I have been an Intern in the analysis department of the Houston Dynamo under the very knowledgeable Oliver Gage for the past five months. If you haven’t heard of him before, he’s active on Twitter or can be found through a simple Google search. He’s well known and very respected in the industry for his work, both in public blogs and presentations and private analytics (club work).
These past 5 months have been amazing and I can say with confidence that I have learned a lot under Oliver. Especially how the ins and out of a professional football club function on a day to day basis when considering performance analysis and statistics. In this blog post, there are a number of areas that I would like to address which will hopefully give fans and aspiring analysts an insight into what to expect in MLS.
1. Major League Soccer (MLS)
What is the league all about, how does it work and how good is it?
2. The work of a Performance Analyst
What do we actually do? Why are we wanted/needed?
3. Weekly challenges in MLS
Weather, Travel, Games & Finances
4. An example of a day in the life of an analyst
Major League Soccer (MLS)
One of the reasons I choose to go to the USA instead of the UK (where most of the job experiences and openings are) was because I longed for something different, I wanted to stand out to future employers. Oliver and the Houston Dynamo also gave me the option of deciding how long I wanted to stay for which was a huge benefit, especially as I also had to plan finances. Other benefits so to say included a smaller analysis department, thus allowing me to learn, assist and add ideas to the club. This was definitely something I don’t think many other aspiring analysts, especially in the UK would have had the chance to experience.
As some people might already know, MLS is quite different to the Premier League, La Liga or even the new money crazy Chinese Super League. How different? This video explains it better than I ever could (and also saves me explaining every detail in writing). If you fancy learning about MLS’s history I would recommend “The Beckham Experiment” by Grant Wahl. It’s a good introduction.
First and foremost, I would argue the biggest difference is what is known as the salary cap. There are of course other factors that affect the teams such as travel, weather conditions and turnover of games (as they always do). The league recently became more flexible after introducing the Designated Player (DP) rule in 2007.
Hopefully, the first YouTube video answered all of the questions about MLS and how the league is structured. Teams are distributed to a conference as can be seen in the left image below. While I was in Houston, MLS confirmed that next season (2017), there would be a further two additions to the league: Atlanta and Minnesota United.
The work of a Performance Analyst
So what does a Performance Analyst actually do? Let’s ask some family and friends…
In some ways the above meme isn’t completely true but it’s also not completely false. We do watch a lot of games (for obvious reasons) but there are numerous other projects that we take on. Sure it’s all about those valuable three points is it not?
Since being here at the Dynamo, I quickly learned that our job is still largely misunderstood by a lot of coaches. Nothing against the coaches but I can understand their perspective (why do this?). Essentially, we support the coaching staff with pre, live and post-match analytical information (both video and statistical).
Matthew Lewis of Bristol Rovers explained it as such: “Performance analysis can be defined as many things and be used in many different contexts but ultimately it is the in depth critical analysis of a set performance, and being able to dissect various elements of the game to the clubs’ and coaches’ needs”.
This information could be anything. That reminds me of the quote from William Bruce Cameron “not everything that can be counted counts, and not everything that counts can be counted “. However, we try to provide anything the coach wants to know about his team and the upcoming opposition. Some coaches feel the need to want to know everything while others are flexible and don’t particularly want an IO (Information Overload).
Generally, most of the analysis (pre & post) is video based. This is due to data reliability issues as small samples cannot provide us with such accurate information as quickly as video can. Statistics are more useful in the long-term to use either for benchmarking performances or finding that “needle in a haystack player”. Elsewhere statistics can have an important role to play but NEED to be used correctly (otherwise they can be misinterpreted and could lead to some seriously bad decision-making)!
If you follow football bloggers in the football analytics community on Twitter you will have seen this tweet by Colin Trainor before. He’s quoted this statement directly from Prof Bill Gerrard’s blog “Winning with Analytics”.
Just this week, my supervisor Oliver described an encounter he once had at work and wrote a piece on the current state of analytics. Again, it’s worth the read!
During my five months in Houston, I worked on a number of exciting projects. I won’t go into all of them in detail, yet I have chosen five that I feel helped my development as an analyst the most.
1) Build upon and develop the academy analysis system in accordance with their style of play.
2) Daily filming and coding of Training sessions.
3) Help with general pre-game opposition analysis (video and statistics).
4) In charge of opposition set-piece analysis.
5) Produce and write regular reports for the coaching staff, e.g.: goalkeeper coach.
For home games, we figured out a way to capture live off a camera instead of using the TV feed. Don’t get me wrong, there’s nothing wrong with the TV feed (it’s better than having nothing)! Yet, being able to provide a professional service with whatever you do goes a long way.
Due to the sun, heat and humidity that Houston receives during the summer months, it was enjoyable for a change to stand outside and film training from the roof of the training facility. Certainly very different to back in Ireland and the UK with the rain-jackets, camera protectors and so forth.
However, once up there I got my daily dose of Vitamin D with temperatures hitting 38 – 40 degrees Celsius (plus humidity) and this view. Definitely better than any cold and rainy day back home!
As with everything in life, there are restrictions. Restrictions that can hinder personal or career development. Unlike the big clubs in Europe, performance analysis is relatively new to MLS. As far as I know of right now, most clubs (in MLS) currently only employ one analyst. This makes for long hours at a laptop, preparing footage and other pre-game visualisations the coaches want to see. Even for two analysts per club, the workload would still be intense but more manageable. It certainly is not a job for the average football fan. You really have to enjoy the work and the hours that come with it!
One week in the midday July heat, Oliver and myself decided to pull the scaffold over from the academy pitches (picture below – red dot) to the First Team pitch (green dot). The scaffold would allow us to be closer to the pitch but it ended up not being tall enough for a quality shot. (On a side note, I probably should point out that it was some effort pulling and pushing the scaffold over freshly watered grass. If I ever thought about entering the World’s Strongest Man competition, this would be a good exercise to start with).
Weekly Challenges in MLS
As I discussed earlier in this post, MLS over the course of the season faces a number of challenges (long travel to away games, weather conditions and a short turnover of games). These factors have to be taken into consideration when scheduling and planning for the season ahead. Let’s first take a look at the weather conditions that teams can face when travelling to away games in this league.
There are a number of different weather conditions that MLS teams can expect to face across the US. I think the challenge within this situation is more for teams playing each other outside of their conference rather than in it. This is due because the months of the season may spring some surprises on teams. If you watched the first YouTube video above (under the MLS header) then you should easily understand this next part.
For instance, let’s take LA Galaxy (Western Conference) and New York City FC (NYFC) (Eastern Conference) this season. LA are due to play in Houston twice this season while NYFC are only visiting once. The same statement can be applied to Houston travelling to LA and NYFC or any other team for that matter.
As an example, a team coming to Houston in March/April will most likely experience very different weather conditions than if they play in July/August. The summer temperature can reach up to 37 degrees Celsius and average around 28 degrees Celsius (although the humidity always makes it feel hotter). With this humidity, come thunderstorms and heavy rainfall (which delayed our home game vs Toronto by 24 hours in August). Other examples can be found in March/April when playing Vancouver away (possible freezing temperatures) and all year round in Colorado having to deal with altitude.
Travel & Turnover of games
If the US was not as big as it was and only had to deal with the weather conditions that it has, playing there would not be as difficult. The travelling factor can take it out of anybody and everybody. The hours spent between home and the airport, the waiting around to depart and at baggage reclaim, the journey to the hotel and if needs be accustoming to the new time zone can drain even the best of us. The US is so big that travel can take a while (6+ hours between New York and Los Angeles) and even longer if you are delayed by the extremities of the weather.
I could now go off on a tangent and explain this through examples yet why not read about it straight from the bookshelf. This past week, there was a big discussion about travel in MLS. Most notably why MLS is so behind the times when compared to the other professional sports leagues in the US. The LA Times wrote about travel plans & schedules in MLS while Brendan Kent of American Soccer Analysis (ASA) investigated the same topic from a statistical standpoint. These are not the only links available and certainly won’t be the last ones.
Now that I have mentioned ASA, they are a great bunch of guys with some real knowledge of the game and should be used more often by MLS teams. I’m sure they would be thrilled to help your club find the next cheaper version of a Giovinco, Villa or Wright-Phillips.
Back to travel. Without a doubt, all this travelling can certainly take it out of your players and your team. How does the manager take care of his players in this regard? Thankfully, that was not part of my job description (we as analysts already have enough to do) yet it would still interest me. How might it look from an analyst’s perspective? Oliver might be able to answer that if you ask him nicely.
The implementation of a system such as the salary cap is a common tradition in American sports. This allows the league to primarily thrive on a more balanced level as can be seen by the different winners (10 teams) of the Supporter’s Shield in the league’s 20 year history. The league has undoubtedly received criticism about the legitimacy of using such a system by Consultancy Firm Soccernomics.
MLS is also the first league to my knowledge (feel free to correct me if I am wrong) to publicly release each player’s salary details and this has caused quite a stir as such.
Elsewhere, maybe not directly related under this heading but still worth a mention is assembling an MLS squad based on a salary cap. Opta who collect match statistics in a number of sports, recently partnered with the University of Columbia and Havard in allowing students to build the optimal roster. Clicking here shows the methodology one team used.
Lastly, if you fancy reading through the pages and pages of MLS Roster rules, please click here.
A Day in the Life of a Performance Analyst
So what does a working day really look like? It varies a bit from day to day as it’s not your standard Monday – Friday 9am – 5pm job. The hours vary depending on the upcoming week and the reasons could be due to 1) When is training scheduled, 2) How many games are this week and 3) It’s Game Day!
For the most part, training took place in the mornings (09:30 – 11am) or occasionally in the evenings at 5/6pm. For the morning training sessions, the days started whenever the alarm clock rang (for me it was 06:30am). As long as you were in work for before 8am, it was a ok. When we trained in the evening, the coaching staff usually got to the office at lunchtime and stayed until 9:30pm. Also if you think the work is done during these “office hours”, you are greatly mistaken. Often you will need to complete another bit from home, more or less to the annoyance of your other half!
Nevertheless, below is an example of how a typical day (let’s use Monday) panned out for me in Houston.
- Wake up 6:30am
- Quick breakfast at home
- Get to the training ground for 07:50am
- Check updates on Twitter (the football analytics community does some really cool stuff), quick read of the mornings news (Brexit and so forth) and reply to some emails. This would usually take about 20 – 30 minutes.
- After that I would start managing the data and video on our next opponent and start with the pre-match report.
- After working on the pre-match report for the last hour, training would have just started (09:30am). I grab a bite of “second breakfast” (if the players leave anything) along with water bottles and head out onto the roof.
- 11:00am – Training finishes and I start to breakdown and format the video for the coaching staff. Quick shower and then its lunch time.
- After lunch, I continued with the opposition report unless other tasks were asked of me. Occasionally, the goalkeeper coach wanted to discuss evaluating goalkeepers through data so I jumped on that with reports.
- In cooperation with Oliver and the academy staff, I continued building a form of feedback report for the academy to use.
What does game day feel like?
For the first time being on the other side of the fence, I can honestly say (to me) it doesn’t feel very different to being a fan. I never travelled to away games so I cannot share my experience with you on that but I was present for home games. Sure you are there to work but you are also one of the privileged few to work in the sport you love (although the hours and weekends are taken away from you). Yes we work while the game is on, but deep down, I am still a football fan, excited to see the team play well.
Standing next to the TV cameras where we filmed the games, there were times just before kick-off during my first couple of games where I had doubts regarding the opposition – “How much did we really know?”, for example, maybe they changed the central midfielder and now he’s playing on the wing. The question of ‘have I done enough to prepare the team’ went through my mind. I quickly realised that these were really rhetorical as it came down to the players executing the game plan we developed.
It’s impossible to know and predict of how the opposition might play but we do our best (along with the coaching staff) to prepare the players for what we think is going to happen.
*Reminder to view the screenshot image I took from Colin Trainor above. *
By now, I think each team has their way of managing the extremes in MLS, except maybe per se for newcomers Atlanta & Minnesota United. Overall, all the above factors that were discussed give fans and aspiring analysts an example of what MLS is all about. Clubs will always know who they’ll play in their own conference but outside of their conference is anybody’s guess. (That’s of course until the fixture lists gets released). The only thing that most likely is not going to change is the conference you are assigned to (as this is based primarily on the cities location).
That is all from myself regarding my stay in MLS. I do blog from time to time either for American Soccer Analysis or via my own blog. Currently, I am gathering team and player statistics on the German Bundesliga via my Tableau Public profile. I also plan to write some more in-depth analysis on the Bundesliga soon, so watch out for this space.