World Cup 2022 is progressing, many good matches and much entertainment. Time then for World Cup 2022 predictions of the round of 16 matches from our DIBP model – here the previous update. In the group stage matches the average of the model probabilities for the actual final results was about 0.52.
Here there are the posterior predictive match probabilities for the held-out matches of the Qatar 2022 round of 16 to be played from December 3rd to December 6th, along with some ppd ‘chessboard plots’ for the exact outcomes in gray-scale color – ‘mlo’ in the table denotes the ‘most likely result’ , whereas darker regions in the plots correspond to more likely results. In the plots below, the first team listed in each sub-title is the ‘favorite’ (x-axis), whereas the second team is the ‘underdog’ (y-axis). The 2-way grid displays the 8 held-out matches in such a way that closer matches appear at the top-left of the grid, whereas more unbalanced matches (‘blowouts’) appear at the bottom-right. The matches are then ordered from top-left to bottom-right in terms of increasing winning probability for the favorite teams. The table reports instead the matches according to a chronological order.
Apparently, Brazil is highly favorite against South Korea, and Argentina seems much ahead against Australia, whereas much balance is predicted for Japan-Croatia, Netherlands-United States and Portugal-Switzerland. Note: take in consideration that these probabilities refer to the regular times, then within the 90 minutes. The model does not capture supplementary times probabilities.
You find the complete results, R code and analysis here. Some preliminary notes and model limitations can be found here.
Next steps: we’ll update the predictions for the quarter of finals. We are still discussing about the possibility to report some overall World Cup winning probabilities, even though I am personally not a huge fan of these ahead-predictions (even coding this scenario is not straightforward…!). However, we know those predictions could be really amusing for fans, so maybe we are going to report them after the round of 16. We also could post some pp checks for the model and more predictive performance measures.
Stay tuned!
Thanks!
Netherlands match was not even close and Argentina almost spilled their chance right near the end of the match. I think both predictions got it wrong.
Alan:
These are probabilistic predictions with many possible outcomes so it does not make sense to say the probabilistic prediction “got it wrong.” Unless the model predicts a bunch of unlikely events or is wrong in some systematic ways.
Perhaps Alan is interpreting the probabilities as: the probability of each outcome given how well each team will play. Actually, it would be kind of cute to take advantage of modern sports statistics and have one model for expected goals (and expected outcome) and an additional model for actual goals (and actual outcome) given expected goals. The latter model would be what Alan apparently thinks this model is.
Next round predictions are out. https://www.leonardoegidi.com/_files/ugd/05c43e_f5e646241cd140f88a22718b14f6479a.pdf
> [favorite] / underdog – fav. / draw / und.
> [Brazil] / Croatia – 0.677 / 0.220 / 0.104
> [Argentina] / Netherlands – 0.414 / 0.264 / 0.322
> [Portugal] / Morocco – 0.536 / 0.191 / 0.273
> [England] / France – 0.361 / 0.292 / 0.348
For reference, the implied probabilities from betting odds are the approximately
> [Brazil] / Croatia – 0.71 / 0.19 / 0.10
> [Argentina] / Netherlands – 0.42 / 0.31 / 0.27
> [Portugal] / Morocco – 0.60 / 0.25 / 0.15
> England / [France] – 0.31 / 0.30 / 0.39
The most likely (betting-odds-implied) final seems to be Brazil – France. They are the favourites – paying 2.75 to 1 and 5 to 1 respectively.
Since there aren’t ties in the knockout rounds, is the model calling draws based on “tied at full time” or “tied at extra time (goes to penalties)”?
Would you expect teams to be 50/50 when it goes to penalties, or is there a way of making an informative prior based on, for example, quality of offense as a proxy for shooting skill?
The model is designed for the regular times only,
I guess there is space for informative priors for tthe penalties, as you mention.
I see the new predictions are available.
For the curious, the favorites for the semifinals are Argentina (58% vs 16%) and France (50% vs 20%).
For the final Argentina 45%, France 35%, Croatia 12%, Morocco 8%.