Following on from my article introducing the concept of xFPL I’ve attempted to add some numbers to the defensive side of the game. Last season there were 5 possible scoring outcomes defensively in FPL.
I’ve taken the expected goals against (xGA) data from each game last season and matched it up with each of the above outcomes. This means I can now look at the xGA from a game this season and compare it to the 50 closest xGAs from last season to give me a % chance of each defensive outcome above.
For example, Southampton lost 2-1 to Leicester at the weekend with an xGA of 0.51.
Last season the 50 closest xGAs to 0.51 ranged from 0.45-0.58.
Of those 50
- 29 (58%) resulted in a clean sheet,
- 19 (38%) resulted in 1 goal conceded,
- 2 (4%) resulted in 2 or 3 goals conceded.
- 0 resulted in 4+ goals conceded.
This would indicate Southampton had a better potential for defensive FPL points than the scoreline suggests. To translate this to an xFPL value I’ve used the following calculation:
0.58 (clean sheet %) x 4 (clean sheet points) = 2.32
0.04 (2 or 3 goals conceded %) x -1 (2 or 3 goals conceded points) = -0.04
2.32 + -0.04 = 2.28.
So Southampton are given a defensive xFPL of 2.28.
I have chosen to use the 50 closest for the time being but as I gather more data throughout this season, and possibly add the data from the 16/17 season, there is scope to increase that number and hopefully make the percentages more accurate.
The combined defensive xFPL of each team for the first 3 games of the season is shown in the following table.
Liverpool are yet to concede a goal and so unsurprisingly lead the way in defensive xFPL. Chelsea, Bournemouth and Watford have all registered one clean sheet so far and their xFPL suggests this is fair with them all having a score close to 4. Wolves are yet to record a clean sheet but sit 5th for defensive xFPL. Bennett at 4m and Doherty at 4.4m could be bargains if they can translate that potential into clean sheets.
The bottom of the table makes grim reading for the 24.2% that own De Gea with Utd having a defensive xFPL of -0.69. Although their fragility at the back should give him the opportunity for save points. West Ham and Huddersfield defenders appear to be worth avoiding but have both had to contend with a tough opening schedule of fixtures.
The following table shows the top 25 players for xFPL so far this season.
Salah is way out in front with Liverpool teammates Mane and Robertson also in the top 4. Callum Wilson in 2nd is outperforming his budget price tag and repaying the faith of those managers that were willing to overlook his underwhelming performances in the lead up to the first blank gameweek last season. Aymeric Laporte could be an interesting City option at 5.5m and only 2.7%TSB. He’s currently the best performing City player in terms of xFPL and if anything looks less likely to fall victim to ‘Pep Roulette’ than their more expensive midfield options. Agüero and Mendy are both in the top 25 and with ownership percentages of 48.4 and 33.1 respectively look essential going forward. Chelsea have some good upcoming fixtures so Alonso (6th) and Hazard (17th) will undoubtedly attract plenty of attention. Alonso’s attacking intent doesn’t seem to have been dampened under Sarri and Hazard’s numbers are particularly impressive given he’s played just 133 minutes so far this season. For those looking for a Richarlison replacement this week there are 3 candidates in the top 25 at a similar price bracket. Fraser (11th) leads the league for expected assists with 1.7, Milner (13th) surprisingly took penalty duties against Crystal Palace so could be of interest if you aren’t already tripled on Liverpool and Walcott (18th) who has already been transferred in by more than half a million managers this week.
It is still too early to draw any strong conclusions from the data after just 3 matches but hopefully xFPL can help to identify some potential differential gems in amongst the obvious bandwagons. I will definitely be strongly considering Laporte for my 3rd City option and Doherty as a budget defender in my team in the forthcoming weeks.