There are a few questions that I get asked from time to time, so I thought I'd share my answers here. 1) How can I get a career in football analytics? 2) Are there resources available to help me learn how to do football analytics? 3) Is any of your work public?
Looking forward to presenting at
@Harvard_Sports
tonight. Will be discussing everything from pitch control to passing probabilities to off-ball scoring opportunities.
@HudlAnalysis
I don't normally post about my personal life, but some events are too big not to mention. I recently got married to an absolutely amazing woman,
@kirakira1712
, and now that the photos are in, I wanted to share.
This is an extremely insightful interview with the inimitable
@analyseFooty
. Anyone interested in what analytics actually looks like at a club should give a listen:
2) There are a lot of good resources for getting started with football analytics. One of the best is the Github repository created by
@devinpleuler
. It has a ton of invaluable information and links to other content:
This might be among the best line-up I've seen at a football analytics conference. A lot of distinguished speakers who are out there putting the theory into practice at the club level in addition to some other great minds in academia and industry.
The 2019
@SoundersFC
Analytics Conference Schedule 📊⚽️📈✅
We are beyond excited to listen and learn from this group of excellent practitioners and thankful for their time. 🙏🏽
June 22-23 in Seattle
Register here 👇🏽
#SoccerAnalytics
#SSFCAnalytics
3) Here are some links to my own public work. Quantifying Pitch Control - OptaPro Analytics Forum 2016:
This talk introduces a data-driven model of pitch control based on time-to-intercept.
Proud to have
@ChampionsLeague
winner Dafydd Steele from
@LFC
talking “Efficient Player Recruitment” - surely they know a bit about this topic
#SSFCAnalytics
I know I'm not the first to say it, but amazing to have
@analyseFooty
and
@devinpleuler
in the MLS Cup Final! What a win for analytics and what a great run both teams have had so far.
Loads of content up on Friends of Tracking now. In terms of tracking data content, here's where we stand:
Part 1 - Gaining deeper insights on shots & passes.
Part 2 - Measuring the physical performance of players.
Part 3 - Pitch control.
Part 4 - Evaluating passing options.
This is an excellent piece by
@EveryTeam_Mark
that does a deep dive into Reep’s career, the impact he had on the field, and how he, along with Richard Pollard, were developing models decades ago that are still considered new and interesting today.
Beyond Expected Goals - MIT Sloan Sports Analytics Conference 2018:
This paper extends the concept of pitch control to create a spatial value model in football.
Next Tuesday 15th
@LukeBornn
and I will be presenting our latest paper at ECML-PKDD 2020 conference. We introduce SoccerMap, a flexible deep learning architecture for predicting full probability maps from raw tracking-data.
Some professional news...at the end of this month I'll be leaving Harvard to join
@BrianPrestidge
's Data Insights and Decision Technology team at City Football Group
I'm very excited to join CFG and have the opportunity to work at the cutting edge of data & analytics in football
Poland plays Senegal in soccer.
70% of respondents identified Senegal as the more athletic team when watching video
62% of respondents said Poland was the more athletic team when using pose/computer vision
It was the same clip of the same game
#nessis21
@GregorydSam
1) I got my start in sports analytics thanks to working at
@Hudl
. Before that I did my Ph.D. in particle physics at
@CERN
. My route isn’t the most common way to get into the field and
@mixedknuts
has good advice on the topic:
Physics Based Model of Pass Probabilities in Soccer - MIT Sloan Sports Analytics Conference 2017:
This paper introduces a simulation approach to quantifying the control players exert on the ball in dynamic situations.
Very excited to present "Extracting Player Tracking data from video using non-stationary cameras and a combination of computer vision techniques" at
#NESSIS
on Sept. 28th! Here's a sneak peak of what it looks like:
@TiagoEstv
Pretty sure it refers to the mountains between the Black Sea and Caspian Sea. Traditionally the Democratic candidates in Iowa must run to the summit of Mount Elbrus and the first to get there wins the party nomination.
As a result, I am currently looking for a new challenge, preferably within football analytics or sports analytics. If you are interested in having a chat with me, then please feel free to get in touch!
It's very easy to get a COVID-19 test in Liverpool. I walked in, was tested 5-min later, and had results 30-min after that. Worthwhile if you haven't done it already.
I've put together a Soccer Analytics Handbook.
It provides some guidelines on what I think is worth learning, a bunch of interactive notebooks via
@GoogleColab
, and an extensive reading list of historical analytics writing and research.
Thanks to
@Soccermatics
&
@hugoriosneto
for their webinar on Football Analytics today! We hope you found it insightful + also see how to work in a football club + the experience when inside one. You can still access it for 24h by registering here:
New Friends of Tracking video tutorial: How to build your own pitch control model in python.
Based on the model of
@the_spearman
, I show you how to build a pitch control model and use it to evaluate a player's passing options.
#FoT
New video tutorial in our Friends of Tracking series: "Delving deeper into the tracking data - measuring the physical performance of players."
#FoT
I describe how to make movies from the data, measure player velocity, and create physical summary reports
It was really excited to present our paper at
@SloanSportsConf
this year, and am absolutely thrilled that
@suds_g
and I finished in first place in the research paper competition.
#SSAC21
Delighted that
@FA
data scientist
@AlexThomasTheFA
has joined our roster of speakers for 'Big Data: The Future of Football?' in Manchester on Nov 22nd
England teams are doing a lot of excellent work in this area/ will be fascinating to find out more...
In less than 24 hours, we meet again -- and for the first time.
See you at 7pm at
@LirOnBoylston
for the Soccer Analytics Meetup.
Drinks and food on us!
@LukeBornn
Guess I'll have to buy scientific journal access now 🤷♂️
In seriousness, thank you for your work. It's hard to overstate your influence on my career and those of many others.