Evaluating Mohamed Salah
When you call a transfer the “Signing of the Summer” and then the player proceeds to score 32 league goals and 10 Champions League goals in a season, it’s fair to say you can be happy with your work. Hell, I even mocked up a picture of Mo Salah in a Liverpool tracksuit for the accompanying artwork. I was high on this guy and Salah chipped in on his end of the deal. Good lad.
This season had been progressing very well too, at least until recently. He has recorded 20 goals in the league and Champions League (albeit with four penalties), it’s just that while last season saw an overwhelming overperformance–StatsBomb model had him at 31 non-penalty goals from about 20 expected– this season he’s basically matched expectation. As such, there is a view in the darkest corners of the fan world that perhaps he’s been a bit disappointing. In general this would be wrong. But there is no doubt that specifically recently, his contribution to scoring both in expected terms and in reality has somewhat slowed down. He has only one goal and one assist in his last ten club matches, and he’s been out there: he has only been substituted once. As I discussed last week, in the aggregate, all this might not matter, since Liverpool’s dynamo front three are scoring at a rate commensurate with what came before, it’s just Sadio Mané has received the goodwill of the footballing gods this season, and his goalscoring has matched that of Salah. Two forwards, 20 goals each.
Let’s have a look at Salah from a data perspective and see if we can identify any aspects of his game that have changed.
Here are his Premier League shot maps year to year (prior to the Fulham game):
To get one thing out of the way; we’re not being ignorant to the slightly different position he’s played at times this season, more central. The general shape of his shot map hasn’t really changed, there’s still a skew to his most usual right sided starting position. I’ve highlighted some segments that I think are of note though, so let’s work through them:
1. Salah’s opposite side saw him finish at a fantastic rate last season, by my reckoning he converted nine goals from around 27 shots in that zone left of the penalty spot. So he had a decent volume there and finished at a very high rate. This season he simply hasn’t gained shots over on that side at anything like the same rate. He has a couple of goals in there, but despite being nominally more central at times, it hasn’t translated into shots or goals.
2. This zone fascinates me because it’s the zone I envisage Salah operating in; cutting in from the right, perhaps beating a defender or creating space for a shot, nicely faced up in front of the keeper. Another nine or so of his goals last season came from this area, whereas this season he’s barely even shooting in there. He has one goal from that zone but there’s a huge gap where last season shots existed, and this season they just don’t. If there is one specific aspect in which the league as a whole may have become somewhat accustomed to Salah’s shooting tendencies, this might be it, the Robben zone. Whatever you do, don’t let him onto his left foot in that area, if you can help it.
3. This feeds into a smaller factor which I think is borne out in zone three. If Salah is diverted away from that golden shooting spot, where does he go? That little cluster of shots derived from throughballs in there feels like the resulting aspect of Salah having had his shooting angle closed off, before ending up with a tight finish with his right foot further in. Even right footers struggle to finish from these kind of zones, so for Salah to struggle there with his off foot, would be quite normal.
Slowing down is normal, and was always likely, but we have a window there into where shooting locations are actually different.
Here’s an simple experimental viz to identify where players move to generate their own shots. It shows the position in which they received the ball (start of the line) and the resulting shot location as well as a footed indicator for the shot. It’s important to recognise that the actual movement of the player may or may not be in a straight line (as represented), so what we’re mainly interested in are the starting and finishing locations. There’s a cut off of 3 metres so as not to overload the visualisation and also to highlight events in which the player has made a significant move, be it a dribble or a carry, between receiving or recovering the ball and shooting it.
Now some of this will be in the eye of the beholder, but it looks to me that Salah was more proficient at generating shots from his right sided starting position, these kind of 5 to 10 yard snappy bursts in a North West direction during 2017-18, while this season, has tended to move inside or take it down towards the touchline. This chimes with what we saw earlier regarding overall location of shots too. He’s perhaps starting his shooting moves in that back quarter of the penalty box less frequently too.
Let’s try and segment the season to see what’s changing elsewhere:
Usefully Salah has appeared in every one of Liverpool’s games this season, so we can slice up his season into tidy ten game segments (data is from prior to the Fulham game, he has one substitute appearance in there too). Ten games is probably about as small as you might want to slice this for analysis purposes, and the schedule is fairly balanced through here too (3/3/2 games vs top 6 rivals) but we can see significant differences.
First ten games: High shot rate, over four per game (same as last season) feeding into high xG (nearly 0.6 per 90) and solid but unspectacular xG assisted (0.26 per 90).
Next ten games: Shot rate has come down by 1.5 per game to 2.6, and xG has followed (0.37) however he’s taken some slack in creativity (0.40 xG Assisted per 90)
Last ten games: Tough times. Shots moved closer to three per game, but xG is down again (0.31 per 90) powered by a big drop off in xG per shot (0.14 in both prior segments, 0.10 in this one) and the xG assisted has gone AWOL (0.14 per 90).
Segmented shot maps bring out that xG per shot decline clearly:
First ten games: few too many blocked shots from range, nice xG boost from a bunch of shots deep in the six yard box.
Next ten games: where did all the blocks go? Really quite a nice mix of shots, then that odd little cluster in a bit of a no man’s zone outside the edge of the six yard box.
Last ten games: I don’t like all these wide shots from range. It’s hard to score from out there, even with an open look at goal. Not saying it’s easy to achieve but Salah somehow needs to work to shift these 5 yards closer to the centre. Look at the lack of shots closer in than the penalty spot and within the width of the six yard box.
It’s worth reframing this whole discussion around expectation too. Without last season, this season would look like an excellent return for a new wide forward. Mané is another player who has had stick at times, entirely unfairly when factored against his general contribution as a non-traditional forward rather than a typical goal-getter. And the team itself is having one of the great seasons of the Premier League era, yet amazingly remain in a title race. I had a look at Salah’s passing high up the pitch, but could see little difference season to season in how he interacts with those around him, or how he progresses the ball. He will always turn the ball over quite frequently, it’s a trait related to his tendency to try and carve out his own chances through ball carrying, but he is a good passer in dangerous areas (ninth in the league for open play passes into the box).
Salah is going through his first major quiet period as a Liverpool player. It’s to his credit that it’s taken the best part of two seasons before he has reached this point. One concern could be around tiredness. He played a full season last year then got injured in the Champions League final and rushed back to the World Cup then has been nearly ever present again in 2018-19. The recent drift towards wider and rangier shots is something to keep abreast of moving forward but luckily, we well know that Liverpool have some of the smartest guys around working on analysis, so one would presume it’s in hand. And spending time with the video will shed further light on the story too.
It’s tough for Mo. He had a season which lit up like Messi, and as we saw again this past weekend, nobody is a Messi but Messi. Salah will shin one in soon enough and people will forget about this blip. The battle between defenders desperately trying to shift him onto his right foot will rage on, no doubt for years to come. How well he adapts to that will be fascinating to follow.