BookmarkSubscribeRSS Feed

Using SAS Viya to Select the British & Irish Lions Rugby Team: Part 4

Started ‎07-22-2021 by
Modified ‎07-22-2021 by
Views 3,914

The Simulation Dashboard

Since there is a strong element of subjectivity in the model process it makes sense to move this out of a code oriented environment and into the hands of the Subject Matter Experts directly. To give an example of how this might work we use an interactive dashboard with SAS Visual Analytics which calls the Job Execution Service behind the scenes.

 

The Job Execution Service allows you to run SAS code as an HTTP endpoint with a back-end script for logic, and a front-end HTML form. We then simply embed this as a web component within the Visual Analytics dashboard. In Figure 27 you can see that the HTML form renders as a native front-end. Because there are so many parameters to fiddle with it made sense to do this as a CSV file upload rather than changing the parameters via a web interface. Analysts can simply toy with the variables, upload a file to the interface and then re-run the simulation. The file is only read on input as a transient table so there is no storage headache either.

HarrySnart_0-1626945668894.png

Figure 27 - Job Execution Service Front End in Visual Analytics

 

The simulation output shows in a new browser tab so you can validate that the model has ran. A nice feature as well is the ability to download a PowerPoint report from the Job Execution Service output. This PowerPoint file is produced using ODS while the simulation script runs in the back-end

HarrySnart_1-1626945668904.png

Figure 28 - Job Execution Service Output with Download Link

 

HarrySnart_2-1626945668982.png

Figure 29 - PowerPoint Report Download Output

 

We are able to then interactively compare the results of multiple model simulations in an interactive Visual Analytics tab. Each time the model is ran it gets added to a group table which labels the model by the file weights. In Figure 30 you can see I have multiple model weights depending on the type of team I am looking to select. Then in Figure 31 you can see the interactive report output where we visualize our selection. The starting line-up are visualized using a raster coordinate system to mimic the field positions.

HarrySnart_3-1626945668985.png

Figure 30 - Model Weight Files

 

HarrySnart_4-1626945669086.png

Figure 31 - Interactive Dashboard for Model Simulations

 

How Well Did The Model Perform?

I ran five simulations based on differing preference weights: Attacking, Defensive, Experienced, Neutral and Youthful.

 

Looking at the player selection counts, several players were selected in many of the scenarios, as shown in Figure 32.

HarrySnart_5-1626945669093.png

Figure 32 - Player Selections across models

 

So, how does the model perform against the actual team selection?

 

As I write this blog the starting line-up for the first test match has been revealed.

 

Drum roll, please…

 

The actual team selected for the first test match on Saturday July 24th is:

 

1

Wyn Jones (selected 1/5 times in the SAS models)

2

Luke Cowan-Dickie (selected 4/5 times in the SAS models)

3

Tadhg Furlong (selected 5/5 times in the SAS models)

4

Maro Itoje (selected 5/5 times in the SAS models)

5

Alun Wyn Jones (selected 1/5 times in the SAS models)

6

Courtney Lawes (selected 5/5 times in the SAS models)

7

Tom Curry (selected 2/5 times in the SAS models)

8

Jack Conan (selected 1/5 times in the SAS models)

9

Ali Price (selected 5/5 times in the SAS models)

10

Dan Biggar (selected 4/5 times in the SAS models)

11

Duhan Van Der Merwe (selected 2/5 times in the SAS models)

12

Robbie Henshaw (selected 5/5 times in the SAS models)

13

Elliot Daly (selected 5/5 times in the SAS models)

14

Anthony Watson (selected 2/5 times in the SAS models)

15

Stuart Hogg (selected 4/5 times in the SAS models)

16

Ken Owens (selected 1/5 times in the SAS models)

17

Rory Sutherland (selected 5/5 times in the SAS models)

18

Kyle Sinckler (injury cover, not in SAS dataset)

19

Tadhg Beirne (selected 4/5 times in the SAS models)

20

Hamish Watson (selected 5/5 times in the SAS models)

21

Conor Murray (selected 2/5 times in the SAS models)

22

Owen Farrell (selected 4/5 times in the SAS models)

23

Liam Williams (selected 4/5 times in the SAS models)

 

We can see that, with the exception of Kyle Sinckler who was not in the dataset, the models frequently select the players who actually made the side.

 

In fact, if we look at the histogram of player selections in Figure 33, you can see that every player in the real selected side was selected at least once by a SAS simulation, with more than 30% of players being selected in every single model. The aggregated model inclusion is that, on average, the real team are selected in 70% of SAS models. Given that there was far less data available than sports teams would actually have access to the models perform surprisingly well.

HarrySnart_6-1626945669095.png

Figure 33 - Overall Model Performance

 

Looking at the real-life selected team it is clear that Warren Gatland is opting for more of an experienced side for the first test by starting players like Anthony Watson, Elliot Daly, Dan Biggar and Courtney Lawes. In fact, the key bolters have mostly failed to make the team altogether. Perhaps they will be given an opportunity in the following two test matches.

 

Another interesting observation is that Ali Price has been selected as the starting scrumhalf. He wins his first ever cap for the Lions, and gets the starting spot ahead of three time Lions tourist Conor Murray (who briefly held the captaincy for the Lions whilst Alun Wyn Jones was injured). It is, therefore, very interesting to see that Ali Price was named in 100% of the SAS models despite being a relative underdog – clearly his stats speak for him!

 

Summary

Overall, the SAS models provided a reasonable simulation of the teams. Given the relative lack of data I was not expecting it to correctly pick the full line-up, but the models actually do a very good job – especially when considered in aggregate.  This re-iterates the point I made in the introduction: a model is best served as a guide, with which you can then make an informed decision.

 

Reading the press release from Warren Gatland this is also how he picks his team – he does not select the full side himself, he asks the rest of the coaching team to come up with their own 23 and then they compare and debate the results.

 

The full press release on how they made their selections is here: https://www.lionsrugby.com/2021/07/21/lions-selection-for-first-test-hardest-ever-for-gatland/  

 

I hope you have enjoyed this blog series. To see more of what SAS can do for Sports Analytics, see our industry page here: https://www.sas.com/en_gb/industry/sports.html

Version history
Last update:
‎07-22-2021 05:31 AM
Updated by:
Contributors

sas-innovate-2024.png

Available on demand!

Missed SAS Innovate Las Vegas? Watch all the action for free! View the keynotes, general sessions and 22 breakouts on demand.

 

Register now!

Free course: Data Literacy Essentials

Data Literacy is for all, even absolute beginners. Jump on board with this free e-learning  and boost your career prospects.

Get Started

Article Tags