Race For The Top Four: An Analysis

I recently appeared on an Anfield Index Analytics podcast (which you can listen to here) in which host Dan Kennett and I ran through the pros and cons of the five contenders who are aiming for a third or fourth placed finish in the Premier League this season. As there is only one third of 2014/15 now remaining, I thought I'd share the stats and my thoughts here. I will be looking at how the five teams fare in three stats: big chances, shots on target, and defensive errors, and look at how they compare to the rest of the league, but also how they have fared in the mini league comprising these teams. Needless to say, I will be paying particularly close attention to Liverpool. Before we go on, a quick look at the current standings and results in the mini-league I will be referring to.

We can see that whilst Manchester United and Liverpool have the most games left to play in this group, maybe that's no bad thing as they also have the best records here too, so they will perhaps view it as an opportunity to strike a blow to a direct rival rather than fearing a match against a tough opponent. Let's start our look through the stats with big chances which, lest we forget, are defined as “situations where a player should reasonably be expected to score”.

We can see that three of the contenders here are within the Premier League top four for this stat, which passes the common sense test, and also that the two sides that aren't are the pair currently outside of the top five in the actual league table. Liverpool's position of seventh in the top flight for this stat may seem underwhelming, but as recently as December they were in the bottom three, so they have been making great strides in recent times. In mini-league terms, whilst United have the best big chance difference, it's Liverpool who perhaps have the most impressive record here, as essentially one third of their big chances this season have come in just five matches against their rivals in the above table, which is quite remarkable. How have the teams done with regards to converting and saving big chances? The conversion figures in attack are for all attempts, but in defence I have only used on target efforts to measure the impact of the goalkeepers.

Liverpool are a mixture of the sublime and the ridiculous; only Chelsea have converted a higher proportion of their top quality chances, but only four teams have saved a lower percentage of the opposition's big chances. Meanwhile, David de Gea has saved 60% of these chances compared to a league average of 37%, and is a massive reason as to why United are currently third in the table. Southampton can also look to their 'keeper (Fraser Forster) for inspiration ahead of their strikers, as the above figures demonstrate. A lot of big chances come about through (Opta-defined) defensive errors, so let's see how the teams compare on this front.

Southampton's record is simply stunning here. No team in the league has made fewer errors (and as I noted on the podcast, Dejan Lovren has made more in the penalty box on his own than the Saints have this season), and up front only Liverpool have forced more errors out of their opponents. Although Liverpool have benefitted from more opposition errors than any other top flight team this season, they (along with Spurs) have only profited with seven goals whilst Southampton have eight. I noted in this piece about errors in 2013/14 that it seems to be teams that press high up the pitch that benefit most from opposition mistakes, and it would appear that this trend has continued this season. The Reds may have made the second most defensive errors in the Premier League this season, but the good news for them is that their current form is far better; Liverpool have only made five errors in their last thirteen matches (which shows how bad they were at the start of the season). Manchester United are the league's worst team for errors, and once again have David de Gea to thank for saving their bacon; they have conceded two from mistakes when even a league average conversion would've seen them gift six goals to their opponents. Finally, I'll take a look at the shots on target ratios. This stat is the percentage of the shots on target in a team's games that the team has, and I wrote at length on this for These Turbulent Times (a collection of writing from The Tomkins Times). What I found was that this is a highly predictive metric, which gives a strong indication of where a team should finish. A few teams will inevitably over- or under-perform, but the shots on target ratio (SOTR) aligns well to each league position on average:

(It's one season out of date, but you get the idea)

We can see that 60% is the benchmark for a top four finish, so how have the five teams done this season? They are sorted below by standard SOTR, but I have included the figures for location adjusted shots on target too, purely out of interest.

The figures broadly follow the pattern that I previously highlighted e.g. that 60% is a top four performance. Spurs have clearly outperformed their SOTR this season, and I think is probably due to their large number of late match winners, as they have been taking three points when they perhaps statistically didn't deserve them. Tottenham have also scored a sixth of the direct free-kick goals in the Premier League this season (three of the eighteen) which has no doubt helped too. From a Liverpool perspective, we can see that their SOTR has put them deservedly in the top four hunt, and they have been the best team in the mini-league; battering Spurs away and having countless on-target efforts at Old Trafford will have helped here. Before I conclude with who I think will finish in the top four, let's take a look at what some of the stats model websites think. Here's how EuroClubIndex think the table will look at the end of the season. The main difference from the current table is that Arsenal surge into the top three whilst Southampton drop out of the Champions League places:

Meanwhile Sports Club Stats only alteration to the current table is for Southampton to finish ahead of United:

If you've been paying attention to the above tables of stats, or if you've listened to the podcast, then you probably know who I think will finish third and fourth: Arsenal and Southampton. These two teams are at or near to the top of all of the three stats rankings, and this is why I think they'll prevail. My money is on Arsenal to finish third, mainly as they've been so good with their big chances whereas Southampton (whilst very strong in defence) look short of the finishing prowess needed for a top three finish. I would then expect Liverpool to finish fifth, as their form is excellent at the moment. You can see that in their underlying stats which are very good now when for half of the campaign to date they were dreadful. The Reds are powering along nicely, but will likely fall short as it will be very hard to get past three of the four teams (at least) and they probably have the hardest run-in too. I then think it'll be Manchester United (as they are doing well with big chances) in sixth and Spurs in seventh (as their stats are the weakest of the five). Of course, the stats can't take account of the impact of lengthy cup runs (which Arsenal, Liverpool and Spurs may yet still have to deal with) or injuries and suspensions to key players, and those difficulties could certainly play a part here. But those problems notwithstanding, my money would be on Arsenal and Southampton. Please check out my other articles, and follow me on Twitter or via Facebook. Thanks.