written by – Rangers Report
One way to look at statistics is that they are objective facts. Kenny Miller scored eleven goals in the Scottish Premiership last season & averaged 0.36 goals per 90 minutes. That’s a fact. Bruno Alves won 70% of his tackles last season for Cagliari. Fact.
Common sense suggests that managers make decisions based on facts that are aligned with objectives they want to achieve. Of course, hunches will always play a part in decision making – but given the short shelf life of football management…maybe they should lean more towards facts then hunches.
On of the most respected writers in the NHL analytics world, Ryan Stimson, recently published his Jerry Maguire style manifesto on how to use analytics to influence the way teams play & how they build their teams.
In The Stimson System, he wrote – “The traits of successful teams and players in addition to new, proven metrics should be analyzed alongside each other to discover what can be done from a tactical viewpoint to optimize a team’s play.”
“In a field that’s ever-increasing in terms of competition, the advantages found in the margins are often the difference between winning and losing.”
When managers are developing tactics for an upcoming opponent, their ultimate goal is to find that advantage that will hopefully lead to a successful outcome. Stimson contends that statistical analysis can help identify the tactical approach that will give you the best chance of creating that advantage.
“How can we ensure the way we want to play will lead to the larger share of shots and expected goals?”
This isn’t rocket science…look at the facts & plan accordingly.
One example of this on a micro-level would be using Possession Efficiency Stats to identify which opposing players to aggressively target & exploit in a pressing system. (I explained & applied Possession Efficiency Stats in a write-up of last week’s Champions League Final).
It’s a very straight forward & easily digestible measurement. At what rate do players lose possession when they have the ball?
For the purposes of this post, I decided to look at West Bromwich Albion’s final four matches of the EPL season to highlight how to apply this approach.
Why West Brom? One, the stats are available for free online (unlike the SPFL), & two, in an ongoing project in which I’m developing a statistic to rate center backs — I’ve concluded that West Brom’s center backs are not very good. West Brom may be the type of team that a pressing system would be really effective againsty.
The following results come from West Brom’s final four matches against Swansea, Chelsea, Manchester City, & Burnley. This sample works out quite nicely given that they played two very good teams & two bad teams. Stats are courtesy of FourFour Two’s StatsZone.
Below you’ll find the total possessions for each player & their rate of losing possession as a result of being on the ball.
Hmmm….the four players who are on the ball the most are midfielders & attacking players & all sustain possession about 75% of the time. Maybe a visual representation of these stats would help identify players to exploit?
This helps a little more. Craig Dawson & Marc Wilson, who are both defenders, are on the ball a lot & also have a tendency to lose possession at a high rate. James McClean, a winger who featured mainly as a sub in these four matches, is also a player who isn’t very efficient while on the ball.
West Brom, who alternated between a backline of four players & three center backs in the games tracked, had two defenders who lose the ball a third of the time & another, Jonny Evans, who only lost possession 15% of the time.
The problem with this data is that we don’t know where these possessions are occurring. Passes are easier to complete the further away from goal you are.
In order to truly determine which players could be exploited for their lack of efficiency while in possession, we need to see their Possession Efficiency Stats in the different zones on the pitch. For example, if a player tends to give the ball away in the defensive third – there should be an automatic green light to press these players at every opportunity.
Here are West Brom’s Possession Efficiency Stats that only take into account possessions in the defensive third:
You can see that if Marc Wilson’s name is in the first eleven, then he is a player to aggressively target with a press. To a lesser extent, if Darren Fletcher has dropped back into support than he is also a player who may be susceptible to pressure.
That’s not to say you should only press these two players, rather it should be an indicator to players who are supporting the press to be more aggressive in order to pounce on any turnovers created. Conversely, the player who had the most possessions in the final third, Jonny Evans, never lost possession in the four games tracked. He should still be pressed, but those supporting players should be much more passive because Evans isn’t likely to make a bad decision on the ball.
Now let’s move further up the pitch to midfield. Loss of possession here can really dictate a team’s transition play or force the opposition into aimless long balls up the pitch.
- If Marc Wilson is on the ball….press, press, press!!!
- McClean & Brunt are also players who can be exploited. Given that they are usually out wide, they have less options while on the ball then a player who is more central. Players who are pressing can basically use the touchline as an extra defender.
- Craig Dawson, who is relatively effective on the ball in his own third, becomes more of a liability to exploit as his moves up the pitch.
- Of your central midfielders, Fletcher is the least efficient on the ball & the good news for opposing teams is that he is on the ball a lot.
- If Claudio Yacob is on the ball, the press needs to be more conservative given how efficient he is.
For all of the Possession Efficiency Stats from this four match study — click here.
Now that we have identified which players to aggressively press, I figured this would be time to share another system that I am stealing from the NHL. This is kind of my modus operandi.
The most common forecheck/pressing system in hockey is a 2-1-2 forecheck.
In this system:
The first forward in (F1) intensely pressures the puck carrier & is looking to take away his strong side play
The second forward (F2) adds pressure to where he thinks the puck carrier will look to pass (usually his defensive partner)
The F3 offers support defending the puck carrier’s strong side play & looks to cut off where the puck carrier may want to skate. He can also function as an outlet if the F2 steals the puck
The other two players are spaced out in support, looking to steal an aimless pass along the boards or to pressure the puck carrier if he manages to skate past the pressuring forwards
Any press begins with an opposing player on the ball. Let’s pretend it’s Marc Wilson…because the stats show we have a very good chance of forcing him into a loss of possession.
With Wilson on the ball, you can see who the 2-1-2 players are. It is imperative that these players play as a five man unit, while the other five players understand their role now that a 2-1-2 system has been initiated.
The 2-1-2 press has begun. Like in hockey, the first player pressures the player on the ball, while the second pressing player begins to close down the best passing option. The third player begins to support the press by closing down the space where the ball carrier may go if he gets by the first layer of pressure.
Those back two players are reading what happens & identifying how to support the play, either in transition or to fill space left by the pressing players.
The ultimate objective is to obviously win possession here; either by creating a turnover & a quick transition…or to force the player on the ball into a long ball up the pitch.
Coming full circle back to the Stimson System….stats could also dictate the kind of players a team should recruit in order to run a pressing system like this.
When looking to identify midfielders, the player’s tackles won ratio should be considered. If he is going to be pressuring the player on the ball, you want a player who could win that tackle if given the chance. Other stats to seriously incorporate would be interceptions & recoveries. A player’s ability to anticipate where the play will develop is a key to accumulating a significant number of interceptions &/or recoveries.
I’m sure there are other entry points to analyze if a system like this is implemented & that is ultimately the point. Tactical systems & statistical analysis should be connected in order to maximize a team’s chance at success.