Analytics

Background

In the last two decades there has been a gradual change in how we refer to the observation, recording and analysis of performance in sport.

This Site explores some of the changes that have occurred. We tend to hear and read less about notational analysis now and talk more about analytics. This indicates an important change in the community of practice that analyses performance in sport. Bill Gerard (2015) has provided an overview of these changes.

In 2005, the Journal of Quantitative Analysis in Sports appeared "as the first academic journal dedicated to statistical analysis in sports" (Benjamin Alamar, 2005). Jim Albert (2012) noted in an editorial in the Journal "In 2005, there were 32 manuscripts submitted, and submissions have steadily increased with 106 manuscripts submitted in 2011". The Journal is an official journal of the American Statistical Association. The first volume of the Journal of Sports Analytics appeared in 2015.

The first MIT Sloan Sports Analytics Conference was held in 2007. The goal of the annual conference is to provide a forum for industry professionals and students to discuss the role of sport analytics in the global sports industry. There were 175 attendeesat the inaugural conference convened by Daryl Morey and hosted in classrooms on the MIT campus. In 2013 there were 2,700 attendees and it took place in the Boston Convention and Exhibition Centre.

An MIT report of the 2014 Conference asserted:

The sports analytics revolution has happened. The question is no longer whether to use analytics to measure and monitor team performance and value. The competition from here on out—on the court, on the field, or in the front office—is a matter of how those analytics are used.

There was a Sports Analytics Innovation Summit in London in March, 2014. Since then there has been an explosion of events in this growing field of study.

Introduction

Benjamin Alamar and Vijay Mehrotra (2011a, 2011b, 2012) provide insights into the origins and development of sport analytics.

They define sport analytics as:

the management of structured historical data, the application of predictive analytic models that utilize that data, and the use of information systems to inform decision makers and enable them to help their organizations in gaining a competitive advantage on the field of play. (2011a)

This definition has three components:

Data management
Predictive models
Information systems

In the second of their articles, Benjamin and Vijay (2011b) observe:

as we look at the world of sports today, there appears to be a profound dichotomy: despite a rapidly growing interest in applying analytics, an explosion in data, a plethora of companies peddling information delivery systems and making promises, the actual impact of analytics on the world of professional sports is still somewhat limited.

Benjamin and Vijay (2012) conclude their three part look at sport analytics with this observation about the industry context:

Sports analytics will continue to evolve as a field. The pace of this evolution and adoption, however, will depend largely on how quickly leaders in sports become convinced that significant investments into analytics (data, models, information systems, and skilled personnel) will deliver a true competitive advantage.

Chris Anderson (2014) defines sports analytics as:

The discovery, communication, and implementation of actionable insights derived from structured information in order to improve the quality of decisions and performance in an organization.

Jeremy Abramson (2014) suggests:

A more official definition might be "the discovery and communication of meaningful patterns in data”, which seems reasonable enough. One thing to note, though, is the part about communication … This is actually an important point, and I think the short term future of sports analytics lies less in the development of new mathematical techniques — very few of which are created in sports contexts — and more with coherent storytelling, information distillation and visualization. Regardless of whether you’re a media outlet, the director of analytics for a team or just an armchair statistician, “analytics" doesn’t mean anything if your results only reach as far as your computer screen.

Bill Gerard (2016b) argues for "a narrow definition of sports analytics" as the analysis of tactical data to support tactics-related sporting decisions. He suggests "this narrow definition captures the uniqueness and the innovatory nature of sports analytics as the analysis of tactical performance data."

Online Courses

There is a proliferation of online courses to inform the theory and practice of sport informatics and analytics. Many of these courses are open and charge no fee for service.

Masters Courses

In addition to the many online courses available, there are examples of fee for service courses at Masters level. La Trobe University, in Australia enrolled its first students in 2017 in their Master of Sports Analytics course. The course extends over 1.5 years of full-time study and includes the following components:

Principles of Sports Analytics
Data warehousing and big data
Decision support systems
Object oriented programming fundamentals
Predictive analytics
Visual analytics
Sports data capture
Sports analytics in practice

Year two of the course offers opportunities to select an elective and to undertake two research projects.

Blogging and Microblogging

There is a vibrant community of bloggers and microbloggers writing about the observation and analysis of performance.

Employment Examples

This is a time of growth in employment opportunities data science (Data Crunch, 2017) in Sport Analytics.

Senior Analytics Specialist

An advert for a Senior Analytics Specialist in January 2017 exemplifies some of the contexts in which Anayltics occurs:

The Sports Analytics Team is a multidisciplinary group where all members have a deep knowledge of sports, statistics, databases, programming logic and languages, and storytelling. Combining best-in-the-industry data with advanced mathematics and statistical modeling skills, the Sports Analytics Team has created storytelling tools and metrics that have improved the evaluation of team and player performance ...

The role:

The Senior Analytics Specialist will assist the leadership team in evaluating potential projects by giving their informed opinion. The Senior Analytics Specialist will work with the Director to define project timelines and delegate duties. The Senior Analytics Specialist will be the lead architect in designing team and player performance metrics and their underlying framework. He or she must demonstrate critical thinking about the execution of the projects and keep analytics leadership informed of progress. Completing work on deadlines is a crucial requirement. A portfolio of successful analytics projects and metrics is necessary to be considered for this position. The Senior Analytics Specialist will have the skills to partner with other members of the Sports Analytics Team as well as Stats & Information staffers outside the Sports Analytics Team. Candidates need to have a background in constructing algorithms and complex formulas that explore new areas of analytics, ideally in sports.

The post holder's responsibilities are:

Use advanced mathematics and statistical modeling to formulate sports performance metrics
Help set the direction of the department by partnering with Sports Analytics leadership to set project priorities
Identify the key components in the algorithms and turn them into language that allows them to be readily used in storytelling
Work collaboratively through conflicts and gain support from internal project partners and stakeholders with various interests and priorities
Support the department’s training plan by contributing to the education of other members of the SIG, as well as other members of the Production department, on Sports Analytics storytelling tools
Collaborate with the other members of the Sports Analytics Team and technology partners to write scripts, queries and basic code to transform data into automated metrics across our digital platforms and internal research tools
Identify trends and react to news stories on a daily basis, using the new tools the Sports Analytics Team develops, other products available to the Stats & Information Group and tools available in the marketplace
Analyze third-party statistics providers with an eye toward recommending potential deals, and collaborate with third-party vendors to produce metrics
Use strategic analytical approaches and techniques using data to answer sports questions

The basic qualifications expected:

College degree
Five years’ experience with sports analytics, or any equivalent combination of education and experience that provides the applicant with the knowledge, skills and ability to perform the job
Two years’ experience constructing algorithms to explain or predict performance
Two years’ experience using predictive modeling, statistics, trend analysis, and other data analysis techniques
Two years’ experience in a programming language such as R, PL/SQL or Python
Strong knowledge in using SQL to query and modify data from databases
Ability to help develop data warehouses by identifying the appropriate categories and granularity of content
A passion for sports and statistics; candidates need to have a working knowledge of players, teams and the rules of the games
Full availability for this position, which will include nights, weekends and holidays
Demonstrated ability to ensure accuracy of content
Effective communication and leadership skills
A thorough knowledge of sports statistical analysis in the marketplace
Strong sports knowledge, both historical and current
Bachelor’s degree in a quantitative discipline such as Statistics, Computer Science, Mathematics or Economics.

The post holder will have the following preferred qualifications:

Advanced degree in a quantitative discipline such as Statistics, Computer Science, Mathematics or Economics.
Expert level knowledge and experience with R
Experience turning sports analytics into storylines
Experience working with teams in an analytics capacity
Experience in digital and/or television sports production
Experience writing about sports analytics

Machine Learning Research - Sports Analytics and Predictions

This opportunity is for a position within a company that works to develop and implement statistical models for analysing and predicting the outcome of sporting events. The company uses the signals generated by their models to inform sports betting by professional traders.

The advertisement noted that candidates with experience of modelling sporting events or financial quantitative analysis were encouraged to apply, but candidates were sought who were "eager to take their first steps on a career path, or looking to move from an academic to a commercial environment".

The role:

We are seeking talented and motivated data scientists with a background or experience in machine learning. You should be excited by intellectual challenges and confident enough to contribute ideas and challenge assumptions from your first day. You will work in group of high-achieving quantitative researchers who love sport and who love solving difficult problems. You will be able to demonstrate your abilities through tangible work and discrete projects, with constant feedback from real world events.
Developing and prototyping new algorithms for machine learning training and inference
Applying these algorithms for computer vision and autonomous trading systems
Collaborating with machine learning researchers in leading universities and research labs. We have formal partnerships with 4 universities in the UK
As a research scientist, you will lead a research project to conduct research in a small team, collaborating on global projects in a friendly work environment. Some of the research projects can involve topic modeling, deep learning, discriminative learning, structured predictions, belief propagation for dynamic networks, discriminative learning etc

Candidates were required to have:

Masters or PhD (preferred) in statistics, computer science, engineering or related field
2+ years experience in working with large data sets in a related field (demonstrable)
Experience with a wide range of research designs and statistical methods (e.g., RCT, retrospective cohort studies, reinforcement learning, deep learning, propensity score matching, regression discontinuity, linear and logistic regression, survival analysis)
Excellent written and oral presentation skills
Record of publishing in peer-reviewed journals
Familiarity with Python, PySpark, and machine learning approaches to predictive modeling applied in a Hadoop environment a plus
Excellent knowledge of 3 or more year of experience in Optimization, Machine Learning, or Computational statistics.

Data Analyst Lead

Kelly Burdine was interviewed in March 2017 about her role as a Data Analyst Lead at Hudl.

Kelly arrived at Hudl after six years in the military as an intelligence analyst and a Masters Degree at Creighton University. At Hudl she worked on business intelligence issues (including subscription models and the Hudl Assist service).

Database Developer

In May 2017, the Budesliga advertised the position of Database Developer for the Budesliga Databank. The job description had these elements:

The opportunity that awaited the successful applicant:

Vielfältige Aufgaben zusammen mit dem nach agilen Methoden arbeitenden IT-Entwicklungsteam
Entwicklung von neuen und Weiterentwicklung von bestehenden Datenbanksystemen, insbesondere der
Bundesliga Data Library, als performantes und stabiles System, welches die Grundlage unserer Produkte für Kunden,
Partner und Lizenznehmer der Bundesliga darstellt
Erstellung und Optimierung von Datenbankdesigns
- Analyse und Verbesserung der Systeme für die Echtzeitverarbeitung von großen Datenmengen, insbesondere mit Hilfe von regelmäßig durchgeführten Last-und Performance Tests
Integration von video - und sensorgestützten Daten, sowie die Anwendung von Big Data Science
Methoden in Zusammenarbeit mit der Data Analytic Abteilung.

The experience expected of applicants (Das hast du im Gepäck):

Ein abgeschlossenes Informatikstudium oder eine vergleichbare Berufsausbildung
Sehr gute Kompetenzen in der Datenbankentwicklung - und administration von relationalen und NoSQL
Gute Kenntnisse im Umgang mit DB2, Informix oder Oracle Datenbanken sind von Vorteil
Gute Kenntnisse und sicherer Umgang mit höheren Programmiersprachen, wie z.B. C# und Java, sowie mit gängigen Entwicklungswerkzeugen (wie z.B. Eclipse und Visual Studio)
Know-How in der testgetriebenen Entwicklung
Sehr gute Deutsch -und gute Englisch kenntnisse in Wort und Schrift

Other requirements (in addition to enthusiasm for the opportunity):

Ein sehr guter Teamplayer mit hohem Verantwortungsbewusstsein und einer hohen Problemlösungskompetenz bist
Zielorientiertes und selbstständiges Arbeiten mit Liebe für das Detail mitbringst und dabei eine hohe
Eigeninitiative zeigst
Erste Erfahrungen mit agilen Entwicklungsmethoden, wie Scrum, gesammelt hast
Basiswissen im Bereich Bundesligafußball und deren Regeln besitzt

What the successful applicant could expect of the employer:

Werde Teil der Erfolgsgeschichte der Bundesliga
Ein kollegiales und hochmotiviertes Team, das täglich sein Bestes gibt
Flache Hierarchien, offene Kommunikation und flexible Arbeitszeiten
Verantwortungsvolle und abwechslungsreiche Aufgaben
Kurze Entscheidungswege
Eine Unternehmenskultur, die durch Toleranz, Respekt und Wertschätzung gekennzeichnet ist

Head of Sports Analytics

The Aspire Academy in Doha advertised the position of Head of Sports Analytics in June 2017. The job description included this information:

Responsible for analytical templates and responsive to requests, evidence, and data to support and define operational and strategic direction of the Department of Sport and Sports Science. Analyses the needs of the departments involved in sports and develops, implements and documents processes policies and procedures relating to data.

•Leads and coordinates the Data Operators.

•Directly reports to the Director of Sport and the Director General for special projects.

•Serves as a focal point for all data management applications and projects.

• Develops automatic reports from database to serve the needs of each department and the top management.

•Develops and maintains handbooks in terms of processes and procedures for the various projects to enhance transparency.

•Sets motivating and challenging objectives for the staff. Gives feedback to the staff and motivates them to continuously improve their skills and competencies.

Applicants:

•Bachelor/Master degree in sport management or a related field is a plus.

•3-5 years professional experience in a similar position, preferably in the same industry.

•Preferable access to a wide network in sport business.

•Conceptual working ability.

•Knowledge of common football (management tools).

•Proficient in oral and written English.

•Knowledge of additional languages, especially Arabic is a plus.

•Technical understanding and abstract thinking.

Data Scientist

In July 2017, PUSH advertised for a data scientist to join the company. The advertisement included the following:

The Data Scientist is a key role at PUSH, you will be working closely with our product team to help implement machine learning techniques to improve our product's accuracy, reliability, and user experience. You will be working closely with everything from our raw motion data (Terabytes worth of data) to user meta data (15 million repetitions and counting). You will work closely with our Director of Sport Science to learn more about the domain and ensure that everything you do has the sport science stamp on it. You will also work closely with our Algorithms Engineer to ensure that you have a good understanding of how the raw signal data collected is transformed into meaningful meta data for coaches and athletes to use in their training.

At PUSH, you will:

Get to solve really really challenging problems for elite athletes and coaches all over the world.
Work with world-class sports science researchers on not only testing existing hypotheses in the space but developing new paradigms of thinking.
Work closely with developers who are very passionate about data and analytics.
Make a real impact at the highest level of sport for athletes and teams across the NFL, NHL, MLB, and NBA.

You have:

Graduate degree in Biomechanical/Biomedical engineering, Robotics, or Biomechanics
Extensive experience working with time series motion data
Applied multi-body modelling in different dynamical systems
Familiarity with human movement patterns and motion characteristics
Strong knowledge of Bayesian theory and probabilistic estimation methods
Experience in machine learning for analytics and prediction
Experience implementing predictive dynamics methods for different applications
Competent scripting skills in R and/or Python.
Experience with Agile development methodology
Demonstrated strong ability to deliver in a fast-paced environment

Nice to haves that we'd love to see:

Enjoys reading sport analytics (e.g., fivethirtyeight)
Worked in a tech startup
Know your way around C++

Lead Data Scientist

In September 2017, Liverpool FC advertised for a Lead Data Scientist to be based at the club's Melwood training ground. The advertisement noted that the successful applicant "will create, implement and manage mathematical modelling research on tracking data, and develop products and visualisations for analysts, coaches and other key stakeholders".

Role Accountabilities

Project managing research programmes, from conceptualisation to delivery;
Leading the development of mathematical and statistical models to make performance predictions and recommendations for analysts and coaches;
Leading and managing the implementation of appropriate performance tools and models, including creating visualisations for communication purposes;
Researching, identifying, reviewing and assisting in the implementation of new technologies aimed at improving the Club’s research programme;
Delivering pre-match and post-match performance analysis, including player performance information and data management;
Liaising with colleagues in relation to use, effectiveness and appropriateness of performance tools and models; and further development/review of such tools and models.

Personal Qualities

The successful candidate will have an excellent academic record, holding a PhD and a first degree in mathematics or physics. They will have extensive experience in developing mathematical models, preferably Bayesian statistical models.
Experience must also include developing predictive models using football tracking data, with a demonstrable record of high quality research in the field.
A high degree of competency in numerical computing, with experience in Python, R and/or commercial numerical libraries is essential.
The successful candidate also have excellent skills using data science tools such as MapReduce and Keras. Preferred candidates will have experience with database management e.g. MySQL, object oriented software development e.g. C#, and experience with web development e.g. JavaScript libraries such as d3 and three. Experience of working with Sportscode is an advantage.
Excellent English language skills are necessary
- Candidates who are invited for interview will be asked to present examples of their tracking data analytics projects.

Sports Data Analyst Internship

In September 2017, Third Fuse Media advertised two paid internship positions. The twelve-month positions offered successful applicants "an opportunity to assist in the collection, verification and analysis of large amounts of data emanating from a particular sporting code". The advertisement noted:

The 12 month internship, with succession planning in mind, ultimately aims to produce a skilled technical expert who will be able to provide reliable statistical analysis, data interpretation and scenario outcome projections.

The responsibilities of the interns:

Collection, verification and analysis of Soccer / Football data
Manage the integration of current analytical tools into a central data management system
Experiment with new and creative performance metric comparisons
Perform regular correlative data studies looking for trends in the performance of Football /Soccer teams and players, both individually and collectively.

The required skills expectations of the successful applicants were:

Diploma in Sports Science / Sports Management or any other relevant qualification
Passion for Sport and Data Analytics (with a particular emphasis on Soccer / Football)
Maintain highest level of professionalism within a professional sports environment
Dedicated work ethic and an open mind
High levels of competency and proficiency with Computer hardware and software
General knowledge of Sports and familiarity with Sport data trends
Strong organizational skills, time management skills and attention to detail
Ability to work and adapt in a fast-paced and flexible environment
Ability to work independently and take initiative where necessary to complete tasks
Passion for developing methods to streamline processes and data flow through automation
Comfort with ambiguity in data

Senior Insights Researcher

In September 2017, High Performance Sport New Zealand (HPSNZ) sought to recruit "an academically strong applicant, with strong research experience, familiarity with a range of data storage and analysis systems' and a passion for excellence in sport performance to join our 'Knowledge for Tokyo' research team".

The position particulars included:

The team is dedicated to capturing and analysing the preparation and performance evidence of how New Zealand athletes, coaches and support prepare for competition at the highest level. We record and explore data across NZ Olympic sports and key targeted sports as athletes, coaches and support team members prepare for and compete in pinnacle events. We support sports by helping to accelerate learning, uncovering what it takes for our athletes to win at events through to the Tokyo Olympics and beyond.

In collaboration with National Sport Organisations, we perform quantitative and qualitative research and analysis to uncover patterns in how people perform, behave and think. We provide insights back to New Zealand's best coaches, athletes and sports leaders.

The skills you bring are important to help New Zealand succeed. You will have strength in research and analysis systems. In this role you would help capture and explore preparation and performance data. You will have a strong qualitative research background with some experience of quantitative analysis. Alongside experts in their field you will help to build, refine and utilise analysis tools, and will lead the development of research methods and reporting/visualisation processes to advance evidence-based best practice within HPSNZ.

This is a full-time role through the Tokyo Olympics to December 2020, however there is potential for the position to become permanent through to Paris 2024.The role will be based at either Auckland, Cambridge or Dunedin. If you have a strong research and reporting background, If you are driven by improving performance and would be proud to help our kiwi athletes gain a competitive advantage then this is an outstanding professional opportunity for the successful applicant.

An advertisement for a Senior Insights Researcher appeared in January 2018. Position particulars included:

Functions of the role

· Designing and implementing short-term and longitudinal research projects to capture qualitative and quantitative data to uncover insights and patterns to inform pinnacle event preparation.

· Uncover and validate patterns, behaviours and interventions which best support elite sport performance and accelerate sport development.

· Conduct qualitative analysis on semi-structured and unstructured text, utilising manual coding and machine-learning methods.

· Utilise a SQL database to store, clean, combine, augment, analyse, and interact with data; including sourcing and incorporating external data sets.

· Utilise data visualisation tools to present complex stories and produce research summaries and reports that clearly communicate key findings and trends.

· Research existing qualitative and quantitative data and related literature for emerging patterns and insights.

Commitments

· Collaborate effectively with stakeholders (National Sport Organisations, HPSNZ, NZOC, Paralympics NZ, external contractors).

· Work within the code of ethics, conduct, standards and confidentiality, relevant to HPSNZ.

· Undertake any other reasonable duties, as requested.

· Comply with HPSNZ Health and Safety policy, procedures and safety instructions at all times.

Experience, Skills and Knowledge

· Strong curiosity and passion for high performance sport.

· A comprehensive knowledge and analysis metholodgy with extensive experience in qualitative and quantitive research.

· Proven background in effectively reporting and articulating complex insights into actionable findings.

· Strong passion for New Zealand’s international sport reputation.

· Effective with various software platforms in provision of research and analysis outputs.

· Demonstrated ability to interpret scientific information and large data sets.

· Proven ability applying analysis and reporting outputs in a meaningful way.

· An understanding and working familiarity with database design and platforms that support qualitative and quantitative research (i.e. nVivo, SQL, R, etc).

· An understanding of critically appraising investigative methodology and data.

· An ability to communicate complex findings through intuitive and engaging media including visualisations and written reports.

Qualifications

Essential

Graduate degree and experience in a research based field:

Computer Science, Statistics, Sport Science, Psychology, Social Science etc.

Data Scientist - Physiological Signals

In October 2017, Firstbeat, a company that offers "physiology-based solutions for performance and well-being", advertised a position for a data scientist that "involves using your analytical mindset to process and visualize large amounts of physiological signals monitored during different settings like exercise lab tests, sports, sleep and ordinary daily life to gain understanding on the physiological phenomena".

The position description included these role expectations:

Experience on time series analysis and signal processing as well as applying Python or MATLAB in your analyses.
C/C++ skills are a plus.
You may have background or degree in physics, applied mathematics, technical engineering, computational sciences or another applicable field.
Strong ability to present and communicate data findings in an understandable manner.
Fluent in written and spoken English.
Ability and motivation to learn new things to further develop your professional skills.

Performance Intelligence and Analysis Lead

The Lawn Tennis Association (LTA) advertised for a Performance Intelligence and Analysis Lead in November 2017. The LTA were seeking:

a positive and proactive individual to join the Performance Science and Medicine Team at the LTA in a pivotal new role created as a key part of the LTA new 10-year performance strategy for British Tennis.

The position:

will have wide reaching impact throughout the Performance Directorate, developing the use and integration of data driven intelligence to directly influence decision-making, coaching and player development and education programmes with the aim of making the LTA the most respected pathway for player development in the world.

The successful candidate:

will have significant experience in leading a team of Data Scientists, Notational Analysts, PhD staff and external partners in the development of world leading intelligence and insight within High Performance Sport. They will also have experience of managing a department through an evolutionary change in strategic direction.

The role required:

someone who is highly organised with proven experience and expertise in sport or business analysis. They will have excellent communication skills (written and verbal) with outstanding attention to detail and ability to adapt and prioritise workflow to meet the needs of the team.

The job description included these key responsibilities:

Management: Analysis Team, Systems, Processes and Relationships

Inspire and effectively line manage a team of Performance Analysts, Data Scientists PhD and Interns, aligning and prioritising work streams, providing guidance, education, development opportunities and promoting a passionate and enjoyable working atmosphere.
Create world leading solutions to ensure infrastructure, environment, processes and staff roles and responsibilities are optimised
Introduce a regular process of review and development to ensure continuous process of improvement and to ensure work programmes and outcomes remain tightly matched to the Performance Directorate strategy.
Build strong relationships of high influence and impact with British Tennis players, Coaches, Support Staff, external stakeholders and other tennis and non-tennis National Governing Bodies.
Develop a cloud based database and platform to house and aggregate data from multiple sources in a future proofed manner through working closely with the Head of Performance Science and Medicine, IT department and identified external partners.
Work with the Head of Performance Science and Medicine in the establishment of relationships with subject matter experts in the areas of data management and data analysis, including Universities, Companies, Partners and competing Nations.

Delivery: Player and Coach Support, Education

Strategically lead in the delivery of services to supported players, including direct analysis support of players where work flow demands dictate
Review and evolve all elements of player support to ensure the Analysis Team deliver a world- leading, sustainable support programme.
Act as principle point of contact for internal stakeholders and Elite Coaches and Players, engaging in a frequent process of open dialogue to optimise relationships and explore new and innovative methods of developing the impact of Analysis for each player and coach
Contribute to a world-class inter-disciplinary education programme for players, coaches and parents through the Pathway, building the highest levels of player autonomy and ensuring the development of self reflective players. This will include delivery of internal CPD, National Conferencing and the production of virtual education
Undertake appropriate professional development to keep abreast of world’s best practice.

Strategic: Data Science, Data Presentation and Departmental

Work in close collaboration with the Head of Performance Science and Medicine to continue to evolve a detailed strategic vision and programme of work to efficiently support the Performance Directorate, building a world leading model of critical determinants of success in high performance tennis
Align the work programmes of a full-time Data Scientist (to be appointed) and PhD student and oversee short and long term projects that deliver the highest levels of inteliigence and insight into the determinants of success in high performance tennis
Communicate exceptionally, translating data driven intelligence into game changing information that resonates with sportspeople and stakeholders and transforms our understanding of what it takes to win
Ensure project outcomes are reported in a meaningful and useful manner for the customer. This will include identifying significant trends and factors with predictive validity, and relationships between them related to athlete development and progression.
Develop world class methods of displaying data by working in collaboration with the Analysis Team, IT department and external partners.

Analyst, Sport Analytics

In November 2017, Canadian Tire Bank advertised for an analyst, sport analytics post. The advert was:

Position Summary

Canadian Tire Bank (CTB) has partnered with Own the Podium (OTP) and the Canadian Olympic Committee (COC) to provide advanced Sports Analytics for Canada’s athletes. CTB is looking for talented individuals to help leverage analytic insights to support Canadian athletes at upcoming Olympic events. This position will report to the Manager of Sports Analytics. Primary responsibilities for the position of Analyst, Sports Analytics will focus on analysis of sports performance data, interpretation of athlete performance through the creation of statistical models and relevant metrics. The position also requires research of sport-specific details and impacts on analyses, creation and validation of datasets and the creation of athlete development tools to leverage analytical results.

Position Opportunities

The Sports Analytics team at CTB is a new, exciting and unique opportunity to support Canada’s Olympic effort. Analysts will apply their analytical and modeling skills to athletic performance data and provide insights to Own the Podium and the COC. Analysts will develop an understanding for performance-based milestones for Canada’s Olympic athletes and how these milestones are produced. As a team member of Sports Analytics, you will have the opportunity to create interactive tools driven by analytical results to assist Canada’s athletes own the podium.

Responsibilities

Conduct exploratory analyses on national and international sports performance data to identify sport-specific patterns, athlete and team strategies, and future trends.
Create and maintain statistical models to predict future athlete performance and development of training milestones.
Develop data-driven metrics to assist in interpretation of athlete results and implementation of performance goals.
Research and document sport-specific issues and evaluate potential impacts to analytics.
Investigate potential sources of performance data and data collection techniques.
Construct and maintain athlete performance visualization tools and provide support to Sport representatives.
Consult and present analytical results and insights internally, with OTP and Sport Organizations.

Qualifications

University degree in Mathematics, Statistics or analytical discipline.
Demonstrated superior analytical, mathematical and problem-solving skills.
Past experience of developing statistical models and predictive methods.
Proven ability to translate business goals into analytical results.
Programming experience, especially with data extraction/mining tools (e.g. SAS/SQL/R/Python/VBA).
Experience building reports and visualizations in Tableau or equivalent.
Good communication skills both oral and written.
Passion for Olympic Sport in Canada.

SQL/R - Insight Analyst

In January 2018, a professional sports team in England advertised for an SQL/R- Insights Analyst. The advert included this information:

This is an outstanding opportunity to join a leading sports team as an Analyst in their growing analytics division. You will be coming in at a lead level, playing a critical role delivering insight on player performance and success. You will have the chance to work in a fun, ambitious and fast paced environment, building on your technical skills, whilst working with senior stakeholders.

SQL/R - INSIGHT ANALYST - WHAT TO EXPECT

You will be working in an exciting, driven analytics team utilising your SQL skills, as well as getting the opportunity to use Python and R to analyse athlete performance and provide insight and recommendations
As the Senior Analyst, you will be responsible for leading and working on a range of statistical projects, designing algorithms and reporting on data
You will have the opportunity to work with teams and stakeholders across the business as this team delivers a range of actionable insight!

YOUR SKILLS AND EXPERIENCE

Educated to degree level - minimum 2.1 in maths , stats, economics etc.
Experience working with data using one of the following SQL and VBA
Experience working with stakeholders and delivering strategic recommendations
Interest in Sport is required!
Excellent communications and interpersonal skills

Data Scientist

Catapult Sports advertised for a data scientist in January 2018. The advertisement contained this information:

COMPANY SUMMARY

Catapult empowers elite coaches globally with scientifically-validated metrics for the advancement of athlete performance. The company engineers wearable technology that provides objective information behind athlete risk, readiness and return to play. Born out of the Australian Institute of Sport (AIS) and a scientific research organization, Catapult now works with over 1250 elite teams and institutes around the world and is based in Australia, the US and the UK.

We offer software, analytics and services that enable sports organizations at all levels to better scout, recruit, teach and win. Our products can be found internationally at practice facilities and arenas, in meeting rooms and on the road, serving premier teams in the NFL, NBA, NHL, MLS, NCAA and more.

At the heart of our continued growth and success is our employees. Our experienced team, comprised of the industry's brightest minds, is equally as passionate and competitive as the organizations we are proud to call our partners.

POSITION SUMMARY

The purpose of this role is to create and deliver innovative solutions in how data shapes athlete performance. The role will complement our strong data science skills in data visualization, machine learning and applied sport science with a focus on a software engineering background.

The role will work closely with the relevant sports scientist who will define the problem and collect the data. Following the initial analyses, the solutions will then be prototyped and tested. Once fully validated, these solutions will be implemented in our real-time firmware or software applications which sit in our technology. Initial analyses and prototyping of solutions may be conducted using R with the final code to be written in C++ or python.

ESSENTIAL DUTIES

Identify, develop, prototype and validate new client-focused data-driven tools and solutions
Facilitate the development of prototypes and their implementation into our technology. Employ programming skills to help to translate algorithms implemented in high level, mathematically oriented languages such as R into object oriented C++/Python
Each data scientist will report to their analytics team leader, through to the CEO – Rest of World

DESIRED SKILLS AND EXPERIENCE

Minimum of 2+ years of recent professional industry experience in data science
Expertise in statistical modelling, including machine learning, with large data sets is highly preferred
Experience with data science tools R or Python strongly preferred
Hands-on experience with at least one database querying language like SQL
Proficient with small and/or big data modeling work
Experience with version control system like GitHub
Communicate concisely and persuasively with engineers and product managers
Great communication skills to be able to articulate your findings, or the approach that you used, to both technical and non-technical audiences
Passion to answer Product/Engineering questions with data
Comfortable working in a fast paced, highly collaborative, dynamic work environment
The desire to directly influence the performance management of elite sports
Ability to work well in a small group which is part of a large international team
Ability to think creatively and solve problems
Ability to spot patterns in the data and strong attention to detail
Bachelor’s degree in a quantitative field such as Computer Science, Statistics, or Mathematics is required
Postgraduate degree (Doctorate or Masters) is highly desirable.

Data Scientist (Football)

Arsenal Football Club advertised for a data scientist in February 2018. The position description was:

JOB PURPOSE

To work as a data scientist at our London Colney training centre. Main duties will be analysing data from tracking systems, GPS training data and an increasing number of wearables and diagnostic technologies to improve injury prevention and performance optimization. The role will work directly with our sports science department as well as our analytics group.

KEY RESPONSIBILITIES

Statistics

Apply statistical rigour to current sports science, analysis and coaching practices

Modelling

Build statistical models using geo-spatial, medical and fitness data in order to assist AFC High Performance Team with, amongst other tasks, injury prevention and performance enhancement
Understand needs of football staff and implement in the modelling
Communicate clearly and concisely results of work
Write code

Data cleaning and maintenance

Clean, manipulate and merge data from a variety of sources.
Perform extract/transform/load (ETL) tasks

Visualizations and reporting

Provide user friendly access to reports from modelling that update automatically as new data becomes available
Build simple visualizations using Tableau, Shiny, d3.js, or other tools

MAIN JOB REQUIREMENTS AND PERSON SPECIFICATION

Education/Qualifications/Training:

Master's or PhD in a quantitative field: Computer Science, Engineering, Physics, Applied Mathematics
Experience in elite sporting environment would be desirable

Abilities/Skills/Knowledge:

The candidate should be proficient in a statistical programming languages and appropriate packages (Python is preferred but R is acceptable).
The candidate should feel comfortable performing ETL tasks, data wrangling, data cleansing and feature engineering
The candidate should have experience modelling rare events and dealing with sparse or noisy data sets
The candidate should be knowledgeable of methods for regression, classification and time series analysis
The candidate should have basic data visualization skills and be able to clearly and concisely communicate their findings to stakeholders with varying technical backgrounds. They should be able to explain the pros and cons of the chosen model as well as alternative, variable importance and goodness of fit criteria.
It is preferable if the candidate has experience with large scale data science technologies like Spark (PySpark is preferred).
Candidates with experience analysing geo-spatial data (including sports tracking data) will receive priority consideration.

JOB EVALUATION FACTORS - (examples of)

1. Complexity and decision making

Quality of modelling - presents alternative approaches to modelling and explains why the final method was chosen including appropriate error measurement techniques.
Understands the different sources of data we can use, how to correctly merge them and remove data inconsistencies and do so in an efficient, repeatable and well documented manor
Delivers models and accompanying reports/visualizations that can be easily accessed and understood by practitioner.

2. Delivering results - commercial awareness, business and strategic impact

Jobholder is charged with ensuring prompt analysis and delivery of relevant information with the goal player, team and club high performance.

3. Freedom to act and accountability

Makes appropriate tradeoffs on own among competing priorities/models asked for by various departments and stakeholders, consulting with supervisors as needed.
Can bring work to completion in a reasonable amount of time
Makes appropriate tradeoffs between efficiency of work and gains in accuracy.
Communicates status of work frequently to supervisors, updating as deadlines shift.

4. Networks, Relationships and Teams

Communicates in a way in verbal and written format that is comfortable for coaches, backroom staff as well as analytics staff.
Respects and draws out expertise of all different areas of the football staff.
Persuades others about his results in a way that is accessible for the quantitative comfort of the given audience
Thoughtful yet tenacious about dealing with objections from others about their work.

Junior Sport Intelligence Analyst

In March 2018, UK Sport advertised for a junior sport intelligence analyst. The advertisement:

The Junior Sport Intelligence Analyst will:

• Support the Analysis team on the development and maintenance of analysis, results interpretation and reporting, to ensure the Performance Directorate can reliably monitor World Class Programmes (WCP) and their competitors and enable UK Sport and Sports to make better informed decisions.

We are looking for someone who has the following skills/experience:

• Support the delivery of Sport Intelligence (SI) outputs and models through the creation and automation of reports and dashboards

• Liaise with BI developer to ensure that data for outputs is robust and clear

• Support the delivery of statistical research projects including the parameterisation of problems, methods of analysis, description and interpretation of results.

• Ensure that analysis is understood by both internal and external stakeholders. Deliver post analysis presentations to ensure clarity to stakeholders

• Report project outcomes in a meaningful and useful manner to relevant stakeholders. This will include identifying significant trends and factors with predictive validity and relationships between them, related to athlete development and progression.

• Work with the other Sport Intelligence team members on the design and management of methods for reliable data collection, manipulation and presentation from a range of fields such as competition results, athlete demographics, athlete training data, injury surveillance, and management accounts.

• Support projects to develop the analytical capability of WCP by consulting, solving and implement solutions using best practice analytical and management techniques.

• Proactively contribute to the development of Sport Intelligence strategy to objectively monitor WCPs and international high performance environments.

• Support the analytical team’s strategy by developing and maintaining documentation such as project trackers, assumptions logs and quality assurance for own work.

• Help the Sport intelligence team guide and advise GB High Performance staff in the use of data and analysis.

Research Fellow - Sport/Data Analytics

In April 2018, the University of Michigan advertised a research fellowship opportunity. The advertisement for the position was:

At the University of Michigan (UM), the Exercise and Sport Science Initiative (ESSI: www.essi.umich.edu) in partnership with UM Athletics seeks a highly motivated postdoctoral research fellow to focus on sport and performance data analytics development and integration.

Responsibilities

Responsibilities will include but are not limited to:

Design, develop, test, implement, and maintain predictive models and metrics using appropriate tools and techniques
Work with ESSI and Athletics to integrate new statistical analyses, models, and data visualizations into existing and new applications
Keep up-to-date on new predictive modeling techniques and evaluate their potential for application to Athletics’ data sets for multiple sports, teams, and student-athletes
Collaborate with UM/ESSI researchers and UM Athletics personnel (e.g., Senior Associate Athletic Director & Student Athlete Health & Wellness Officer, coaches, athletic trainers, and data analysts) to design and implement statistical analyses
Collaboratively implement new methods and technology into the student-athlete and team development processes to optimize player and team performances and reduce injury
Work with data analytics experts at the UM and with Athletics to research, develop, and test predictive models as well as bring objective measures to assess student-athlete performance and reduce injuries
This position will not entail developing a formal curriculum and teaching regularly scheduled classes, but the candidate may be asked to occasionally present at selected events

Required Qualifications

A PhD or similar degree (ScD) in a computational field such as: data science, mathematics, engineering, quantitative social sciences, quantitative sport science, or analytics
Strong knowledge of statistical analysis, machine learning, and predictive modeling
Demonstrated experience with statistical software (e.g., R and Python) and database querying (SQL)
Ability to communicate effectively with UM Athletics staff
Experience with spatiotemporal analysis or Bayesian statistics—preferable, but not required
Experience with wearable sensor data (e.g., Catapult)—preferable, but not required
Understanding of typical athletic and performance data structures, plus knowledge of current sports performance research, statistics, and strategy
Experience or strong interest in formulating, writing, and managing research grants/contracts
Ability to communicate complex ideas to non-technical audiences using data visualization
Strong interpersonal and written/verbal communication skills and experience—the candidate will play an integral role in and serve as a leading member of a collaborative team, including individuals from ESSI, UM Office of Research, Athletics, and various schools and colleges on campus

Additional Information

The Postdoctoral Research Fellow will be appointed in 1 year increments. Funding for a 1-year postdoctoral fellow position is immediately available, with the potential to extend the appointment for a second year, depending on availability of funds and performance in the first year. The decision to extend for a second year will be made six months into the first year of funding.

Analytics

Background

Introduction

Online Courses

Masters Courses

Blogging and Microblogging

Employment Examples

Senior Analytics Specialist

Machine Learning Research - Sports Analytics and Predictions

Data Analyst Lead

Database Developer

Head of Sports Analytics

Data Scientist

Lead Data Scientist

Sports Data Analyst Internship

Senior Insights Researcher

Data Scientist - Physiological Signals

Performance Intelligence and Analysis Lead

Analyst, Sport Analytics

SQL/R - Insight Analyst

Data Scientist

Data Scientist (Football)

Junior Sport Intelligence Analyst

Research Fellow - Sport/Data Analytics

Recommended Reading

Suggested Reading