Welcome to part 18 of the season review. We are still on the topic of expected events versus actual events using possession as the pivot.
Today we shall be looking at shots against and how each team used their possession. Join me after the jump to see which teams exceeded their expected shots conceded totals and which did not.
A quick explanation is needed for how we work out what each teams expected shots conceded total was.
Average EPL possession over the seasons course 50%
Total shots against in the EPL 2011/12 10932
Average per team 546.6
Average per 1% of possession 10.932 (546.6/50)
Arsenal as an example- 398 shots against, 6.77 shots per 1% of Arsenals possession (58.76%). We see that Arsenal's actual shots against number easily beat their expected number.
Actual shots against per 1% of possession was 9.65 vs 10.93 expected shots conceded per 1%. To get that figure we use Arsenal's shots against total/possession conceded (41.24)=9.65 per 1% of possession.
Expected shots against is a teams possession% x 10.932 (the League's average shots per 1% of possession)
Shots Against With Possession
The above graph shows us the raw shots against numbers. The totals are for the 38 game EPL season. Team and possession in brackets are listed along the bottom of the chart. Possession declines going from left to right. As that possession declines we roughly see shots against increase.
Shots Against Per Teams 1% Of Possession
We see some strange results here, Stoke and Everton concede fewer shots than their individual possession number expected them to do.
Swansea despite the plaudits were leaking shots against at an alarming rate per 1% of possession.
This is the actual shots against per 1% of possession vs the expected shots against per 1%. We can see if a team exceeded the 10.93 average per 1% of possession or failed to match that number.
We see some interesting results here. Everton lead the league in actual shots against vs expected shots against. We know they are a stubborn team and we expect them to be difficult to create against. But still, this number is a surprise.
Stoke are the most interesting team featured above. Without this excellent +1.84 per 1% number, it's doubtful they stay in the EPL.
Swansea, who had the terrible expected shots for +/-, once again struggle by this measure. They don't fit the pattern of a normal high possession team. They had less shots for than expected and now they have more shots against than their possession expected of them. Swansea struggled with shots for and shots against despite their excellent 57% possession number.
Actual Shots Against v Expected Shots Against (38 Games).
So, expected shots is the ascending blue bar. Simply put, a team is expected to concede more shots the less possession they have. We can see above who is a + or - by this metric. But let's make it easier.
Actual Shots Against vs Expected Shots Against (38 games)
Here we see the +/- differential for the 2011/12 EPL season.
We expect the better sides in the League to have a + number, Arsenal, Man City, Liverpool and Chelsea prove that theory right. But what of Man Utd?
They conceded 48 shots more than their possession number indicated they should have. Why was this? Inconsistent selection in defense, the absence of Vidic, anyhow, United conceded only 8 shots on target more than was expected. The point being they gave up 48 more shots than they should of but those shots didn't translate into shots on target against. Were they allowing teams to shoot from distance and thus allowing those low% scoring shots?
Stoke, again are sensational. Everton, ditto. Aston Villa are a + team here conceding 30 shots less than expected, this was important for a team who didn't score a lot.
Actual Shots Against vs Expected Shots Against (per game)
Swansea are struggling, and although these charts show that their possession % was wasted somewhat, overall possession is a decent indicator over 38 games. Swansea are the anomaly for a high possession team in that they were poor against their expected numbers.
Everton conceded 2.55 shots per game less than was expected. Can it be sustained? Probably if tactics persist.
Stoke survived in help due to their 2.89 less shots conceded per game. Over a season that number matters in terms of goals against and thus points.
8 out of the top 10 EPL teams posted a + number by this measure ( ie conceded less than was expected). The odd teams out were West Brom with 0.05 more shots per game against than was expected, a measly number, and Man Utd who conceded 1.26 more shots per game than was expected.
What was happening at Man Utd for that team to be so much worse than the rest of the EPL's top ten in terms of actual shots against vs expected shots against?