#8 Player Partnerships and Compatibility

Throughout the years there have been many partnerships between players that seem to stand out and are more memorable than others. The classic Yorke and Cole strike partnership for Man Utd’s treble winning season, Xavi and Iniesta passing rings round everyone for fun and Ferdinand and Vidic who became arguably the Premier League’s best defensive pairing. The main thing in common with these players is that they were successful, playing at the highest level of the game for years and so were the best of their eras. As they were successful, they must have been pretty good. Individually you need to be very good to make it at the top, however since football is a game that requires eleven players all working towards the same goal it’s how well you can fit into a system alongside other talented players which can make a good team a great team. If the aim of a team is to be greater than the sum of its parts, when that occurs very good things tend to happen, eg. Leicester City in 2015/16.

The problem arises at every football club of how to fit together all of your best, talented players into the same team. If you can get those players working well together then you will be working near the optimal level that your team can achieve, and that’s the aim. However, this frequently doesn’t turn out to be the case. You have two-star players who both play as strikers, however each prefers to lead the line on their own and so when you force them to play together, both of their performances drop and as a result the team’s performance drops. It feels like a lose-lose situation, either you drop one of your star players who will then be unhappy or you play them both and suffer bad performances. The ideal solution as suggested is to get them both playing together, however sometimes players are incompatible and so the next best solution is to work out which of those players is more compatible with your team and offload the other.

On a football pitch there are eleven players and so you want to get optimal performance from the whole team rather than just having your right full back and right winger link up well every now and then. Each player has a partnership to some degree with every other player on their team, those players that are in frequent contact and close proximity will usually have stronger partnerships since they interact more. These are pairings such as the two centre backs, two centre midfielders, full back and their respective winger and two strikers. This of course will depend on formations, if you are playing three central midfielders then it’s how well those three can play together which is important.

Figure 1 – Football Manager represents player partnerships with green lines

One way to measure the compatibility of all eleven players would be to compare the results and performances of each combination of eleven players on the pitch. So, one starting line-up would have a set of results, whilst if you just changed one player you assume that is a completely different starting line up and have a separate set of results for that group. It’s debatable how much of an impact changing one player could be, it depends how influential the player you replaced was. The problem with this is that there won’t be a large enough sample size since rarely do the exact same players play every week due to injury or rotating players due to tactics.

Another option could be to compare the results and performance of specific partnerships that you are interested in, such as when your central defenders are the same or when your strikers are the same. In looking at specific partnerships, it allows you to look into more specific areas of performance when assessing. For example, when comparing two sets of central midfielder partnerships, you may want to compare how much possession or passes you had in those matches whereas if you’re comparing two sets of striker partnerships you may want to compare how many goals were scored in those games.

As mentioned earlier, there aren’t just partnerships between those players in positions in close proximity but between players all across the pitch. Maybe a right full back has an understanding with the left winger and likes to make a long diagonal pass or maybe the winger likes to come inside and receive passes from the central defenders.

So far, we have looked at how to compare sets of partnerships with each other, using metrics such as goals or passes as proxies for compatibility since they are seen as productive outputs. Without realising, these are all on the ball metrics. There is nothing wrong with looking at those, however they need to be just part of the answer. Most of the game for every player is played off the ball so it’s arguably more important to assess their performance off the ball than on it.

When assessing partnerships off the ball performance, as I’ve discussed in my Defensive Metrics posts, it’s more about how you can be in the right place at the right time and what decisions you make. If it so happens that when you are playing with a certain player that collectively you are able to maintain correct distances, are capable of covering for mistakes and therefore force the opposition to make worse decisions as a result then you would be more compatible with that player than with another. The problem is that it’s hard to quantify these, I have attempted to outline a few ways in which you can start to get some insights in previous posts however it’s still very early. If we can effectively measure player performance of the ball and interactions between players that don’t involve the ball then we will get more of an insight into what makes some partnerships work so well and why some others don’t.

There are many examples of passing networks in football, based on specific matches and representing the distribution of passes between each player. They provide a certain aspect of the partnership between players, it would be interesting to see other examples of networks in football with connecting lines that represent how compatible the two players are. How exactly we can measure that is up for debate.

@TLMAnalytics

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.