Performance Analytics and Pedagogy

Some recent posts have encouraged me to think about pedagogy for a new age of performance analytics in sport.

It started with Mine Cetinkaya-Rundel‘s speakerdeck Let them eat cake (first)! (link). Slide 16:

Slide 61 really pushed me to think about how we might share with a different kind of pedagogy:

… and brought back memories of Jo Ito‘s observation “education is what people do to you, learning is what you do to yourself”.

Next up was Karen Gold’s Transforming the First Ten Minutes of Class (link). In her post, she notes:

After attending Penny Kittle’s workshop on 180 Days: Two Teachers and the Quest to Engage and Empower Adolescents last summer, I made the decision to shift my teaching-. Like most teachers, I’ve done a lot of professional development. I’d come away refreshed and excited to try something new, but too often, it was challenging to incorporate a big, new idea into the fast-paced routine of school. Penny’s workshop was different. Something resonated with me that summer morning, and I thought, “I can do this. I WILL do this.”

Karen’s story shares her experiences of encouraging children to read at the start of a lesson. Day 1:

Instead of going over a syllabus or introducing course expectations, the librarians and I gave brief book talks, sharing novels we had read or that we knew were well-received by young adults.

This sounded like Mine’s cake to me. As did Solomon Kingsworth’s discussion of reading comprehension (link), he proposed:

If reading comprehension relies on background knowledge and mental models of the world, then the purpose of our lessons should be to leave the child with more knowledge and mental models.

Solomon talks about the pedagogy that shares the treasure that lies within each book.

This pushed me to think how we share treasure in our domain and epistemic culture in a new information age. And how, as The Economist suggested recently, our first step is “to understand that it is not data that are valuable. It is you” (link).

Three examples from sport appeared as I was pondering these issues:

Laura Seth shared news of a webinar hosted by the FA in January to discuss Performance Analysis & Effective Observations (link).

Mladen Jovanovic published Predicting non-contact hamstring injuries by using training load data and machine learning models (link).

Sam Robertson tweeted a list “of the type of sports science/analytics research I think we need to see more of in 2019”:

  • Optimising the structure, efficiency and communication practices of practitioner teams
  • More club, institution, university and manufacturer collaboration to address ‘whole of sport’ problems
  • Longitudinal skill/learning interventions in team sport settings.
  • New and better methods for coaches to improve communication, rapport & trust with athletes
  • Analysis of raw tracking data.
  • Interdisciplinary collaboration,  psychophysics (utility of visuals in reporting and learning), cognitive science.
  • Field application of work undertaken in other disciplines (deep learning & unstructured data), automation and semi-automation of many manual processes currently faced by sports practitioners, and human and machine integration.

Laura, Mladen and Sam are actively engaged in service delivery in high performance sport. As I read their posts I was thinking about how a pedagogy of praxis might engage the next generation of performance analytics.

I am thinking that my pedagogy will move even more strongly to an unmeeting approach with lots of mention of cake.

Photo Credit

Person holding black fruit near cake  (Alex Loup on Unsplash)

Western Bulldogs’ Ballarat #Hackathon

A year ago today, I was celebrating Leicester City’s Tactical Insights day.

I thought it was a delightful adventure in sharing and exploring data.

Tonight, under the leadership of Sam Robertson, it is the Western Bulldogs’ opportunity to be innovative.

The football club and the City of Ballarat are hosting a weekend #hackathon.

There is a website to provide information.

Some of the data used at the event will not be in the public domain and attendees will sign a non-disclosure agreement.

The fun starts this evening at 6pm at the Ballarat Library.

I am hoping it has this kind of energy found in Ballarat.

Momentum into and away from the 2015 #AFLFinals

13904755286_fd0bf88738_z

The 2015 AFL Finals rounds are about to start.

The top 4 teams going into the Finals are Fremantle, West Coast, Hawthorn and Sydney. Their momentum throughout the twenty-three rounds of the regular season was:

AFL Top Four 2015

At the other end of the AFL Ladder, the bottom four team were Essendon, Gold Coast, Brisbane and Carlton. Their momentum was:

Bottom Four AFL 2015

Sam Robertson and his colleagues (2015) have reported their analysis of match outcomes in the 2013 and 2014 AFL seasons. They indicate:

Logistic regression and decision tree (chi-squared automatic interaction detection) analyses both revealed relative differences between opposing teams for “kicks” and “goal conversion” as the most influential in explaining match outcome, with two models achieving 88.3% and 89.8% classification accuracies, respectively.

 

In the 2015 season, the third placed team, Hawthorn, were the team that took most kicks and their kicks at goal were the most accurate (55.1%). Essendon were the least accurate team kicking at goal in the 2015 season (46.3%). The two bottom teams Brisbane and Carlton scored the least Behinds in the season. Both teams had the highest losing average margin too (Carlton 51.7, Brisbane 48.2).

Fremantle, the Minor Premiers, unlike West Coast and Hawthorn, did not have a high average winning margin. In fact, they were ranked just above Carlton and Essendon at (29.2 points per victory).

Photo Credit

Footy (Beau Lebens, CC BY-NC 2.0)