I have had a long-held belief that there are periods of the season when defenders, particularly attacking fullbacks, are the best value players in the FPL game.   A case in point was at the start of the 2018/19 season many teams went with the RAM (Robbo, Alonso, Mendy) line up at the back.  I personally took it one further with TRAM (Trippier in as well).  Those that took this heavy defence approach appeared to reap the rewards, with £6-7m players outscoring the £10m+ Midfield and forward premium options week after week.   Did they actually benefit from this heavy in defence approach?

As an advocate of “defenders are the best value” for many years I am convinced that there has been a tangible shift in the value of defenders over the last few seasons.   This analysis aims to test the hypothesis that defenders have changed in terms of their output, regardless of price, and that for the 2019/20 season we need to be thinking of four or even five at the back as opposed to the traditional three at the back approach most FPL managers default to.

The overriding question I am looking to answer is:

Should we invest heavily in defence for the 2019/20 season?

There are a number of hypotheses to test to answer this question:

  • Have defenders seen the greatest points improvement, of all positions, in Points and Points per £m (PP£m) over recent seasons?
  • How do defenders perform in terms of overall points scored?
  • How do the top six club players perform across each position?
  • How do the top 20 scoring players from each position perform?
  • Has there been a shift in defender points over the last five years?
  • Do we see the same higher points trend from the top six club and top 20 defenders in the last two years?
  • Which defenders should we target and where are these points sourced from?
  • Have we seen similar trends for other positions?

 

My high-level conclusion is:

There has been a shift in the way points are scored and defenders have increased their output at a greater rate than other positions.  There are periods of the season where we should be investing in the back line more heavily than the rest of the pitch.  This marks a move away from the traditional 3-4-3 approach the majority of managers have adopted over the years.  In fact, I would say that 5-4-1 or 4-5-1 should now be the default formation at the start of the season, so long as you invest in the premium defenders (ideally full backs). 

Identifying the best defenders for the season may be hard but investing in top 6/8 club fullbacks is where I would put my money for the GW1 deadline. 

Rules applied during the analysis:

  • Only players playing 60 minutes in a game have been included in the analysis
  • Gameweek grouping is based on eight sections of the season based on five GWs for rounds 1-6 and four GWs for round 7 and 8
  • Top six clubs are Arsenal, Chelsea, Liverpool, Man City, Man Utd and Spurs
  • Top performing players are based on the top 20 highest scoring players (top 10 for GKs) from each season in each position
  • Double Gameweeks are separated so average scores are based on points per games not points per game week
  • Attacking defenders are identified as those who have at least 20% of touches in the final third over the course of that season
  • Attacking midfielders are defined as having 30%+ of touches in the final third

Have defenders seen the greatest points improvements in Points and Points per £m (PP£m) over recent seasons?

In terms of points change (fig 1.0) of all players, goalkeepers have seen the greatest points change.  Among the best FPL assets and top six club players its defenders and midfielders.  When we bring value into the equation (in the form of points per million) there is a clear winner and that’s the defenders, especially when looking at top six club and the best points scoring players.

It does indeed look like the defenders are the place to investigate for this analysis.

Fig 1.0

If you want a lot more detail on player value, please see my article on the subject from last season here:

How do defenders perform in terms of overall points scored?

Initially, I looked at this round by round (fig 1.1) but it proved quite hard to determine the trends due to quite an extreme score variation week by week (particularly around the double game weeks).  To try and identify clearer information I broke the season into 8 periods over the course of the season (Fig 1.2).

Fig 1.1 & 1.2

Game week grouping is based on 8 sections of the season based on 5 GWs for rounds 1-6 and 4 GWs for round 7 and 8

 

Over the last five seasons, defenders are the lowest scoring players of all positions.  This does make sense when you factor this includes all clubs so will include defenders from relegation clubs as well as the top six.  There however there does seem to be a peak for defender scoring at the start (GW 1-5), middle (GW21-25) and end (GW35-38) of the season.   Interestingly this trend is only seen in goalkeepers at the start of the season (the rest is fairly stable) which may suggest it’s the attacking outputs that differentiate these latter periods.

How do the top six club players perform across each position?

Applying a top six club lens we see that the average scoring chart changes substantially (Fig 1.3)

Fig 1.3

Top six clubs defenders begin to become comparable with keepers and almost midfielders with peak performance extending to all bar round 3 (GW11-15) and 6 (26-30).  Top six forwards look to be the consistently highest points scorers and its worth highlighting their performance in round groups 4-6 (GW 15-30).

How do the top scoring players perform across each position?

If we look at the information in a slightly different way, by looking at the top 20 players in each position (top 10 for GK’s), then we see that defenders now outscore keepers and midfielders perform at the same level as forwards (fig 1.4).

Fig 1.4

Again, we see the similar trends for defenders peak scoring time and only round 6 (GW 26-30) being lower than the keepers average weekly score.

Interestingly, the point at which top six strikers seemed to hit real highs (GW 15-30) is contradicted by the best 20 strikers where we see a dip in average points over the same period.  This suggests it may be a time to invest in premium forwards and avoid mid/budget ones.

Has there been a shift in defender points over the last 5 years?

Having identified how defenders score in comparison to other positions it’s now important to look at the year on year trends to see if we can observe any patterns.

There does seem to be a slight shift in the time when defenders score more points (it is quite hard to spot due to season variation across 5 seasons).  The last two seasons defenders have seen higher GW 1-10 and 35-38 points than in previous years, but lower points on average from GW 11-34.

Fig 2.0 & 2.1

 

This becomes far clearer when we look at the last two seasons (17/18 – 18/19) vs the older 3 seasons (14/15 – 16/17) from the data.  From the chart below (See Fig 2.1) it appears that defenders score fewer points now than they did in previous years with the exception of the first 10 GWs and the last 4 GWs.   This is very interesting and possibly reflects the impact of the shift of power in the last 5 years with big clubs scoring significantly more and everyone else conceding significantly more (fig 2.2)

 

Fig 2.2

 

Given this change its important to see the impact of the top six teams defenders over the last 5 years (fig 2.3), the last two seasons (fig 2.4) and the change over time of the top six defenders (Fig 2.5)

When looking at groups of clubs (top six, stable ones and the others/relegation candidates) we can see that over the last 5 years there is a clear gap (of 1 point per GW) between the top six clubs which is very significant as it represents an average of almost 40 points more over the season.

 

Fig 2.3 & 2.4

When looking at just the last two seasons (fig 2.4) it looks as if the top six players score higher and the remaining defenders score lower with the gap increasing to an average of ~1.3pts extra per GW.

Given the shift in the last two years investing in the top six defenders looks like a very sensible approach when investing at the back.  This gap is most pronounced in the first 10 GWs of the season where top six club defenders appear to score almost 0.6 points per week more in GW1-10 and 35-38 than they did in previous season (this represents between 1.5-2 more points per week on non-top six defenders).

A more detailed comparison of other positions can be seen at the end of this article in fig 3.3

 

Fig 2.5

Looking further at the top six club defenders in isolation (fig 2.5) we can see that there has been an uplift in their points in the last two seasons.  This is most pronounced in the first 10 GWs (group 1 & 2) and last 4 (group 8) weeks of the season.  This poses the question is this driven from an increase in clean sheets or attacking returns…. which we will come back to in a minute.

Before heading down that tangent we need to check:

Do we see the same higher points trend from the top 20 defenders in the last two years?

Fig 2.6

We do and the points difference is large with the gap being as much as 1 point per GW better in the early weeks vs previous seasons.

At this point it’s fairly safe to conclude that investing in the right type of defenders in the early part and very end of the season is better now than it ever has been before.

 

So which defenders should we target and where are these points sourced from?

With support from Peter Blake I have been able to identify attacking defenders, i.e. full backs, based on the number of touches in the final third of the pitch from OPTA stats.  Applying this we can see (fig 2.7) that investing in attacking defenders is a better way to return points (regardless of club or being a top scoring or just average FPL asset).   This applies when looking at all or the top six club (fig 2.8) or top 20 performing defenders (fig 2.9).

The message here is invest in offensive defenders where possible, regardless of the team.  Now this is probably a no-brainer but it needs to be noted.

 

Fig 2.7, Fig 2.8 & Fig 2.9

 

Going a step further and looking at the difference over the last two years it looks like there are periods in the season when the full backs significantly outscore the centre back peers (fig 3.0).  This appears to be in the first 5 GWs, and across the second half of the season.  One theory could be that we have seen a change in playing style recently, combined with teams approaches to games at points in the season.  At the start and end of the season there, from my own perspective, seems to be more focus on using full backs to attack.  Weeks 6-20 tend to be more about staying solid defensively.

 

Fig 3.0

Assuming we buy into the fact that premium defenders are now more valuable assets than they ever have been before the next question is

Where do the increased points come from?

Looking at the top six club defenders (fig 3.1) we can see that there are more clean sheets at the start of the season but for the rest of the season clean sheet rates are consistent with previous years.  Among full backs we see significantly higher assist rates than in previous years which I believe reflects a change of playing style for many clubs (getting the width from the full backs).  Interestingly Centre backs goal threat seems to have increased during the mid-part of the season GW15-30 so an interesting approach could be to start with fullbacks and switch to centre backs around November/December.

 

Fig 3.1

By contrast, when looking at all defenders (fig 3.2) it appears there is no difference in assists over the seasons.  Clean sheets seem to be prominent at the start of the season but then align to previous seasons.  Interestingly full backs have scored considerably less goals and centre backs have scored more goals.   This doesn’t align to the top six views (despite the impact of the wolves’ wing backs last year).

 

Fig 3.2

 

Have we seen similar uplifts for other positions?

Very interestingly midfielders have seen the largest uplift in points scored and seem to be the standout area of improvement (fig 3.3).  So I have to ask myself has this article been worth it at all?

Digging further this points uplift has predominantly been driven by the classification of high priced wide attacking midfielders such as Salah, Stirling and Hazard who could just as easily be classified as forwards.  When you factor in points per million it is still the defenders who come out on top on the greatest proportional improvement in recent years.  Thankfully after all the effort put into this piece of analysis, the best returns are still the best defenders i.e. top six club or best 20 performers (fig 1.0).

 

Fig 3.3

Fig 1.0

Conclusion

There has been a shift in the way points are scored and defenders have increased their output at a greater rate than other positions.  There are periods of the season where we should be investing in the back line more heavily than the rest of the pitch.  This highlights a move away from the traditional 3-4-3 approach the majority of FPL managers have adopted over the years.  In fact, I would go so far to say that 5-4-1 or 4-5-1 should now be the default formation at the start of the season, so long as you invest in the premium defenders (ideally full backs).

Identifying the best defenders for the season may be hard but investing in top 6/8 club fullbacks is where I would put my money for the GW1 deadline.

A note of caution is there are periods where forwards and midfielders seem to dominate so don’t commit to this tactic without the flexibility to scale up your FPL teams attacking options.

In simple terms, investing heavily in Premium defenders for GW1-10 and switching to a slightly more offensive approach between GW 15-30 before reverting back to a heavy defence for the last few weeks would be my conclusion for this year.

Thanks for reading and good luck for the coming season

Luke (@elldubya1 on twitter)

Become a Fantasy Football Hub member for just £1

You can join hundreds of fantasy managers and access this and all our membership content for just £1. Simply enter coupon code: PRESEASON at checkout. You’ll get:

  • OPTA Data (NEW).
  • Entry into £5,000 prize league (NEW).
  • Members’ slack Channel (NEW)
  • Ben Crellin’s FPL planning sheet.
  • How To Win at FPL guides.
  • Player comparison tool.
  • Team reveals form top-ranked players.
  • Rate my team tool.
  • Advanced fixture ticker.
  • FPL watchlist tool.
  • Minimum 100 extra articles each year.
  • The Algorithm forecasting model for FPL.
  • A lot more!

Membership Registration

Follow the Hub

Twitter

Facebook

YouTube

Instagram

View all our latest fantasy football tips on our Homepage