Tactics in MLS (And/Or The Supplemental Nature Of OptaStats)


Logo from optasports.com

I tend to consider tactics and Major League Soccer like an unabridged dictionary in a fourth grade classroom. While the study and employment of tactics within MLS happens, there is less tactical sophistication in MLS than in the more established leagues of the world.

One thing is clear about tactics: the biggest key to their successful use is that an advantage or disadvantage must be identified by a manager in order for that to be exploited or minimized. In my opinion, the talent level in MLS is relatively flat. There aren’t many “game changers” in MLS. 10 years ago Thierry Henry was a game changer in any league. At this point, he is a game changer in MLS. Those players can be difficult to find (and/or afford) in MLS. Across the board, players are pretty evenly matched. There are definitely some differences, but they tend to be subtle. That’s one reason why, in my opinion, MLS tends to have a higher prevalence of results that are dictated by chance (set piece, penalty call, outright mistake) rather than upon strategy.

Having said all of that, it is changing. Most of the clubs still play a version of the 4-4-2 or 4-4-2 diamond. Formation is only one small piece of the tactical puzzle though. Everything from direction of attack, to tempo, to players playing out of position can factor into the strategic decisions of a manager on the field. As better players and more sophisticated managers enter the picture, I think we’ll find meaningful tactical shifts more prevalent.

So to Opta. The statistics information provide by Opta cannot alone allow you to discover a team’s strategy. It’s very difficult to sit down after a match, grab the Chalkboard, and go. Opta’s information is best used as a supplement. With all of the observations you make, you can formulate your hypotheses for how a team coped or dominated. At this point, Opta becomes very helpful. You can then take your hard work and scrutinize it against the hard stats, and find out whether they jibe with your argument.

Here would be an example. I know you are probably sick of analysis of the Philadelphia/Chivas match this week, but as I pointed out there was an interesting substitution at halftime. Jordan Harvey, the team’s starting left fullback, was removed for Keon Daniel, who is a left midfielder by trade. Why do I feel that Nowak made this substitution? My initial hypothesis was that it was because Daniel, being a midfielder, would be more likely to get forward with better pace and width. I was able to speak with Keon, and all he indicated was that he believes Nowak hasn’t put him in training to play that position. Let’s examine this in these Opta chalkboards, analyzing approximately the final third for the entire timeframe of the match (this works because they were exchanged):

Harvey's Final Third

Daniel's Final Third

It may be difficult to read in this picture, but the difference between Harvey and Daniel in the final third is quite interesting. Harvey played more passes in the final third in his 45 minutes, and they played the same number of crosses. So the question begs: does this ruin my hypothesis?

The complicating factor is the score. The Union were behind for 20+ minutes while Harvey was on, while they were ahead for a similar amount of time while Daniel was on the pitch. If you look further at Daniel’s numbers, you see his runs forward subsided slightly after the 68th minute, when Ruiz broke the tie. The number of Harvey’s runs were similar before and after the opening Chivas goal at the 28th minute, but given that there was less time after the goal than before, the rate would be higher upon falling behind.

This is hand-waving of course, and to the best of my ability I’d say that my argument is plausible, that Nowak had expected Daniel to give more width, forward push, and linkup with LeToux on the left side. Of course in football you can never predict if and when your efforts will prove fruitful, and in this case a Veljko Paunovic goal three minutes after the break changed the urgency for the fullbacks to surge.

To reiterate my point, I don’t believe you can independently use OptaStats to dig into a manager’s brain without a knowledge of the match itself. I’m sure there are cases where it’s possible, but it is a great tool for verifying and supplementing your visual analysis of the match. If you have your own analysis you’d like to provide for MLS matches on this site, feel free to contact the Gaffer or myself, or leave a comment below.

15 Responses to Tactics in MLS (And/Or The Supplemental Nature Of OptaStats)

  1. TFC Blogger says:

    Great break down! I love talking formations!

  2. I think the McCarty DeRo trade is a prime example of strategy and tactics in MLS. DC United was working well with Fred and Simms, Najar and Pontius on the wings and Davies/Brettscneider/WOllf up top

    Fred leaves and they get rid of McCarty for DeRo. They did get McDonald for the backline but they let a central midfielder go for a goal scorer. I have never seen DeRo as a midfield general and that is what DC needs.
    With Pontius, Davies/Ngwenya,Wolff/Brettscneider,and Najar around the center, you dont need a goal scorer , you need a midfield general, with two others.
    lastly a 33 year old, for a 25 year old, I feel like if they would have kept mccarty and put in mcdonald, with jakovic coming back at some point, they would have had enough. where is quaranta? Patience is a hidden trait with successful teams too, the MLS clubs tend to have too little of it

    MLS’s problem is personnel decisions based on ambitions and not the needs of the team at the moment.


    • nicc says:

      we liked McCarty in DC but it was becoming plain to see that it wasnt really working. we needed someone with a winning attitude that could distribute the ball well to the Najar/Pontius/Davies’s of the squad as well as be an offensive threat on his own and McCarty just couldnt handle that aspect.
      McCarty often seemed as if he was playing the DM position in an advanced role…

      • I agree, McCarty wasnt offensively potent, though look at Barca, how often do Iniesta and Xavi really score? with Najar, Pontius, Wollf, Brettscnheider, Davies, do we need a midfielder who can score? A defensive midfielder in the advanced role is more of what we need when we have 4 players who are so offensively potent, that leaves 2 midfielders and the back 4 for defense. When DeRo pushes, najar and Pontius will push, they are young, and wollf or brettschneider and davies or ngwenya are up top.
        That will leave one midfielder and our back four on the defense, our back four which is not so strong right now. Will DeRo go back to defend strongly, especially as he will be in the center of that offensive 5? He will need to.

        since the NBA is out and the NFL might be out, do you think the MLS can have divisions based on economic strength? I wonder your thoughts?

  3. Charles says:

    “I tend to consider tactics and Major League Soccer like an unabridged dictionary in a fourth grade classroom. While the study and employment of tactics within MLS happens, there is less tactical sophistication in MLS than in the more established leagues of the world.”

    And you lost me…..maybe the rest of your article was golden but your opening was SO stupid, I coulnd’t keep going.

  4. prindi says:

    Interesting piece. I agree with your initial hypothesis regarding why Nowak chose to play Daniel as a full back. Presumably, that hypothesis, was based on your seasoned knowledge and understanding of the game. The Opta stats, however, tell a different story by showing Harvey to be the more offensive and dangerous fullback.

    Why Daniel did not generate more offense, as you correctly pointed out, was likely due to the change in score. As the score changed, the tactics changed. Once Philly went up, any good centerback would’ve told his fullbacks to play smart and hold off on the runs forward. Which is what happened here.

    The score is just one of the variables that changes an individual and team’s play. The glory of football is that there are infinite variables that can all have an affect on how the game is played – weather, injuries, cards, substittuions, referee tendencies, etc. Understanding the cause and effect of these variables is what makes football the complex and amazing game that it is.

    Stat machines like Opta, while ok as a supplement, hurt the game of football. Too many pundits are relying on stats more and more to explain how and why a team plays the way it does. In my view, such heavy reliance on statistics indicates an individual who does not truly understand the game. Perhaps this comes from not playing the game growing up, never having proper coaching or just plain inexperience.

    Regardless, the bottom line is that the game of football is an art and deserves to be analyzed as such. Let’s leave the heavy reliance on statistics to less intricate and more quantifiable games like baseball.

    • Earl Reed says:

      Thank you for commenting, and I’m glad you found this interesting.

      I think another aspect I failed to consider, but may explain your observation: comfort level. A player like Keon Daniel being fielded in an unfamiliar position is likely to be uncomfortable in certain disciplines. Given the state of the match due to the score, he may have been so concerned with defensive responsibility that he was insecure with going forward in order not to get caught out of position.

      Thanks again!

  5. Ben says:

    Fascinating article. I’m just wondering, wouldn’t it be more useful to compare Harvey’s last third to Daniel’s first third. I feel like that would yield the answer to your hypothesis most clearly, with the score change and all.

  6. Mike M says:

    The chalkboard feature by OPTA on the MLS website is really awesome, ever since i heard about it i cant get enough.

    I am a Red Bulls fan and as you know there has been a major goalkeeper controversy. The major point i have heard for Sutton is that his distribution out of the back is sooo much better than Coundoul, and that is why he is preferred. So using the chalkboard, i was able to look at the distribution stats for each keeper, comparing successful and unsuccessful passes in multiple games. And what do you know, Sutton is no better than Coundoul.

    One thing I also learned, after looking at some other keepers distribution stats, keepers should never try and boot the ball past the halfway line, 95% of the time those passes are unsuccessful.

  7. Earl Reed says:

    I’ll combine two responses into one comment.

    Ben: That’s a good point, unfortunately Keon didn’t get touches early on in the 2nd half. He had four, and they were definitely not in advanced positions. And that is somewhat of the downside at this moment of OptaStats, is that you don’t get position data if there are no passes, dribbles, or challenges. Basically, there’s still a long way to go, but Opta seems to be at the forefront of implementing data acquisition.

    Mike M: Great! I think that’s a good observation about keepers and the direct “kick to the target man” method. It’s effectiveness isn’t great, but sometimes teams are wed to a philosophy trying to win 1-0. In that regard, if it can yield the one goal without causing the team to commit men forward, then tactically the victory is won.

    • Dave C says:

      To be honest, I’m surprised that 95% of long kicks are unsuccessful. I would have thought the very lowest the completion rate could be would be 50% (i.e. essentially evenly, randomly distributed between the two teams).

      Also @ Earl – you say because of the relative parity in MLS, games are often decided by “freak” events (mistakes, penalties, etc) rather than tactics. I don’t really follow MLS closely enough to argue, but I would have thought that if two teams are closely matched in terms of ability, then that should mean tactics are even more important, rather than less important.

      • @Dave, no tactics are less important, that is the open secret. When teams are matched in physical ability and technical skills, logic says strategy should then count the most, but functionally, they dont. Teams with equal skills and abilities dont overcome other teams with equal skills and ability by strategy because they dont have the skills and abilities to implement their strategy good enough to do so.
        DC United always talks about playing attacking soccer, but strategy isnt enough, they need better personnel, they have them now and are doing better, they havent changed strategically from last year really, but the quality of players, and their skill level has and thus the strategy is implemented better.

  8. predrag says:

    Great article and analysis.Although ,I would like to see at least one MLS team which plays classic catenaccio(score then park bus in front of goal…)This would be great opportunity to compare which strategy is the most lucrative.As former player I can only say that save lead by bunker is not that hard as it seems.But for opponent with score down and time running out is very difficult to equalize.As they say “best by test”.From Europe.

Leave a Reply

Your email address will not be published. Required fields are marked *