Updating the Rugby World Cup 2019 Network Diagrams … and future plans.

Now that all the changes to the squads have been added, admittedly after the tournament has finished, I can update the figures. In general, the changes made very little difference.

The quarterfinals:

rdStjP.png

(Sorry to any New Zealanders for cutting off the N of the country name.)

The only difference adding Rob Herring for Sean Cronin for Ireland makes is that Leinster and Munster now both have the most players left in with 12, followed by Crusaders, Scarlets and Saracens with 11.

All 8 teams are their own communities.

Japan are the team closest to the centre still, but Yamaha Jubilo are now the team closest to the centre.

In the total players used up to the quarterfinals diagram, Canada and France have still added the most (4), then Ireland, Italy, Samoa, Scotland, South Africa and Tonga with 2 then Argentina, Fiji, New Zealand, the United States and Wales who have all added one.

rdSSws.png

All 20 teams remain their own communities.

Jaguares have the most (27) players at the World Cup, followed by Welwitschias (20) and Glasgow warriors and Benetton (16).

Scotland and Pau are the teams closest to the centre.

Semifinals:

The addition of Owen Lane for Josh Navidi changes nothing because it swapped a Cardiff Blues player for a Blues player. Therefore, the teams in the centre haven’t changed from the previous version (England and Harlequins) and the teams with the most players haven’t changed.

rdSReF.png

In the total players used up to the semifinals diagram, Canada and France have still added the most (4), then Ireland, Italy, Samoa, Scotland, South Africa, Tonga and Wales with 2 then Argentina, Fiji, New Zealand and the United States who have all added one.

rdSqjQ.png

All 20 teams remain their own communities.

Jaguares have the most players at the World Cup (27), followed by Welwitschias (20) and Glasgow warriors and Benetton (16).

Scotland and Pau are the teams closest to the centre.

Final:

No changes to the diagram showing just the finalists because I made the Ben Spencer for Willi Heinz change in the original diagram. I made it because it had an effect on how close the teams were to each other as Willi Heinz and one of the South Africans both play for Gloucester while Ben Spencer plays for Saracens.

In the total players used up to the final diagram, Canada and France have still added the most (4), then Ireland, Italy, Samoa, Scotland, South Africa, Tonga and Wales with 2 then Argentina, England, Fiji, New Zealand and the United States who have all added one.

rdSlIJ.png

All 20 teams remain their own communities.

Jaguares have the most players (27), followed by Welwitschias (20) and Glasgow warriors, Benetton and Saracens (16) (due to addition of Ben Spencer of Saracens for Willi Heinz).

Scotland and Pau are the teams closest to the centre.

~~~~

I wanted to see if there was any correlation between final result and players named to the squad. Obviously, teams that went further in the tournament played more games which increases the risk of injuries. Therefore, I divided the numbers of total players (and total players/original players) by the number of games played to try to account for that.

If you look at total players named to squads divided by games played versus the team final positions it looks like this:

rdSObO.png

You can see an obvious pattern. There is a similar pattern if you plot starting number of players named divided by total players named then divided by games played against final positions.

rdSA9N.png

I’m not sure what to do with the information. Dividing by the number of games played has a huge effect and I don’t know if the effect is out-sized. Also, it’s all well and good to be able to see patterns at the end but it would be interesting to see if you could predict final positions from this sort of information at the end of the group stage.

Another interesting question, raised by L, is whether you can predict anything from number of players actually played and which teams maintained the most continuity, in terms of players who were on the pitch with each other. It’s something you could probably work out from easily available data, but it will take time to do so it is being put into the future plans folder.

Watch this space, but don’t hold your breath 😉

Other forthcoming plans for this data include trying to make a video showing the changes throughout the competition – the first few dry runs look very pretty but that might also take some time to perfect, but the results of that should be out sooner than the other analysis.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s