Last season, using data obtained from WhoScored, I wrote an article which looked at whether or not Brendan Rodgers was tactically flexible in 2012/13; the evidence suggested he was, and you can read the piece here.
The aforementioned stats website has the formation data, including results, dating back to 2009/10, so I have now compiled Liverpool’s statistics for the previous four seasons to see which set up has provided them with the best results. The answer may surprise you.
Before we go on, clearly this analysis is not to be taken too seriously thanks to the host of caveats that have to be applied. I don’t know which teams the Reds faced in each batch of matches, how accurate the formation data is, which players were available or injured (or indeed suspended, eh Luis?), and I assume that any in-game tactical switches are not accounted for either
Surely some of the variations in formation are purely in the eye of the beholder too; is a 4-5-1 much different from a 4-2-3-1, or a 4-4-2 from a 4-4-1-1? Clearly there are subtle differences between the similar formation pairs, but two people may differ on which one was being employed in any one game.
The first interesting thing to notice is the risk and rewards that can occur with tactical experimentation, as the following table shows:
As it resulted in a 5-0 win, why not try 4-1-2-1-2 more often?! Equally though, it’s perhaps clear why 3-5-2 (in 2010/11) and 5-4-1 (the infamous beachball defeat at Sunderland, which was four years ago today strangely enough) have not been seen more than once, as they didn’t work when they were tried.
Three or five at the back has only been used seven out of a possible 152 times in the last four seasons, and saw an average of just 0.03 points per game earned more than with the standard four at the back. Two wins from two with a back three this season may see the system used more frequently in future though of course.
There is more of a difference when looking at formations with one, two or three at the front:
Having two forwards has seen the most goals scored per game, and indeed Liverpool have scored three in both games that they have played with two forwards this season, though of course the fact that the two are Daniel Sturridge and Luis Suárez, and the opposition Sunderland and Crystal Palace has obviously helped there.
But the most fascinating, and perhaps surprising finding is which formation has registered the best points-per-game tally. I have limited the below table to those formations which have been used at least twenty times, with the remaining twenty-three matches bundled up together.
Well, well, well. For all that it is associated with outdated football, and the unsuccessful reigns of Roy Hodgson and Kenny Dalglish, 4-4-2 has yielded the highest points and goal difference return per game of the four most commonly used set-ups (though also a higher loss percentage than 4-2-3-1 too; for whatever reason, Liverpool haven’t had many draws using 4-4-2).
Let’s take a closer look at the 4-4-2 record:
In fact, of the 26 formation/season combinations in the study, Benitez’ 4-4-2 in 2009/10 has the highest points-per-game return of any formation used more than once. Based on their own scoring system, this is the 4-4-2 that WhoScored believe Benitez should have employed that season:
It will be interesting to see if Brendan Rodgers ever plays with a 4-4-2 at some point, now that he has his ‘SAS’ partnership working well in tandem. The evidence here suggests it works well for the Reds, so maybe he should.
Recent and related posts you might like:
Tipping Point? – If Liverpool defeat Newcastle, then Rodgers might have the plane high enough…
Second Half Slump? Controlling The Result – Although Liverpool have had some poor second halves in 2013/14, they’ve had better control than you might think…
Reds Bounce Back – Brendan Rodgers has an impressive record when following a league defeat.
LFC Pass Combination Heatmaps 2013/14 – A look at which players have been most involved pass-wise, and who they’ve linked up with in every league match this season.