So I like stats. I’m an unabashed stats lover. Whenever I can I write using stats, about stats or using stats in my articles. One of the benefits of being in an LFC community is that there are people who are smarter than me on Twitter who occasionally post interesting stats. This is one of those occasions, and one of those articles.
So AI’s own Dan Kennett recently did this thread highlighting goalkeeper error rates:
Thread looking at defensive errors and penalties conceded in the #EPL 2014/15 to 2017/18
— Dan Kennett (@DanKennett) July 12, 2018
My reaction to this thread was naturally to write an article about it. So let’s talk Liverpool goalies: Karius, Mignolet, Klopp, possible replacements and, because why not, Danny Ward.
So on average, as Kennett states, a goalie makes an error every 800 minutes, which is just less than nine games – so a keeper can be expected to make about four errors per Premier League season. Liverpool though is a top club, so would expect less than that from their top Premier League goalkeeper.
Since joining the club, Karius has made four errors, two of which led to a goal and two were penalties conceded. This is over 2610 minutes, so an error every 653 minutes, making Karius the 11th worst keeper over that period. That being said, 2610 minutes is a lot smaller as far as sample sizes go, compared to for example Simon Mignolet, who over the period covered – since the start of the 14/15 season, has played 10514 minutes. Mignolet has made 15 errors over that period, including eight leading to goals and one penalty conceded. That’s a ratio of one every 701 minutes.
There are obviously other things to factor in when considering goalkeepers. With Mignolet especially, it’s not necessarily statistical errors, so much as a less than superior shot-stopping ability. Errors don’t always encompass things such as being beaten at your near post, being caught out of position, not coming for crosses, etc.
Nevertheless, these numbers are less than pretty reading for a Liverpool fan. Seeing David De Gea at the other end of the list, in fourth, conceding just three errors leading to goals and not a single penalty in nearly 13,000 worth of minutes is pretty damning.
One number I found particularly interesting was the penalties situation because it’s a problem that Liverpool fans have addressed with regards to Karius before. It’s one that we’ve seen in the back end of this season. He tends to charge off his line when exposed in one-on-one situations, and against both Roma and Crystal Palace in the run-in, we saw Karius completely get it wrong, as well as against Spurs.
This is one of the features of a team playing a high line, something that will worry Klopp as it’s something Karius needs to get better at dealing with. Mignolet, for all of his errors, has only given away one penalty since the start of 14/15.
The next factor to take into account is how Liverpool’s style affects these numbers. Manchester United or Burnley play with high volume lower percentage shots and less pressing systems tend to take the pressure off of their goalkeepers, and therefore will have lower error counts. As Kennett notes, one of the issues with LFC is that their goalkeeper errors tend to be high anyway, albeit less so under Klopp. It has to be said that given the number of poor goalkeepers Liverpool have had over that period, Mignolet especially, it’s not necessarily fair to blame the system for Karius’ mistakes, but it’s certainly got to be taken into account as a mitigating factor.
It also cannot help when you look at the names near the top of the defensive errors list – with Dejan Lovren and Martin Skrtel both near the top of that list too. Having someone like Lovren in front of you cannot help your confidence and pressure as a goalkeeper, although that goes the other way too and having Mignolet behind Lovren works the other way – creating a vicious cycle.
So is Danny Ward the man to break the trend of Liverpool’s error-prone goalkeepers? Well, no. Not really. A quick glance at his Opta defensive errors for the 16/17 where he played 4008 minutes for Huddersfield indicates that he made six defensive errors – which at 668 minutes per error would put him slap bang between Mignolet and Karius.
So what about any potential replacements then? Nick Pope is a man who stands out in just about every conceivable metric, but as noted, Burnley’s system is extremely different to LFC’s and might not suit him at all, as evidenced by Heaton also being in the top three. Matt Ryan and Kasper Schmeichel are both names that have been put out there but neither have overall impressive metrics.
Alisson on the other hand, a name mentioned a lot with regards to LFC, has made just five errors, one more than Karius, in his 8000 career minutes, as this data indicates:
This was the data perspective for the GK shortlist on the AI Committee podcast #Karius #LFC pic.twitter.com/nzD7esYFwL
— AIUnderPressure (@AIUnderPressure) July 10, 2018
Whilst errors directly leading to goals are only one aspect of a goalkeeper’s job, it cannot in any way shape or form be a good thing that Karius’ numbers are not particularly inspiring. I would definitely not say that this data indicates we should sell or scrap Karius for a multitude of reasons – the sample size, the fact both of his errors leading to goals were at the start of last season, the team effect, the Lovren effect, the fact that some top keepers are on that list. However, when you have question marks over a goalkeeper, it makes sense to look at more data, and this data corroborates the idea that even before Kiev, Karius had far from had an error-free season.
When looking at replacing your goalkeeper, there are lots of factors to take into consideration, but ideally, at a top club like Liverpool, you want your goalie to be putting in positive metrics, not having to try and justify or explain away poor ones.