A few weeks ago we entered into the debate surrounding Real Goals vs Expected Goals and posted a list of players who had scored more than the average player would have in that situation as well as which players underscored over the last two seasons. This is a topic that has caused much discussion in the football statistics world with arguments for and against this kind of data being given.
Thinking out the box is something that we like to do as often as we can here at SportsMatrix so we decided to take the concept of Real Goals vs Expected Goals and used it to look at the same shots from a different angle – that of the goalkeeper!
How do you judge a goalkeeper? The first and most popular answer is always against goals conceded. Of course, the keeper of a big team, with a world class defence in front of him will, invariably have better stats than those of a keeper who plays for a weaker team when judged by this standard . This is one of the main reasons that the goalkeeper of the year is almost always from one of the big teams.
In order to get a more statistical sound view on who was the best goalkeeper really over the last two years, we examined all shots taken from the 2011/2012 and 2012/2013 seasons and how the various keepers across Europe handled them. The reasoning behind this was simple – a goalkeeper standing behind a world class defence will, generally speaking, face less dangerous shots as his defence would have dealt with the threat first. Thus it was important to separate the shots by a difficulty level. The various difficulty levels were determined on the basis of a number of different parameters, not to dissimilar to those used in the expected goals post. The parameters used here included the goalkeeper’s positioning, position of the player taking the shot and the power and accuracy of the shot, amongst others.
Once we had separated the saves into the various difficulty levels, we could see how many saves the average goalkeeper would make in each situation. The next step was to compare those stats with the actual stats by the many goalkeepers standing guard for their various teams. This comparison basically split the keepers into two categories – those who saved more shots and thus prevented more goals than the average keeper and those who conceded more goals than the average keeper would have when in the same situation.
Perhaps unsurprisingly, the best performing keeper of the last two seasons was Juventus’ Gianluigi Buffon who conceded 9.5 goals less than the average keeper would have when in the same situation. In third place, and proving our point that good keepers need to be judged on more than just goals conceded was Michael Agazzi of Cagliari in Italy who despite conceding almost three times the amount of Buffon, actually saved his team 8.3 goals that an average keeper would have conceded.
When looking further down the list, we saw some fascinating facts regarding ‘big goalkeepers’. A few eyebrows were raised when Liverpool manager sent Liverpool keeper and fan favourite, Pepe Reina away on loan after signing Simon Mignolet from Sunderland. Based on the stats, this was actually a masterstroke and could certainly be one of the reasons for Liverpool’s resurgence this season. Over the last two seasons, Reina has conceded 4.2 goals more than the average keeper whereas Mignolet actually saves 6.2 goals more than the average keeper. The bottom line here is that by changing keepers, Liverpool are 10 goals better off.
In order to allow you to come to your own conclusions, we have created a special infographic showcasing the stats of the top 3 keepers in this regard over the last 3 seasons, with another three keepers of interest!
Picture Courtesy of: