By David Butler
There is a debate in Ireland about who should be our national football (soccer) teams number one goalkeeper. Luckily, we have two young Premier League goalkeepers to call on. Gavin Bazunu plays regularly for relegation candidates Southampton while Caoimhin Kelleher is the reserve goalkeeper for Liverpool.
The discussion is that Bazunu is receiving game time and gaining valuable experience but he is shipping a lot of goals. Kelleher is not playing – we do not know how many goals he would have conceded. Importantly, Bazunu is not solely responsible for every goal conceded – the entire Southampton defence should shoulder the blame. Ireland’s last famous goalkeeper knew this all too well; Shay Given was known as an amazing ‘shot-stopper’ (although that is his job!), largely because he had plenty of practice with a weak Newcastle defence in front of him.
It is hard to disentangle individual from team performance, but more and more we are getting access to statistics that allow us to do this.
The ‘post shot expected goals’ stat (PSxG) quantifies a goalkeeper’s ability at making saves and is expected goals based on how likely the goalkeeper is to save the shot. This is calculated by OPTA and other football stats providers. The xG includes penalty kicks, but does not include penalty shootouts. In a nutshell, this stat takes into account the quality of the shot, the distance from the goal, angle, type of shot (head, foot, etc), the position of the defenders and goalkeeper, and other variables. The stats companies then calculate the probability of a goal being scored after a shot has been taken.
By subtracting goals conceded from the PSxG, we can determine whether a goalkeeper is overperforming or underperforming compared to their expected performance. Positive numbers indicate better performance and maybe luck. The evidence would suggest that Bazunu is not doing well.
The table below is ranked from smallest to largest by the difference in goals to PSxG for the current Premier League season. I removed the goalkeepers who only played a handful of matches. Unfortunately, Bazunu is bottom of the list. He has conceded about 15 more goals than he 'should have'. The Liverpool number 1 goalkeeper – Alisson - is performing best. Caoimhin Kelleher may have to warm the bench for some time yet. We don't have an observation for Kelleher, so I guess the debate will rumble on.
For more on the use of goalkeeper performance stats and how they correlate to pay see a new paper in the European Journal of Sports Management here.
There is a debate in Ireland about who should be our national football (soccer) teams number one goalkeeper. Luckily, we have two young Premier League goalkeepers to call on. Gavin Bazunu plays regularly for relegation candidates Southampton while Caoimhin Kelleher is the reserve goalkeeper for Liverpool.
The discussion is that Bazunu is receiving game time and gaining valuable experience but he is shipping a lot of goals. Kelleher is not playing – we do not know how many goals he would have conceded. Importantly, Bazunu is not solely responsible for every goal conceded – the entire Southampton defence should shoulder the blame. Ireland’s last famous goalkeeper knew this all too well; Shay Given was known as an amazing ‘shot-stopper’ (although that is his job!), largely because he had plenty of practice with a weak Newcastle defence in front of him.
It is hard to disentangle individual from team performance, but more and more we are getting access to statistics that allow us to do this.
The ‘post shot expected goals’ stat (PSxG) quantifies a goalkeeper’s ability at making saves and is expected goals based on how likely the goalkeeper is to save the shot. This is calculated by OPTA and other football stats providers. The xG includes penalty kicks, but does not include penalty shootouts. In a nutshell, this stat takes into account the quality of the shot, the distance from the goal, angle, type of shot (head, foot, etc), the position of the defenders and goalkeeper, and other variables. The stats companies then calculate the probability of a goal being scored after a shot has been taken.
By subtracting goals conceded from the PSxG, we can determine whether a goalkeeper is overperforming or underperforming compared to their expected performance. Positive numbers indicate better performance and maybe luck. The evidence would suggest that Bazunu is not doing well.
The table below is ranked from smallest to largest by the difference in goals to PSxG for the current Premier League season. I removed the goalkeepers who only played a handful of matches. Unfortunately, Bazunu is bottom of the list. He has conceded about 15 more goals than he 'should have'. The Liverpool number 1 goalkeeper – Alisson - is performing best. Caoimhin Kelleher may have to warm the bench for some time yet. We don't have an observation for Kelleher, so I guess the debate will rumble on.
For more on the use of goalkeeper performance stats and how they correlate to pay see a new paper in the European Journal of Sports Management here.
RSS Feed
