HSE311 - Applied Sports Science 1: Performance Analysis in Sport Assignment

Assignment Task

Introduction

This practical provides an introduction to a component of performance analysis in sport. This is an opportunity for you to explore the Dartfish tagging software and experiment with different coding or ‘tagging’ approaches. Dartfish is one type of notational analysis software available, there are certainly other common software available (e.g. HUDL SportsCode, LongoMatch). In this practical we will introduce you to important concepts and skills that are relevant to notational analysis; these skills are not limited to using Dartfish. It is intended that, upon completion of this practical, you will be able to apply these skills to other types of notational analysis software. It is not meant to be a comprehensive guide to notational analysis – rather a brief introduction to give you some understanding of tagging and a chance to experiment with the software.

Tagging player actions/events in a video file is an important process and it is imperative as a sport scientist that you have a good understanding of how this is done, and the relevant software that is commonly used by professional performance analysts. In the evolving area of sport science, more specifically performance analysis, there has been a growth in the collection of data. This results in large datasets that require a specific skillset to manage, manipulate and analyse. The second part of this practical will teach you the basic skills that you will need to be able to analyse these types of datasets. This is not designed to be an advanced analytics unit and therefore we will only go through the fundamentals.

Aims

At the end of this practical you should be able to:

• Understand how notational analysis can be applied in sport
• Create and use a tagging panel within Dartfish software
• Export the results of tagged video for later use/analysis
• Prepare you for Assessment Tasks- (notational analysis section) and 4 (final exam)

Aligned Assessment Tasks

This Practical Class will prepare you for:

• Assessment Task 1 (notational analysis section); and
• Assessment Task 4 (Final Exam).

Pre-Practical Tasks

Online module 2.2 on notational analysis will provide you with the background knowledge to undertake this practical class and give you an understanding of the basic functions of Dartfish.

Tasks

Step 1: Understanding Tagging and How to Tag in Dartfish

In Dartfish, the process of “tagging” video is used to create a searchable index of events that happen in a game. The events are “tagged” with a set of labels to identify what happened. This tagging allows for:

• Notational analysis – as events are coded throughout the game, tagging does this using buttons which are clicked to code events
• Video indexing – as events are coded in the video (with a timestamp) they can be found and replayed easily

This lets the analyst code information about events and also find video footage that can be reviewed quickly. For example, if an analyst of a hockey game wanted to find out where the team tends to make ineffective passes, the analyst could code all ineffective passes in a game for location and could also go back and replay and analyse all these passes.

The best resource for understanding how to tag in Dartfish are the Dartfish online tutorials. Here you will find a set of tutorials on tagging, and specifically, how to create a tagging panel and tag a video clip. Review these pages and tutorials before progressing to the next step . It is also worthwhile spending some time to familiarise yourself with the Dartfish user guide, there is a chapter dedicated to tagging (chapter 9). Please note, the newest version of Dartfish has some minor changes that mean the most up to date user guides and tutorials are not identical. To navigate to the tagging panel and enable you to complete tagging of video, you will need to select the ‘Replay’ tab on the right hand side of the screen.

Step 2: View an Already Created Tagging Panel

To get an idea of what a tagging panel looks like, there are some examples provided for different sports within Dartfish. Keep in mind these are simplistic examples and do not employ the full capacity of what a tagging panel can achieve. Your task is to load in some video and trial one of the built in Dartfish tagging panels, trialling the process of tagging events and how this works (i.e. the different buttons, events etc.).

Today you will analyse an international netball match between Australia and New Zealand at the Constellation Cup. First, you will need to download and load the relevant video clip into Dartfish (i.e. Australia versus New Zealand.mov’ available on Cloud Deakin). There is a Dartfish tutorial on importing video into the software, and the Dartfish user guide also has a section on loading and playing video for tagging (section 2.2). Throughout today’s practical

class you will need to refer to the user guide to assist and troubleshoot. In most instances, the user guide will be able to solve any issues that you encounter. To start with, trial one of the pre-existing tagging panels within the Dartfish software.

After trialling the existing tagging panels, inspect them and explore their use by analysing a small portion of the loaded video and answer the following questions.

1. How are the tagging panels laid out? What characteristics (visual, organisation etc.) of the tagging panel make them easy to use? Is there anything you would change about the tagging panel to make them better?

2. What are the different button types used in the tagging panels, and what do these provide (tip: click the different buttons and view the table output at the bottom of the screen)? What keywords do the different buttons use, and how do these work?

Step 3: Create a Basic Tagging Panel

Now that you have viewed some example tagging panels, you can see that none of them are appropriate for our sport of interest, netball. Your next task is to create a tagging panel to analyse the performance for both teams in our netball match. Before creating your tagging panel you should think about the aspects of a netball match that you could effectively tag. A useful method of doing this is to map out your ideas of what may be important to performance and outcomes in netball and then consider how you could tag these events using your own panel. Remember to include as much detail as possible to characterise the scoring opportunities and outcomes within the game – consider such factors as the team taking the shot, where the shot is taken from and the outcome of the shot. Once you have created your tagging panel, use it to code the video.

3. What ‘events’ did you include in your tagging panel? What button types did you use to code these events?

4. What additional characteristics (i.e. keywords) did you tag for the ‘events’? What button types did you use to code these additional characteristics? Describe the tagging process you used to code events with these characteristics (e.g. what order did you press the buttons in).

5. Explore your events list at the bottom of the tagging Can you edit this if you’ve made a mistake or missed something (Note: check section 9.6 in the Dartfish user guide)? What happens if you double click on an event you’ve tagged? Why would this function be useful?

6. How well did your tagging panel perform? How did you organise the panel to make the job easy (or not easy) for yourself? What would you do differently if you were to create this tagging panel again? It is quite likely through several iterations you would create a more efficient panel that effectively capture the aspects of performance you are most interested Keep this in mind, and remember that we are not expecting you to create brilliant tagging panels on your first attempt.

Step 4: Become a Notational Analyst

The next step of today’s practical is to put all of the previous steps into practice and use notational analysis to answer a specific question. As part of this simulated scenario, you will become a performance analyst for an Australian Rules football club and use notational analysis to code events relevant to a specific scenario.

You are a sports scientist working for a company that provides performance analysis services to elite Australian Football League (AFL) teams and the media. You have been asked to provide analysis of an AFL match between Melbourne and Geelong, and tasked with completing an analysis via notational video-coding.

Download the video file Melbourne vs Geelong Q4 from the practical folder on Cloud Deakin and load this into Dartfish for tagging. A tagging panel will need to be created to code events and outcomes for each team . The overall goal is to analyse the scenario with relevant statistics or performance outcomes. Consider the choices you make with what you tag and ensure that it relates to the scenario and performance.

The scenario

A key area of interest is the outcome of forward 50m arc stoppages (i.e. when play is reset by the umpire by a boundary throw-in or ball-up). Investigate the core actions/events, their outcomes and other descriptive factors that may contribute to the outcome (i.e. number of players around the stoppage, relative player count around stoppage, e.g. +1, when home team has an additional player). There are many factors that you could investigate! Take your time and think about how you can best capture information to answer the question.

Create your tagging panel and use it to code the video.

1. How well did your tagging panel perform? How did you organise the panel to make the job easy (or not easy) for yourself? What would you do differently if you were to create this tagging panel again from scratch?

Step 5: Export your Results

Now you need to export the results from your tagging so that they could be analysed further.

First, we will work through a way to publish a specific event(s) from the tagged events. Dartfish provides the option to publish events (one or multiple) as a video or ‘highlights’ file. Review section 9.8.3 of the Dartfish user guide for details on how to do this, and select one or multiple events to publish.

7. How could this ‘highlights’ publishing feature be used in an applied setting to communicate performance to players or coaches?

Second, we will work through a way to publish all of the tagged events and associated information (i.e. keywords etc.). Dartfish provides an option to export the events list as a CSV file (i.e. openable in Microsoft Excel). Review section 9.8.6 of the Dartfish user guide for details on how to do this, and export your events to a file. Once you have done this, clear all of your tagged events and run through the process of importing your events back in.

8. Why is having the capacity to import events back into Dartfish important?

View your exported events in Excel to get an idea of what this looks like (Note: it should look very similar to your tagged events table in Dartfish). The next step in the notational analysis process is to extract some relevant performance metrics (i.e. performance indicators) out of such an events file, and we will be addressing how to do this in our seminar this week .

Note: You will only have access to Dartfish during your practical classes, so it is imperative that you save your work during this time. Once you have saved your working following the instructions to export (follow the procedures stated in the Dartfish User Guide, Section

9. Producing and sharing tagging resources on page 229) close the program and reopen it, to ensure it is a working file.

Part Two: Filtering, manipulating and analysing a sports performance dataset

In today’s practical class, you have created a tagging panel and coded actions/events in both netball and Australian Football. The events you have tagged and exported can be analysed to gain an understanding of the characteristics of performance for a team or individual. If we had you tag a series of games/matches in a particular sport, this begins to create a database. In the current day, data is highly valued. Therefore, an important skill for a sport scientist is having the required skillset to manage the data. We could choose to tag a series of AFL matches and then get you to analyse this data, however, given you have learnt the skill of tagging events already, we will give you a season of AFL match data for you to analyse.

Datasets - AFL Match performance indicator dataset

There are a number of ways that sports data can be obtained in the current day. For today’s example, we will focus on obtaining AFL match data. For most of us it is not possible to obtain licenced AFL match data from statistical providers (i.e Champion Data) due to the associated cost. However, on the AFL website we can view some publicly available data. See the screenshot of AFL website statistics that are available. There are a number of different software programs we could use to analyse the dataset, but for the purposes of this practical class, we are going to take you through the fundamentals using MS Excel for this type of analysis.

Tasks

1. By using the PivotTable, report the team that had the highest average team value for

2. Which team had the lowest average value for Marks?

3. Which team had the greatest number of Inside 50’s for Round 17?

4. Which team had the least and most number of clangers in a match and what round did this instances occur?

This HSE311 - Sport Science has been solved by our PhD Experts at My Uni Paper.