RSS Twitter
Home Blog Page 4

Analytics: theory and practice

Stephen Downes provided a link to Dominic Wyse’s presentation (link) in his newsletter today (link). In his presentation as President of the British Educational Research Association, Dominic looks at the history of education as a discipline. His paper identifies three traditions in the discipline: academic knowledge traditions, practical knowledge traditions, and the interaction between the two traditions.

These traditions resonate with conversation underway about the place of analytics in sport. In his paper, Dominic observes that a discipline is defined “not only as an area of knowledge, but also as a community of scholars with a shared heritage which includes an infrastructure and traditions of published outputs and other modes of communication that underpin the discipline”. In his identification of traditions Dominic notes that Academic knowledge traditions foreground academic knowledge. Practical knowledge traditions includes the ideas behind competences and standards, and ideas that are part of networked professional knowledge. Integrated knowledge traditions attempt to bring academic and practical knowledge into some kind of relationship with each other. This integration involves activities such as action research.

Dominic points out that different knowledge traditions coexist in the history of development of education in any one country or region. I think the same is true of sport analytics too particularly as academic departments in universities seek to integrate knowledge traditions and emerging practices.

He concludes his paper with this observation “education and its research in diverse topics, using a wide range of methods, influenced by a multitude of theories, is seeking to find ways that inequalities in learning, or we might say ‘intergroup learning’, can be addressed through appropriate teaching”. I think this is a fundamental issue for sports analytics too as we start to contemplate how to manage and analyse the vast amounts of data becoming available.

In this regard, the Let Them Eat Cake (link) approaches in backward design encourage us to think how we prepare students for a world of work and how we integrate the knowledge traditions, Dominic identified. We have a great deal to learn from the discipline of education as we contemplate how to integrate academic approaches to analysing data with dynamic practice in sport.

Photo Credits

Photo by Gaelle Marcel on Unsplash

Photo by marif shaik on Unsplash

#OptaPro Forum 2020

The #OptaForum was held in London on 5 February 2020 (link). The proposed presentations for the Forum were judged in November. The judges scored each proposals based on three criteria: innovation, relevance and application. “By the end of the process, six projects had been chosen for presentation, with a further two submitters invited to exhibit their proposals as posters”. In December 2019, the successful proposals were announced (link).

The six main presentations were:

  • Karun Singh: Learning to watch football: Self-supervised representations for tracking data.
  • David Quartey: How do attacking and defending strategies affect goalscoring opportunities from throw-ins?
  • David Perdomo Meza, Daniel Girela and Mark Thompson: Tactical insight through team personas.
  • Santhosh Narayanan: Modelling event sequences in football using multivariate point processes.
  • Dan Barnett: I’m in a wide open space: creating opportunities at set pieces.
  • Vignesh Jayanth: Identifying and evaluating strategies to break down a low-block defence.

The two poster presentations selected were:

  • Pieter Robberechts: Contextualized performance projections for soccer players.
  • Javier Buldú: Using network science to quantify the identifiability of football teams.

Each year, since 2014, the Forum has given a feel for “innovation, relevance and application” and indicates the issues we might need to monitor as analysts. The Opta web site has a link to previous years’ papers (link).

Photo Credit

Opta Forum (Opta Website) (link)

A pedagogy of small

Tanya Elias, Laura Ritchie, Geoffrey Gevalt, and Kate Bowles (2020) (link) have written about the pedagogy of small. Stephen Downes shared the link to their University of Calgary paper in his daily newsletter (link).

Tanya and her colleagues revisit open education theories from the 1970s “in order to understand the ecologies that shape open learning in the present”. These ecologies explore “three value pairs that constitute axes of negotiation in the open classroom”:  

  • Autonomy and interdependence
  • Freedom and responsibility
  • Democracy and participation.

My experience of CCK08 (link) led me to think about how these three value pairs connected as self organising networks (link). I see these networks as the energy and agency that drive open online courses … small or massive.

Tanya and her colleagues addressed the issue of small in their thinking. They observed “small offers a choice to those who have, either consciously or unconsciously, already experienced moments where “the recipes [they have] inherited for the solution of typical problems no longer seems to work”. I sense that these recipes now exist in different ecologies and that have much to do with how we engage with open source communities that “offer a counter-cultural model for learners to interact technologically, economically and socially” (link).

I was particularly interested in their discussion as I have had the particular difficulty of dealing with the term ‘massive’ in the MOOC label. Massive to me always seemed an enormous number to me that challenged us to think about our pedagogy of engagement in open educational resources. It is for this reason that I labelled my own work in this area as SOOC (link) … a small open online course. And one that had the characteristics of a Souq (link).

Photo Credits

Riffing (link)

Souq, Aleppo

#WNBL Regular Season 2019-2020

The WNBL Regular Season ended on 1 February 2020 with Southside’s 78v61 defeat of Sydney (link). The WNBL has moved into the Finals part of the season. The four teams in the Finals are: Southside, Adelaide, Canberra and Melbourne Boomers (link).

During the Regular Season the median winning score was 82 points and the median losing score was 71 points.

My ggplots (link) of the performances of winning and losing teams:

By quarter the scores were:

Photo Credit

Southside (Twitter)

Observation

The National Gallery of Australia is holding a Matisse and Picasso exhibition at the moment (link). The introduction to the exhibition noted:

Matisse & Picasso is the first exhibition in Australia to tell the story of the artistic relationship between two of Europe’s greatest twentieth-century artists. Henri Matisse (1869–1954) and Pablo Picasso (1881–1973) met in 1906 and for more than half a century followed each other’s creative developments and achievements. The sustained rivalry between them was not only key to their individual success, it also changed the course of 20th century Western European art. The exhibition included paintings that will be on display in Australia for the first time. The exhibition features more than 60 paintings and sculptures drawn from public and private collections internationally and in Australia.

What I particularly enjoyed about the exhibition was the juxtaposition of both artists and the narrative evident in their sustained rivalry over a long period of time. I was impressed by both artists’ desire to innovate. It was wonderful seeing so many of their paintings and illustrations. Themes in the exhibition linked to:

  • Different worlds
  • The battle over Cubism
  • Reshaping space
  • Designing for dance
  • The past as the future
  • Exotic worlds
  • Radical chic and the cult of the ugly

One of the program notes indicates that “over time, Matisse and Picasso no longer felt the need to compete to be first with the newest and most radical ideas” and added “each artist now had the freedom to exercise their imagination”. This transition fascinated me.

I believe the art world has a great deal to offer sport analysts. Doctors have included visits to galleries in their training years to learn about observation (link).

As I left the Matisse and Picasso exhibition I came across an indigenous art exhibition about belonging. In the write up about the exhibits, there was a quote from Charles Perkins. He suggested “we know that we cannot live in the past but the past lives in us”. This is what I think going to a gallery does for us. We are able to look carefully at our present and future by understanding we are part of a story.

Photo Credits

Portraits of Henri and Pablo from the exhibition images.

Tactical Insights Day

Leicester City will host a Tactical Insights Conference at the King Power Stadium (link) on Wednesday 19 February, 2020. This is the second Insights conference at the club. The first conference This “was a culmination of six months hard work from those within the analysis department and club”. The Insights day aimed to provide an educational basis for the discussion of tactical approaches in football (link).

I thought it was a great innovation in open sharing and mirrored developments at the Barça Innovation Hub (link). A perspective on the work underway at Leicester appeared in November 2018 when the post of Head of Fooball Analytics was advertised at Leicester (link). Mladen Sormaz was appointed to the position (link).

I am delighted to see developments at Leicester and Barça as I am firmly of the view that open sharing of data and processes is the new competitive edge. This open sharing is happening in rugby league too with Castleford Tigers. In the most recent Analysis Pro newsletter, Josh Bryan shared news of an event on 20th January when Castleford Tigers hosted an evening sharing insights to their performance analysis processes. There is a video of the evening that is available introduced by Ste Mills, the Head of Analysis at Castleford (link). It was the second such evening at Castleford.

The advertisement for the Leicester day observes “in a progressive and fast-moving industry that continues to show its undoubted value to the world of professional sport, this conference is the perfect opportunity for Performance Analysts, data scientists and coaches to continue their own professional development, make key contacts and improve their knowledge and skill set”. My hope is that a “progressive and fast-moving industry” can demonstrate the openness characteristic of open educational initiatives that are central to our discussion of on-going learning.

The program at Filbert Street is available (link) and can be downloaded as a pdf file. The schedule includes presentations from:

  • Paul Balsom and Mladen Sormaz (Leicester City)
  • David Slemen (Opportunities in a data driven era I)
  • Jordi Fernández (An integrative analytical approach)
  • Lisa Fallon(Converting data to match winning margins)
  • James Krause (Analytical approaches in an Academy environment)
  • German FA (Integrative analytical approach II)
  • Sportlogiq (Integrative analytical approach III)

News of the event also includes profiles of the presenters (link). The Football Association will accredit 6 hours CPD for those attending the Tactical Insights conference.

I am hopeful that the gender balance in the conference might change. If we are to be “a progressive and fast-moving industry” we need to address the gender imbalance at such conferences.

Photo Credits

All Leicester images are from the Tactical Insights blog post (link).

Ste Mills, Head of Analysis at Castleford (link)

Defending a large total in #BBL

Sydney Thunder has defeated the Hobart Hurricanes in a BBL eliminator game (link). Sydney batted first and scored 196 runs in the innings. Hobart were all out for 140 in their chase.

196 runs was a large total to set in BBL (there have been six totals in the BBL regular season that were greater than 196). In the regular season there were ten occasions when the team batting first set a rate of 19 runs or more per partnership. The team batting first has won all ten of these games. The result raises the question of how does a team batting second chase down a large total. Thus far, the Sixers (twice) and the Scorchers (once) have been able to chase down a rate of 18 runs per partnership.

My ggplot of the Thunder v Hurricanes game is:

Photo Credit

Jono Cook (Cricinfo)

Analytics' Notes

Introduction

Like most people, I receive pull newsletters in my in box. I find these profoundly helpful. They often contain news of analytics that become tweets (link) or the stimulus for a blog post on Clyde Street. I have been using Mastodon occasionally (link).

Analytics’ Notes

John Danaher has written about algocracy (link). His paper prepared as a chapter in the Oxford Handbook on the Philosophy of Technology considers how algorithmic governance can be “emancipatory and enslaving”. John points out “advances in computing technology have created a technological infrastructure that permeates, shapes and mediates our everyday lives”. John explores “the unavoidable and seemingly ubiquitous use of computer-coded algorithms to understand and control the world in which we live”.

Laurence Goasduff looks at augmented analytics (link). He notes “augmented analytics uses machine learning and artificial intelligence techniques to automatically identify actionable insights”.  Laurence’s post uses data from a Gartner Data and Analytics Summit poll on the potential transformative effect of augmented analytics (link). Gartner’s glossary shares this view of augmented analytics. It is:

“the use of enabling technologies such as machine learning and artificial intelligence to assist with data preparation, insight generation and insight explanation to augment how people explore and analyze data in analytics and business intelligence platforms. It also augments the expert and citizen data scientists by automating many aspects of data science, machine learning, and AI model development, management and deployment” (link).

The Seattle Stats Guy shared on Medium his account of predictive modelling in R (link). The paper links to a 2018 post (link) about the use of ARIMA and ETS in R for modelling purposes. There is also a link to another 2018 about a vocabulary for predictive modelling (link). There are, the Seattle Guy suggests, four important words:

  • Stationarity (a time series that has a consistent mean, variance, and covariance).
  • Autocorrelation (the correlation an observation has between itself and another observation in the time series).
  • Stochastic
  • Differencing (can help stabilize the mean and remove stochastic trends).

Davide Bacciu and his colleagues have shared their tutorial into Deep Learning for Graphs (link). The tutorial seeks to introduce “the basic building blocks that can be combined to design novel and effective neural models for graphs”. I am mindful that i need to look more carefully at graphs as I contemplate the relationship between informatics and analytics in sport. David and his colleagues’ tutorial is most helpful in thus regard.

Conclusion

The volume of push material available is remarkable. Trying to follow links in newsletters is fascinating but requires a great deal of discipline.The examples shared here are meant to look at how we practice analytics and how we position ourselves ethically in the algorithm debate. Andrew Manley and Shaun Williams, amongst others, have looked at data and organisational surveillance (link). As John Danaher indicates, we need to think carefully about our use of machine learning and artificial intelligence. We need to consider our relationship to augmented intelligence as we adapt to the volumes of data available to us.

Photo Credit

Surf boat (Ed Dunans, CC BY 2.0)

Digital Dexterity

Introduction

VALA is holding its 20th bicentennial conference in Melbourne from 11-13 February 2020 (link). The theme of the conference is Focus on the Future.

VALA – Libraries, Technology and the Future Inc. is an independent Australian based not-for-profit organisation that actively supports the use and understanding of information technology in libraries. I have been particularly impressed by the exploration of innovative practice.

VALA’s alert to the conference looked at the digital dexterity framework developed by the Council of Australian University Librarians (CAUL) (link). The conference call included this invitation “our call to action and our challenge to ourselves, our exhibitors, sponsors, speakers and delegates is to be visionaries, to see through the lens of the future and to bring our best, most creative, contemporary and controversial selves to VALA2020”.

I liked VALA’s description of the conference as an environment that “offers inspiration and collaboration. We welcome the diverse community that contributes to our industry, our students, practitioners, academics and vendors, colleagues and competitors. It is a safe and respectful place where you can take risks, challenge yourself, ask questions and invite feedback, and where we can all learn from our shared experiences”.

Digital Dexterity

The Council of Australian University Librarians Digital proposes that dexterity is a critical component in the success of digital societies: it encompasses the “cognitive ability and social practice needed to leverage and employ various types of media, information and technology for advantage in unique and highly innovative ways that optimise personal and business value”.

VALA’s conference theme has enormous importance for how we develop dexterity in sport. The conference “has actively sought the participation of digital dexterity champions to provide core content for the conference program”. This content includes:

VALA also has two conference themes to support the focus of the conference: digital literacy and digital dexterity. The papers in these streams include:

Literacy

Dexterity

The VALA website has comprehensive information about the conference and contains a program of events (link).

Conclusion

I think this year’s VALA conference raises profound issues for how we in sport address digital dexterity and how we share research and practice. The VALA website includes a link to JISC’s Building Digital Capabilities paper (link). In that paper, JISC suggests that one of the aims of the capabilities framework is to “support discussion and build consensus about the capabilities required in a digital organisation, perhaps in order to develop a local framework or a locally adapted version of this framework”. I see this relationship between organisational focus and local interpreation as profoundly enabling as we seek to develop our digital dexterity.

I do think the VALA issues require our attention in sport.

Postscript

Papers for the conference are available at the VALA2020 Proceedings webpage.

Photo Credit

CAUL’s digital dexterity visualisation (link)

Epistemic Vigilance

Introduction

When I was drafting sections of the Sport Informatics and Analytics course on WikiEducator (link), I had an opportunity to look at what is to count as evidence in our work as analysts (link) and to contemplate our epistemic culture, how we come to know (link) and the construction of the machineries of knowledge construction (Karin Knorr Cetina, 1999) (link).

Vigilance

Once we have started to think about our culture as analysts and how we address the construction of knowledge, I think it is a short, epistemological and ontological step to think about epistemic vigilance. Dan Sperber and his colleagues (link) observe “people stand to gain immensely from communication with others, but this leaves them open to the risk of being accidentally or intentionally misinformed, which may reduce, cancel, or even reverse these gains”. Epistemic vigilance seeks to address this misinformation potential.

Gergely Csibra, and György Gergely (2011) (link) discuss the cognitive mechanisms that enable the transmission of cultural knowledge by communication between individuals and regard this as a system of ‘natural pedagogy’. This transmission raises for Louise Gupil (2020) (link) and her colleagues the importance of evaluating the confidence our social partners have in the information they are providing.

This evaluation seems very important when we have an analytics profession that has varying levels of experience. Lisa Scharrer, Marc Stadtler, and Rainer Bromme (2019) (link) note “results of two studies suggest that non experts are not sufficiently aware of the relative importance of source evaluation, which may increase their susceptibility to misinformation”.

Conclusion

Epistemic vigilance is a very important meta-issue for analysts. How we learn to become analysts raises fundamental issues for us. We need to be able to evaluate the information we are given and then apply it in context. Vigilance connects us with our epistemic culture and our awareness of the machineries of knowledge production.

What it is to be an analyst is a technical and a social undertaking. Vigilance suggests to me that we take profound care in what we share and how we learn.

Photo Credit

Gallery (Hernan Pirera, CC BY-SA 2.0)

1...345...201Page 4 of 201
The blog of Keith Lyons, an education technologist from NSW, Australia.
RSS Twitter

Popular Categories