Fantasy Football

Fantasy Premier League Strategy: Four at the Back?

As the 2017/18 season rapidly approaches, there has been much pre-season talk about the virtues of playing four (or more) defenders in Fantasy Premier League (FPL) starting line-ups. The conventional wisdom until this season has been to operate a 3-4-3 system, but the emergence of wing-backs as a feature of many Premier League sides towards the end of the 2016/17 season has lead to an increase in interest of the attacking defenders. As a consequence, many FPL managers are looking at 4-3-3 or 4-4-2 set-up.

The appeal is understandable. These players are defenders in the game, and so qualify for the four points for a clean sheet, but they operate like midfielders who play a key role in attack; players such as Chelsea’s Marcus Alonso and Tottenham’s Danny Rose were frequently among their teams’ furthest forward positioned players last season.

However, I’m not sold on the idea, and the graphic below demonstrates why. The table shows the 2016/17 count of players in each position to score over 100 points. The values in the cells relate to the number of players available at each starting price, with a tolerance of +/-0.5m. In other words, the number of players within a £0.5m range of the starting price. For example, there were four forwards with a starting price of within £0.5m of £9.5m (that is, £9.0m-£10.0m) who scored 100 points or more. I have set the cut-off as 100 points as an arbitrary boundary, because I believe this represents the minimum amount of points for a viable FPL option, although position and personal preference would be variables if others were to perform this analysis. The £0.5m tolerance represents realistic transfers which can be made without disrupting the balance of the squad with a single free transfer.

Figure 1: count of players to score 100+ FPL points in 2016/2017 season (+/- £0.5m tolerance)


What we can see here is several key points which could be instructive for team structure:

  • The most options for viable defenders come in the £5.5m range (26 players within £0.5m of this total)
  • The most options for midfielders are in the £8.0m range (18), and for forwards it is in the £6.5m / £7.0m range (seven each).

How this data is interpreted is up to the reader, but my view is that the budget defenders and midfielders provide the most intrigue (along with the mid-range strikers, but that’s for another post). We can see that there were a lot of options in the mid-upper price defender bracket, and we can expect this to increase this season; in 2016/17 there were 14 defenders priced £6.0m or higher, this coming season there are 20, and the proliferation of the wing-back system in the Premier League makes them attractive options. However, what should be noted is the opportunity cost of picking these premium defenders (that is, what would have been sacrificed elsewhere).

Taking the typical structure of a team (before I continue, I repeat this is ‘typical’; it is by no means definitive and there are numerous ways of structuring a team that will suit different FPL managers, and for an interesting perspective read Firetog’s post on Fantasy Football Scout), it is common for as little money to be tied-up on the substitutes bench as possible, hence the popularity of the £4.0m goalkeepers and defenders in the game, what are commonly referred to as ‘enablers’; they free up money to be spent elsewhere in the squad. In the traditional 3-4-3 dynamic, it would be common to see on the bench:

  • a £4.0m goalkeeper
  • x2 £4.5m defenders (or x1 £4.5m and x1 £4.0m)
  • a £4.5m midfielder

Looking at the table above from last season’s points, we can see that there were 6 goalkeeping options within £0.5m of £4.0m, 14 defenders within range of £4.5m, and 7 midfielders within the same range.

Now let’s suppose that we play a 4-3-3 system. The bench would probably look like:

  • a £4.0m goalkeeper
  • a £4.5m defender
  • x2 £4.5m midfielders

This for me highlights the problem of the four at the back system. In the 3-4-3 formation, you would have been required to find two £4.5m defenders from the 14 viable options (of 122 total players in that range to score at least a point, or 11.4%), but in the 4-3-3 system, you needed to find two £4.5m midfielders from just seven options (of 109, 6.4%). In short, your bench options were significantly reduced in these two positions, limiting the overall effectiveness of the squad.

(As an aside, a 3-5-2 is not an option if you want a strong squad because there were no 100+ options below £6.0m, which I would argue is too much money to have on the bench. If you go for a £4.5m forward, don’t expect him to do much for you).

There are of course arguments against this and ways around it:

  • A counter-argument could be that you don’t need strong options on the bench, and that a strong first XI is the priority. However, the rotation experienced by all Premier League squads finds fault with this logic.
  • One could wait to see who the seven £4.5m midfield options are likely to be and transfer them in when it becomes obvious. Setting aside the question of whether spending a free transfer on a bench option is an efficient use of it, I personally would like to have a greater range of players to choose from, which the defenders afford you. The factor to consider here is that with fewer budget midfielders to choose from, the concentration of ownership is heightened, which can affect price rises and drops more significantly. By contrast, with budget defenders the ownership is spread further, stabilising their prices.
  • The first wildcard can provide a work-around for those looking to wait and see. If the first few weeks highlight a few top quality midfield options at £4.5m, and their form and place in the team looks sustainable (unlike, say, Capoue for Watford in 2016/17) then they could provide the structure required for a long-term budget midfield. It is therefore an option to start with the four at the back and budget midfielders and hope that a couple of candidates cement their positions by the time the first wildcard is used. It is a risk though.

For me, I am looking to play a 3-4-3 in the coming season not because I think there is necessarily the best value to be found in the starting XI using this system, but because this gives me the best chance to load my bench up with value. The appeal of the premium defenders is obvious, however by taking four of them it means that you are reducing your chances of building a strong 15 man squad, and one of my main strategies when playing FPL is to have as many options available as possible both in terms of squad personnel and balance. The likelihood that you will not need to call on your bench during the season is very low considering the level of rotation that can be expected to afflict many Premier League squads, with the principal candidates for this rotation actually being the wing-backs when considering the physical demands placed on them.

Personally, when it comes the fringes of my squad I like to plan for the long-term. This means that I’ve only got 8 or so key players set up to maximise the early fixtures; if I need to wildcard in gameweek six then so be it, but if they start well and I manage the five free transfers correctly, I may be able to extent the length of time needed before the wildcard is played if my fringes are optimised. Taking a risky x2 £4.5m midfielder strategy which has a lower probability of success from the off means you are more likely to be locking yourself into an early wildcard in order to repair the fringes of your squad. With a three at the back system, I may well need to do that also, but if I can reduce the chances of doing so by just a few percent, I’m going to do that. Remember that FPL is a game of probability, so I’m looking to play the odds whenever possible.

If we are to assume that an FPL manager is looking to have as much money in the starting XI as possible but still wants options across the whole of his or her 15 man squad, then historical data suggests the 3-4-3 is the formation that has the greatest probability of delivering this.


As always, find me on Twitter @artemidorus_1


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s