Well, there you have it. Simple, isn’t it?
That, my good friend, is the result of 5,000 simulations of the Rugby World Cup using our machine learning Rugby Algorithm. We use various factors to determine team strength, team form, head to head records and a few other bits and bobs and cram them all together into a veritable soup of rugby statistics. The Algorithm has a look at all this, decides what is important in determining the outcome of a rugby match based on previous matches, and the builds a prediction for future matches and the tournament as a whole. Lovely job.
[For individual predictions for the group stages see here.]
Mainly to annoy my wife. She hates rugby.
Perhaps unsurprisingly, New Zealand come out on top (despite their recent woes). Their historical strength is too great, according to the Algorithm, to be ignored on the basis of a few recent matches. However, it may interest you to note that their chances have dropped sharply on the back of their Rugby Championship performances. We calculate the probability of them winning the title has reduced by 16% over the course of the Championship, from 71% then to 55% today. This will continue to be updated until the tournament begins, and throughout the World Cup – so stay tuned to see further updates.
Lots! Here’s a chart that shows the chances the Algorithm has calculated for each team getting to each stage.
One of the things that can be read from the above is that we have 6 very certain quarter finalists – in the shape of New Zealand, England, South Africa, Australia, Ireland and Wales – and 4 battling it out for the final 2 spots.
Scotland, France, Argentina and Japan will have to fight it out to make the knockout stages. Seem reasonable?
I’ll leave you to tease out any other nuances that you may see in the above then. Please let me know if you spot anything interesting and feel free to ask any questions in the comments below, or email me about anything else. I’m always happy to try and explain things that may not immediately make sense. A few things worth noting are:
Wales are currently second in the World Rankings, but very briefly were ‘unofficially’ top last weekend at one point between matches. However, I feel World Rugby’s rankings system can sometimes overestimate teams given certain schedules and home and away status. Even though Wales have a good chance of going top ‘officially’ this weekend (beat England and they will be ranked first), using Rugby4Cast’s own ranking system we actually calculate their position much lower.
Yes, well. That’s what it says. For more information on the details, how they are calculated and differences between the ranking systems, see here. In short, the Elo system reflects team strength a little better by incorporating match expectation into the calculation, rather than just the final match outcome. In fairness, it’s close in the middle. There really isn’t much between 2nd and 6th and we could see a lot of movement over the next few weekends. Come back to check on how the rest of the warm up matches have affected team ranking.
That’s it for starters. I think the above should have sufficiently annoyed at least 50% of the readership. We’ve got a whole host of other things that we’ll be putting out in the run up over the next few weeks, but let us know if there is anything you’d particularly like to see and we can try and oblige.
You are most welcome. Stay safe out there.