Assessing the Projections: 2013-14 English Premier League
Categories: League Competitions, Soccer Pythagorean: Tables, Team Performance
There is one more projection left to assess in the English professional game this season: the 2013-14 English Premier League.
Here is what we projected at the start of the season:
And here is the final league table:
Pld | W | D | L | GF | GA | GD | Pts | |
Manchester City | 38 | 27 | 5 | 6 | 102 | 37 | 65 | 86 |
Liverpool | 38 | 26 | 6 | 6 | 101 | 50 | 51 | 84 |
Chelsea | 38 | 25 | 7 | 6 | 71 | 27 | 44 | 82 |
Arsenal | 38 | 24 | 7 | 7 | 68 | 41 | 27 | 79 |
Everton | 38 | 21 | 9 | 8 | 61 | 39 | 22 | 72 |
Tottenham Hotspur | 38 | 21 | 6 | 11 | 55 | 51 | 4 | 69 |
Manchester United | 38 | 19 | 7 | 12 | 64 | 43 | 21 | 64 |
Southampton | 38 | 15 | 11 | 12 | 54 | 46 | 8 | 56 |
Stoke City | 38 | 13 | 11 | 14 | 45 | 52 | -7 | 50 |
Newcastle United | 38 | 15 | 4 | 19 | 43 | 59 | -16 | 49 |
Crystal Palace | 38 | 13 | 6 | 19 | 33 | 48 | -15 | 45 |
Swansea City | 38 | 11 | 9 | 18 | 54 | 54 | 0 | 42 |
West Ham United | 38 | 11 | 7 | 20 | 40 | 51 | -11 | 40 |
Sunderland | 38 | 10 | 8 | 20 | 41 | 60 | -19 | 38 |
Aston Villa | 38 | 10 | 8 | 20 | 39 | 61 | -22 | 38 |
Hull City | 38 | 10 | 7 | 21 | 38 | 53 | -15 | 37 |
West Bromwich Albion | 38 | 7 | 15 | 16 | 43 | 59 | -16 | 36 |
Norwich City | 38 | 8 | 9 | 21 | 28 | 62 | -34 | 33 |
Fulham | 38 | 9 | 5 | 24 | 40 | 85 | -45 | 32 |
Cardiff City | 38 | 7 | 9 | 22 | 32 | 74 | -42 | 30 |
In early August I thought Chelsea had the talent to win the league in a close race from Manchester City and Manchester United (Arsenal finishing fourth again), and I picked all three promoted sides to go straight back down. As it turned out, it was a very close race, and Arsenal did finish fourth once more, but Manchester City finished on top, Chelsea ended four points back, Manchester United fell away from the top four and Liverpool replaced them in a thrilling season.
The English Premier League is probably the most scrutinized soccer league in the world, almost certainly the case from a statistical point of view, so it’s not too surprising that the projections were the most accurate of the four leagues in the English professional pyramid. Goals scored were predicted within ten goals for all except five teams, and goals allowed were predicted within ten goals for all but six teams. But it’s the significant over/under-performance of clubs that has made the difference in the Premier League’s final outcome.
If you consider goals allowed to correlate with the strength of a side’s defense — and after all, the objective is to prevent the other side from scoring — the defenses of the top four finishers were pretty much as expected. The difference was offensive production — Liverpool had the SAS, and Manchester City had Yayá Touré, Edin Dzeko, and Sergio Agüero. The result was that City scored 32 more goals than expected and won 11 more points than projected, while Liverpool exceeded their offensive expectation by 39 goals and ran 24 points ahead of their projected totals. Now the league did turn on a few matches and specific events, but over the entire season Liverpool’s free-scoring transformed them into a side with a serious chance of the title.
Few people expected Manchester United to win the league in the first season of the post-Sir Alex Ferguson era, but Manchester United’s failure under David Moyes to finish in the top five alarmed many. In my opinion, the failure wasn’t down to defensive underperformance; rather, it was the inability to get offensive production that caused them to win eight fewer points than expected. Offensive production isn’t just the responsibility of the strikers, although the only United players to rank in the Premier League’s top twenty-five scorers were Wayne Rooney, Robin Van Persie, and Danny Welbeck. It’s also the result of opportunities, or lack thereof, provided them by the midfielders.
The biggest surprise in the division beside the two Liverpool clubs was Crystal Palace. Selected to go straight down after promotion, they probably would have gone down if Tony Pulis didn’t replace Ian Holloway in late October. Offensively, their output was just as expected, but Palace’s defense was hugely overperforming — 16 goals better than expected. It’s not surprising for a Tony Pulis team, and I feel that that performance in defense earned them five wins that they might not have earned otherwise.
In general when I compare the actual league table to our projected one, the teams who score or concede goals well ahead of expectations invariably end up at the top of the table. If those teams fall way behind expectations, they either fall out of contention for a European place or fall into danger of relegation. This last scenario was true of Fulham and Cardiff City, whose defenses were very poor, and Norwich City, who were unable to score.
The statistical summary:
Pred Goals | Actual Goals | % Change | RMSE Pts | RMSE GF | RMSE GA | RMSE Pos |
1015 | 1052 | +3.65% | 10.93 | 13.33 | 10.94 | 3.73 |