The Premier League Shockers – How common are the shocking upsets?

[quote]Football is an incredible game. Sometimes it’s so incredible, it’s unbelievable. – Tom Landry[/quote]

Ok, so Landry was talking about the ‘other’ type of football, however the sentiment remains true for ‘real’ football.  One of the most beautiful things about the beautiful game is the potential for a freak, unexpected result.  An incredible, unbelievable result.

There’s nothing that neutral football fans enjoy more than seeing an underdog grabbing a win against the odds.  So how often does it happen?  And who are the best at it?

The very fact that a shock result generally means a ‘weaker’ team gaining points from a ‘stronger’ team naturally excludes a number of teams from doing well in this scenario.  Manchester United, Manchester City and, to some extent, Chelsea are expected to win almost every game in a season therefore reducing their chances of securing a shock result.  Even Man Citys 2-1 win at Old Trafford in April wasn’t considered a shock.

For the purpose of this post I have defined a shock result as one being given odds of 5/1 or more by bookmakers.  Why 5/1?  Blame my dad.  He once told me that any bet less than 5/1 wasn’t worth the shoe leather for the walk to the bookmakers, and i’ve continued his rule on.  Also, 84% of odds quoted in the Premier League this season (up to 12th May) were at odds of less than 5/1, thus meaning that 16% of results could be classed as a shock.

So, how many results have there been that meet my criteria?  Looking at the last 4 seasons of the Premier League:

Including season totals and percentage of total games played:

The figures show that, on average just over 6% of games end with a shock result and, with the exception of 2009/10, away shocks are vastly more common than home shocks.  This makes sense as it is generally accepted that it is harder to win away, even without the differences in squad abilities.  The number of shock results has reduced massively this season, possibly suggesting that the gulf in quality between the favourites and the underdogs is growing and it is getting harder for underdogs to bite back.

Indeed, whilst it is a very small sample size, the 2012/13 season has seen less shocks even than the trend suggests:

For clarity, the following is the list of games from this season that have had shock results:

Clearly QPR winning at Chelsea was a massive shock (regardless of the odds) and even Norwich winning at home against Arsenal & Man Utd offered odds of over 6/1.

Only 14 (15%) of the 93  shock results in the last 4 seasons have been HUGE shocks, offering odds of over 10/1.  The biggest shock, according to the bookies was Blackburn beating Man Utd 3-2 at Old Trafford on New Years Eve 2011, with Bet365 offering odds of 26/1!!!!!

We have so far established that shocks happen, on average, once every 17 games and HUGE shocks only happen in 1 in 108 games.  So which team are best at pulling off shocks?

Wigan have recorded the highest total number of shock results over the last 4 seasons with 11, Wolves are second on 8 and Villa & Sunderland are 3rd with 7.  I’m not sure whether this can be worn as a badge of honour or not, is the fact that you are regarded as an underdog a good thing or not?  Should Arsenal & Man City be embarrassed that they appear on the chart at all?  Arsenals 1 shock result was the 5-3 win away at Chelsea in 2011 and Man Citys was the 4-2 win in 2010, also away at Chelsea.

Obviously the above chart is weighted in favour of the teams that have spent the most amount of time in the Premier League, so by dividing the number of shocks by the number of seasons in the league we’ll get a better picture of who has the best record.

Blackpool therefore have the best record of shocks per season, with 5 in their only season in the Premier League in the last 4 seasons.  Wigan still fare reasonably well based on this criteria, coming in 2nd with 2.75 shock results per season.

Finally I wanted to have a look at which teams are more likely to be on the wrong end of a shock result.

Of the 93 shock results in the last 4 seasons, 18% (17 in total) have occurred against Chelsea.

If I had a time machine and could go back and have a £10 bet on each of the shock results this season, i’d have a profit of £1176.  If my time machine allowed me to go back to each of the past 4 seasons I would have made £6728.50.

So, does anyone have a time machine I can borrow?