By Thomas Negron
Last season’s Premier League title was all but won last April at Emirates Stadium where Chelsea secured a 0-0 draw to maintain a 10 point league lead over Arsenal with just five matches left to play. As the Chelsea players celebrated the likely league title, Arsenal supporters belted a chorus of “Boring, Boring Chelsea.” Jose Mourinho, the manager of Chelsea, has always been known for his defensive, pragmatic approach to the game, which became very evident in the second half of last season with strong defensive showings week in and week out. However, only four months after winning the Premier League title, Chelsea sits near the bottom of the table and has conceded the second-most goals of any team in the Premier League. I decided to look into exactly what has happened to Chelsea’s in the first six games of this Premier League season.
Last season, Chelsea conceded 0.84 goals per match in the Premier League. Through seven games this season, they have conceded 2 goals per match. A two-sample t-test with this data yields a p-value of 0.02189. For those less comfortable with statistics, this essentially means that if the true mean for goals conceded per match for Chelsea was the same for this season and last season, this would only happen 2.189% of the time. While a tougher schedule could account for conceding more goals and attaining worse results, Chelsea got five more points out of the same seven fixtures from last season.
A lot of this boils down to a statistic known as TSR (total shots ratio) which simply measures the number of shots your team takes divided by the total shots taken in a game. Last season, Chelsea had a TSR of 57.6%. So far this season, their TSR is only 54.1% and this is with their average shots per game increasing from 14.8 to 16.4. Chelsea are conceding 13.9 shots per game on average, much higher than the 10.9 shots per game they conceded last year. This has lead to increased work for their goalkeepers, who are averaging 4.3 saves per game compared to the 2.4 saves per game they had last season. This decline is evident in their defensive statistics. Their tackling percentage (successful tackles/attempted tackles) has fallen about three percentage points, making them one of the worst tackling outfits in the Premier League. Their clearance rate has also dropped significantly (from 27 clearances/game to 18.1 clearances/game) which relates to their inability of stopping teams from getting shot opportunities.
While the quantity of shots that Chelsea have conceded this season has increased, it also seems that the quality of the shots has increased as well. Last season, 29.7% of shots taken by their opponents were on target [(saves+goals)/shots]. This season that number has shot up to 45.3%. Another thing we could look at with these numbers is the save percentage [saves/(saves+goals)]. Their save percentage was 74.1% last season and is 68.3% so far this season. Doing a two-proportion t-test for this data gives us a p-value of 0.227, which is not statistically significant. While we cannot statistically say that goalkeeping skill has declined for Chelsea this season, most people would expect it to have, as starting goalkeeper Thibaut Courtois has been out for a significant amount of time with a knee injury.
While Jose Mourinho continues to pass the blame for poor results from the team doctors, to the referees, to the FA, he is surely worried about the defensive start of the season for his team. It is almost certain that Chelsea has a much better defense and a better team than indicated so far this season. A decline this big with a team that is for the most part the same, would be extremely unlikely. While a regression to the mean is almost certain, this start points out some big issues with the Chelsea side. These trends will be important to look at as the season progresses and could lead to some important transfer business this January from the London club as they attempt to defend their title.