPL site at the end of each Gameweek. These teams will almost always lean more towards a centralised rather than decentralised composition, with more players selected from a fewer number of Premier League teams.

Decentralisation reduces the risk of a blow out because it spreads that risk across a wider range of teams. Using again the example of the three Man City defenders from above, one goal conceded would wipe out all three cleans sheets. With defenders from three different teams, three goals would need to be conceded for the same outcome to occur. From a purely quantitative perspective, three events of a particular type will almost always be less likely to occur than just one.

From an attacking perspective, a decentralised FPL squad can tap into attacking returns from a wide number of teams, reducing the impact that opposition or team form might have on your likelihood to score points.

If you only had one attacking player from a team, statistically-speaking your odds of owning the player who scores the points is low compared to if you owned three. You also cannot score multiple points from a single goal (i.e. if you own both the scorer and the assister of a goal), meaning that you will typically need more goals to be scored in general than a more centralised team.

This effect can, of course, be mitigated based on who you select. For example, if you own Bruno Fernandes (£11.5m) or Jack Grealish (£7.8m), there’s a good chance you will profit from a large proportion of Manchester United or Aston Villa goals, given their high levels of scoring involvement.

Finally, while a decentralised FPL squad featuring many individual Premier League teams will usually be more work to manage as it requires analysis of more sources of information (e.g. looking at team news for up to 15 teams instead of as few as 5), you are far more insulated against FPL’s more unexpected ‘black swan’ events. These can include the cancellation of a match or unexpected results (e.g. Liverpool losing successive home games to Burnley and Brighton) or the general dip in form from an individual team.

Conclusion

Ultimately, whether you choose a centralised or decentralised strategy is likely to depend on your approach to playing FPL in general. If you prefer more consistent scores (“dullard”), you are more likely to opt for a decentralised approach where the risk is spread more widely and, thus, somewhat mitigated. If you are willing to put up with the occasional blowout Gameweek in order to achieve some huge scores (“maverick”), you are more likely to lean towards a centralised approach which, due to its narrow focus, will produce more extreme results. Both, of course, still rely on the qualitative element of you picking the right individual players.

I’ve personally typically leaned towards the decentralised approach as my goal is generally to try and accumulate consistent scores over the course of a season (I also think it’s usually more fun to have players spread across different teams – particularly if you’re watching the matches) but I’d suggest that, if you are struggling towards the end of a season, adopting a more centralised strategy might be worthwhile as a final throw of the dice. At that point, extreme results may well be what you need and you don’t really have much to lose by trying it.

One final note; the tendency of a centralised approach to produce extreme results is also very often evident in daily fantasy sport games where the best scores almost always come from those managers who focus their selections on just a few teams and happen to nail it that Gameweek. I’d expect that their average scores are usually lower than a player who might adopt a more decentralised approach but, in that environment where it’s all about individual gameweek scores, that really doesn’t matter.