In the previous article we looked at how strength of schedule impacts a strikers performance in FPL. The data showed that we should avoid strikers’ games against six because fantasy output is much lower, but we also found that strikers averaged more points against some of the middle-of-the pack teams compared to those near the relegation zone. I ran the same analysis for midfielders, defense, and goalkeepers to see if the striker trends hold for other positions as well. As always here are the datasets and assumptions I’m using:

  • I’m using 2016-2017 FPL data for all players through the entire season. The 2017-2018 data is not yet available for all players.
  • Opponent strength is determined by their league position at the end of the season. While teams fluctuate in their ranking throughout the season, end-of-season league position provides a fairly stable estimate of the team’s quality.
  • Only starting players who played at least 60 minutes. In some cases, subs came on and got some points, but to get a nice robust dataset, I limited the players that were analyzed to those who played at least 60 minutes.

Midfielders
It looks like there is a fairly consistent trend among starting midfielders last year: they scored progressively more points against opponents lower in the table with the exception of Stoke and Burnley (who ranked 9th and 10th in goals conceded last year, but finished 13th and 16th in the table).

Screen Shot 2018-02-08 at 11.09.01 AM

 

Defenders

The difference in defensive players’ FPL output is much more pronounced when we look at strength of schedule. Defensive players averaged between 1.6-2.1 points per game when facing one of the top teams. This is pretty obvious; the top teams have high-powered offenses and are more likely to score. The rest of the table generally follows the trend that the worst teams give up the most points to defensive players. The one outlier I want to point out is Southampton. The Saints finished well last year, reaching 8th place with 46 points, but they only scored 41 goals (half as many as Tottenham or Chelsea). Although they don’t look like a great matchup for your defensive players based on their table position, their anemic offensive production says otherwise. They conceded the 16th most FPL points to starting defenders.

Screen Shot 2018-02-08 at 11.09.51 AM.png

Goalkeepers

Finally, we get to goalkeepers. As you can see from the color-coding in the table below, keepers’ performance does not seem to correlate as well with their opponent’s position in the table. Generally, it looks like you want to avoid playing your keeper against the top 5 teams, and you want to play your keeper against the bottom 5 (obvious, I know). The data gets a little goofy when keepers play teams in the middle of the table. Man United, Southampton, West Brom, and Leceister were all more favorable matchups for keepers despite their relatively high league position. Meanwhile, keepers averaged the same or fewer points against Bournemouth, West Ham, and Stoke, compared to Tottenham and Liverpool. When choosing your keeper, I would be wary of using strength of schedule to make your selection.Screen Shot 2018-02-08 at 11.10.20 AM.png

What to do with all of this information?

When you’re deciding between players to transfer in or who to captain, don’t rely on a player’s strength of schedule alone, especially for keepers and strikers. You’re probably safe to bring in midfielders and defenders who play mid- or bottom-table teams in their next few fixtures. I would warn you, however, that if you only use strength of schedule to make your decisions, you may suffer some poor performances. I recommend you use at least two other data points to decide if an opponent is a good matchup. For defense and keepers, think about data on opponents’ chances created and goals scored. For midfielders and forwards, think about opponents’ goals conceded and clean sheet percentages. Don’t be the guy who says, “Bro, you’re thinking too hard. Just go with strength of schedule.”

Nick Zupan

The Art of The Dive

 

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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