I love stats. Total number of... shots on target, passes completed, crosses attempted, blocks put in and so much more... I love it. To the untrained eye, football does indeed look like it’s just 22 people kicking a leather ball around a grass pitch for 90 minutes or so, but the sheer number of stats out there shows how it’s so much more than that.
Although stats can’t properly predict what will come to pass (pun intended) in a football match (consider me an xG sceptic), they’re great for analysing what’s already happened. Sure, you can’t explain everything that takes place in a match, but you can - if you dig deep enough - usually zero in on a root cause for most things.
That brings us to Reading - specifically Jaap Stam’s Reading - which has had an interesting relationship with numerical analysis over the last 18 months or so. Bizarrely, the stats gurus couldn’t find much data to explain The Royals’ third place finish last season - the previously-mentioned xG (expected goals) metric conclusive in its assessment that Reading were not the third best team in the Championship last season.
Given our underwhelming campaign so far, I’ve seen it suggested a few times that - broadly speaking - Reading got lucky in 2016/17 and our fortunes are evening out in 2017/18. That’s not something I agree with, but it still got me thinking:
Statistically, how different are Reading to last season?
To answer that, we need to head on over to Squawka.com (the relevant data is more readily available there than it is on WhoScored.com) - so almost all the numbers you’ll see quoted below come from Squawka. In short, the differences are so small that they’re barely noticeable.
Jaap Stam’s playing style is often cited as the main source of Reading’s woes at the moment, but the basic stats suggest it’s not that straighforward. Although we saw an average possession rate of 56% last season, that’s dropped just one point to 55% in 2017/18 - barely any change. Similarly, the pass completion rate has fallen from 81% to 79%, and the average distance of a pass from 22m to 21m.
Clearly, the style itself hasn’t changed despite Reading going from third one season to eighteenth now - just how well that style is being implemented on the pitch. Looking at the raw data above how well we keep the ball and move it around, there’s little to suggest that we’re doing anything differently.
The logical next step is to analyse how effective that style is in the final third but, although 2016/17’s performance was noticeably better here, the difference isn’t as huge as you might think. This season, Reading create 8.34 chances per game (19th in the league)- which Squawka defines as either a ‘key pass’ or an assist - side point, that’s the exact same figure as sixth-placed Sheffield United’s. So, is it a lot lower than last year? Nope - that comes out at exactly 8.5 (enough last year for 13th).
However, a minor decrease in shot accuracy (47% -> 42%) has apparently led to Reading being noticeably less lethal in the final third (1.4 goals per game has now become 1.07).
Personally, I think we can put this down to one pretty obvious difference between the two campaigns: the presence of a reliable goalscorer. Last year, Yann Kermorgant was a constantly dangerous presence in the box, taking 2.7 shots per game and hitting the target once per game. However, this season’s leading hitman Modou Barrow can manage numbers of only 1.8 and 0.7 - not bad for a winger, but nowhere near the level required of a main striker. As for Kermorgant, injury problems have made him a shadow of his former self, cutting his figures down to 1.6 and 0.1*.
Oh Nelson Oliveira, how we need you.
*Stats for this entire paragraph came from WhoScored.
Down the other end of the pitch, bizarrely, The Royals have actually got better. It was widely accepted last season that Reading were fairly awful in their own third at times - see Brentford, Fulham and Norwich City away as painful examples - so perhaps an improvement isn’t that surprising.
On the whole, Jaap Stam’s side let in 1.36 goals per game in 2016/17, but that’s now down to 1.26. An interesting side note is that, whilst we conceded three or more goals on nine occasions last campaign, we’re yet to do that since the 3-2 loss at Nottingham Forest in April.
Despite the stats, I’m loath to accept that Reading are a better side defensively - although it’s not really the point, you’d probably have a similar goals conceded record if you cancelled out the random thrashings at Griffin Park, Craven Cottage and Carrow Road. That said, we have demonstrated some surprisingly good defensive solidity at times this season - despite not winning in our last eight league matches, we’ve only conceded six goals in our seven games since Christmas Day.
The problem instead is, quite clearly, sticking the ball in the net, and there are two conclusions that I’m confident in sticking by - at least on the basis of the above stats analysis. First of all, the failure to bring in a new first team striker was evidently huge, and I’m starting to think that it could be the single biggest factor in the embarrassment that’s been 2016/17. How much better would we have been with Nelson Oliveira in the side?
Secondly, confidence. Considering that most of the prior stats are so similar, the psychological difference between last season and this must be huge. After all, in 2016/17 the team was able to turn a very similar number of chances into a much bigger number of goals - that can only happen with more belief and desire.
I’ll leave you with one question to ponder: is all the above evidence that a) Jaap Stam’s style at Reading is fundamentally flawed, or b) the system is fine, but you need the right personnel.