Interview with Professional Data Analyst Joel Salamon

Luke Griffin@SL_TeamTalk

Do you want to work in professional football as a data analyst? Well, hopefully this can help you. SideLineTeamTalk conducted a short, casual interview with professional data analyst Joel Salamon about his job in data analytics with the aim to help aspiring analysts find out more about how to get into the field of professional data analytics in football.

Joel Salamon is a data analyst who has worked at Brentford, FC Midtjylland and Bristol City, but currently looking for work with a professional club. He runs a fast growing YouTube channel called Messi Seconds. It’s one of the first data analytic blogs to move to YouTube and it works fantastically. The format is good and easy to follow with constant good graphics and visualisaitions. You can find him on twitter @MessiSeconds where you can see his opinions and interesting interaction with the analytics community.

Joel Salamon also did a talk at the OptaPro forum this year which you can watch here. He researched into break out ages of footballers and how it affects their careers.

The Interview

1. Which club do you support?
It’s a massive dickhead answer… but whoever I’m working for!

2. Which club are you working at and how did you get the job?
I’m currently a shameless free-agent! So I’m currently talking to new clubs, including two ‘Champions-League-ish’ teams. Previously I’ve been at Bristol City, Brentford and FC Midtjylland.

3. What did your average work day consist of when working with those clubs?
At Bristol: watching, coding and feeding back on games. Some analysis stuff. At Brentford and FCM: lots of coding, lots of numbers, lots of analysis, lots of learning!

4. Which clubs are you talking to about working with?
Nice try.

5. How did you get those jobs?
At Bristol: I had thrown my CV around to clubs for a year or two, unsuccessfully. Eventually an opportunity at BCFC opened up, and with an improved CV and references I got it! At Brentford and FCM: poking them several times from multiple angles, plus a lot of luck, plus the kindness of Matthew Benham and Ted Knutson.
6. What is your current job title and what is your day to day role?

“That schmuck from twitter”. I was preparing to return to Brentford and Midtjylland until the analytics team became defunct last week. Thus, my day to day role now includes talking lots of rubbish on twitter whilst learning more and more about the field, in preparation for wherever I am next.

7. What are you currently working on?
Currently I’m improving my coding, data visualisation work and getting my head around some more complex expected goals stuff.

8. Is coding important for analysts and what language do you use?
Extremely. Predominantly R, some Python, some other stuff. I’ll have to learn much more in future (as will everyone)!

9. How does someone need to start to become an analyst and can you list any landmark articles to help beginners learn more about data analytics and do their own analysis?
Tom Worville has some great stuff for beginners (blog posts + @FanalyticsBlogs). Other than that read lots, ask the right questions and then actively answer those questions!

10. How did you first get involved with football analytics?
I suppose I first got involved in primary school maths lessons. Basically, I’m a massive nerd. More specifically, some of the book Soccernomics mashed several of my interests together; Ted Knutson’s Statsbomb articles consolidated that new-found love and made me realise that analytics wasn’t simply the stuff of fringe paperback; then a tonne of reading, a few crappy videos and luck landed me here.

11. As an analyst do you have to watch a lot of games?
Last year I probably averaged 20 full games per week (though most of them I attempted to watch simultaneously with a second). Now I watch less, but try to catch as many tactical trends and exciting players as possible. I personally haven’t had to watch much for club work, though everyone else does have to watch a lot.

12. Do you need to be great a maths?
Nope. You have to think analytically, and certainly going forwards need to understand more and more data science, but the beginning is mostly about ideas and coding. Most of the pure numbers stuff I’ve done so far probably could have been learned from lots of googling.

13. Did you work closely with coaches and did they use your input?
I personally haven’t done much with coaches, but of course people less junior than me (i.e. everyone else) have to. There’s no point in analytics unless it’s directly actionable, and it can’t be put in to action on the pitch without great communication and understanding between coaches and analysts. Those discussions are just as important as the analytics itself.

14. What are your ambitions for the future?
I think most people who want to get in to this would ideally like to run successful analytics consultancies or go in-house. My ambition is a bit different. In a dream world where I could do everything, I’d want to go into management and use these ideas from “the inside”.

15. What qualifications and degree do you have? Have you done any coaches badges and do you intend to?
I did Maths, Further Maths, Physics and Economics at A Level with a few academic awards. I’m now in my second year of Economics at Bristol and will possibly be applying to US masters programmes next year. I’d love to do coaching badges, but haven’t done so yet (to my frustration).

16. How did you get involved with the OptaPro Forum?
Opta ask for research abstracts around October. I feel very fortunate to have been a speaker – both in terms of the flattering opportunity and also the fortune of being selected by the judges (I preferred the other proposal I made but didn’t send!).

17. How did you grow your YouTube channel to where it is today?
It’s still very small in relative terms, but to get to where it currently is the inputs were just ‘lots of work over time’ plus ‘making stuff I would personally enjoy’.

18. Do you read many books/blogs on football and analytics?
My great embarrassment is that I don’t read books and haven’t for a long time. In terms of football stats blogs I read literally all of them.

19. Which are you favourite books/blogs? 
Soccernomics was great for sparking my interest, but I disagree with much of it now. Numbers game is supposed to be great. But my favourites are @Statsbomb, @deepxg, @willtgm, @goalimpact, @analyticsfc, @mc_of_a and @them_l_g.

20. I’ve recently read Soccernomics, what are a few of the principles you now disagree with?
I can’t remembered most of it, but for a start – the idea transfer spending doesn’t matter; the idea managers usually don’t matter; the idea that English footballers are mostly working class because the working classes have more free time (??????????) – I’m sure there’s more besides that. Simon Kuper is one of my favourite writers though.
SLTT wants to say a big thank you to Joel Salamon for the interview, make sure you check out his YouTube and twitter. I hope this was interesting and useful.
Feel free to give feedback and let me know if you want more interviews at sidelineteamtalk [@] gmail dot com.
Also, make sure you follow me on twitter @SLTeamTalk

One thought on “Interview with Professional Data Analyst Joel Salamon”

Leave a Reply

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

You are commenting using your 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