EPL Analysis and GW 38 Prediction
A data-driven attempt in predicting English Premier League results using xG Statistics
This is the last game week of a season which would be remembered for various reasons. This season has been a roller-coaster ride.
This is the final article on my EPL Prediction series. You can check out the previous predictions here.
Expected Goals or xG is the parameter used for prediction. If you are interested in understanding the algorithm for prediction, I recommend that you check out this article where it is explained in detail.
Data Courtesy: Football Reference
Analysis up to Game-week 37
As we get to the last game week of the season, Man City has already been crowned the champions, and relegation battle is over. These facts can dent the importance of the last game-week usually, but not this season!
This is the second match of the season in front of an audience, and first for the 10 clubs who are hosting this game week.
Apart from the fans sentiments about returning to the stadiums, what makes the league still interesting is the fight for European spots.
Both Machester teams have ensured berth to the Champions League securing the top two spots in the league. One out of the Blues, the Reds or the Foxes will have to settle for UEL.
A draw in the last mach against the Saints can ensure Europa League entry for Moyes’ Westham. However, if they lose, the Spurs can leapfrog them to secure a Europa by spot beating the Foxes at the King Power stadium.
The team in seventh position can secure entry into the inaugural edition of the Europa Conference League. Which gives a small chance for Everton and Arsenal at 8th and 9th position respectively also have chance of competing at the European stage.
After 37 game-weeks, the champions, Man City dominates both attack and defense as it has the highest xG scored and the lowest xG conceded per game.
The Reds are in second position with respect to xG scored. They are followed by the the Blues and the Red Devils who have scored more than 1.5 xG per game.
Crystal Palace, Sheffield United and West Brom lack creativity up front, they were not able to even create 1xG per game.
Manchester City with around 0.75 xG conceded per game, seems to the the best defensive unit despite conceding a few goals in the recent matches.
Chelsea, whose defense has considerably improved after Tuchel took charge, is at a close second position. The Seagulls are the only other team which has conceded less than 1 xG per game.
Most teams concede between 1 to 1.5 xG per game. West Brom, Sheffield United and Leeds United have failed to stop their opponents from creating goal scoring opportunities, having conceded more than 1.5xG per game.
Based on xG Scored and xG Conceded, teams can be grouped into 4 quadrants as shown in the above graph.
The horizontal dotted line shows the average xG scored per game. Teams above the horizontal dotted line are strong attacking sides and the teams below, weak in attack.
The vertical dotted line shows the average xG conceded per game. Teams to the left have a strong defense and the teams to the right have week defense.
Man City is head and shoulders above the other teams with respect to delta xG, with a difference of more than 1 between xG scored and xG conceded. Liverpool and Chelsea are the the only other teams with delta xG above 0.5. West Brom and Sheffield United, with delta xG less than -0.75, on the other hand are struggling at both ends of the pitch with the weak attack and porous defense.
Fulham, the third team to get relegated have performed better than several other teams in both xG scored and xG coceded parameters. Scott Parker and team would be cursing their luck to get relegated after putting up a decent performance.
Leeds United ranks among the top in xG created per match. However, the team has negative delta xG as opponents find it easy to penetrate the Leeds defense. The team has improved their defense in the last few matches. With the experience from this season, next season, Leeds would be a force to reconed with.
Wolves on the other hand is a good team defensively, but they lack the strike force upfront. Raul Jiminez was out for nearly the entirety of the second half of the season due to injury.
Brighton as per the delta-xG table is placed at 5 position, however they are in the 16th position in the league as they lack quality strikers up front.
Game Week 38 Predictions
Before proceeding to the predictions, let me clarify that this is a very simple algorithm just based on past xG, so only baseline performance can be expected. The algorithm also fails to predict high scoring games. The model also does not take into account the team selection, absence of players due to injuries/suspension, formation, tactical changes etc.
However, the model has been performing pretty well in predicting the momentum of the matches. You can check out how the actual performance fared against the predictions made for the previous game week below.
EPL Analysis and GW 38 Prediction
Predictions for Game Week 38 are provided in the table below.
The absolute value of GD shows the competitiveness of the match. The higher the value, more one sided the match is expected to be and higher the accuracy of prediction.
The lower the value of GD, the more the match could be anybody’s game. Positive value of GD means Home win and Negative value means Away team win.
Liverpool vs Crystal Palace is expected to be the most one-sided match of the weekend. The Reds are estimated to steamroll over the Eagles. The champions who host the Toffees in front of their pumped up home fans is expected to result in a comfortable win for the hosts. Leeds is also estimated to secure 3 points from their home outing against the relegated Baggies. Chelsea and Manchester United are expected to win their away outings. The odds are also in favor of the Hammers in their home outing against the Saints.
Fulham vs Newcastle is expected to be the most closely contested match of the game-week, it is literally anybody’s game.
The Foxes hosting the Spurs is the most anticipated game of the game week. European aspirations of several teams are hinged on the outcome of this match. We algorithm estimates a very tight contest with the Foxes having an slight edge over the visitors.
Update: Predictions Vs Actual Results
The algorithm was able to correctly predict the direction xG in all of the games where the expected delta xG was more than 0.20 with 100% accuracy. This is an incredible result considering the simplicity of the model.
In the matches where the predicted delta xG was less than 0.20, two resulted in a balance of delta xG (SHU vs BUR and LEI vs TOT).
Become a Member
I hope you like the article, I would highly recommend signing up for Medium Membership to read more articles by me or stories by thousands of other authors on variety of topics.
Your membership fee directly supports me and other writers you read. You’ll also get full access to every story on Medium.
Other Articles You May Like:
