Predicting Euro 2024 (matchday 2)
In the last weeks I have built 2 different models to predict the Group Stage results of Euro 2024. In this article I explained how to predict the results for the matches played in matchday 1. While in this article I explained how to predict the total goals scored in the Group Stage matches.
Let’s look at the actual result to see how the model behaved in the first matchday of the competition and let’s see the model predictions for matchday two.
Matchday 1 results predictions
See below the table with the predicted and actual results of all matches (except Germany).
| Match | Prediction | Result | Win (Odds) |
|---|---|---|---|
| Hungary-Switzerland | 2 | 2 | +1.16 (2.16) |
| Spain-Croatia | 1 | 1 | +0.94 (1.94) |
| Italy-Albania | 1 | 1 | +0.36 (1.36) |
| Slovenia-Denmark | 2 | X | -1.00 (3.63) |
| Serbia-England | 1 | 2 | -1.00 (1.50) |
| Poland-Netherlands | 1 | 2 | -1.00 (1.51) |
| Austria-France | 1 | 2 | -1.00 (1.57) |
| Romania-Ukraine | 2 | 1 | -1.00 (4.77) |
| Belgium-Slovakia | X | 2 | -1.00 (6.86) |
| Türki̇ye-Georgia | 2 | 1 | -1.00 (1.62) |
| Portugal-Czechia | 1 | 1 | +0.50 (1.50) |
| Final | 4/11 = 36% | -4.04 |
The model had a good start predicting the first 3 matches correctly, but failed to predict England, Netherlands and France win. It predicted also a draw between Belgium and Slovakia, which was pretty close, but failed to predict the majority of the results.
The total correct predictions are 4 out of 11, which is 36% accuracy. Simulating a betting strategy of 1 unit per match, this would have lost 4.04 units.
Matchday 1 goals predictions
See below the table with the predicted and actual Over 2.5 goals of all matches (except Germany).
| Match | Prediction (Over 2.5) | Goals (Exp goals) | Win (Odds) |
|---|---|---|---|
| Hungary-Switzerland | Over | 4 (3.58) | +1.49 (2.49) |
| Spain-Croatia | Over | 3 (4.39) | +1.13 (2.13) |
| Italy-Albania | Under | 3 (2.07) | -1.00 (2.02) |
| Slovenia-Denmark | Over | 2 (2.76) | -1.00 (1.70) |
| Serbia-England | Over | 1 (0.74) | -1.00 (2.06) |
| Poland-Netherlands | Over | 3 (2.78) | +0.99 (1.99) |
| Austria-France | Over | 1 (2.85) | -1.00 (2.07) |
| Romania-Ukraine | Under | 3 (1.95) | -1.00 (2.06) |
| Belgium-Slovakia | Over | 1 (2.27) | -1.00 (2.13) |
| Türki̇ye-Georgia | Over | 4 (4.29) | +0.96 (1.96) |
| Portugal-Czechia | Over | 3 (2.28) | +0.78 (1.78) |
| Final | 5/11 = 45% | -0.65 |
Here, the model predicts only 5 out of 11 results correctly. However, notice how the model predictions are extremely aligned with the expected goals in 9 out of 11 matches, except for Belgium (two disallowed goals) and England matches. This suggests that the model has been affected by a quite high amount variance in this matchday. For example, Romania, Italy and Albania scored more goals than they deserved, while France and Netherlands scored less, according to their Expected Goals.
The total correct predictions are 5 out of 11, a 45% accuracy. Simulating a betting strategy of 1 unit per match, this would have lost 0.65 units.
Matchday 2 results and goals predictions
Using the same approach as for the first matchday, these are the predictions for results and goals for matchday two. These predictions are based on the performance during the Qualifiers and the first match day. In the case of Germany, only on matchday one performance.
| Date | Match | Prediction (Odds) | Over/Under 2.5 (Odds) |
|---|---|---|---|
| 19-06-24 | Croatia-Albania | X (4.46) | Under (1.9) |
| 19-06-24 | Germany-Hungary | 1 (1.27) | Over (1.51) |
| 19-06-24 | Scotland-Switzerland | 2 (1.89) | Over (2.25) |
| 20-06-24 | Slovenia-Serbia | 1 (4.92) | Over (2.06) |
| 20-06-24 | Denmark-England | 1 (5.61) | Over (2.20) |
| 20-06-24 | Spain-Italy | 1 (2.14) | Under (1.70) |
| 21-06-24 | Slovakia-Ukraine | 2 (2.10) | Under (1.73) |
| 21-06-24 | Poland-Austria | 2 (1.98) | Over (1.96) |
| 21-06-24 | Netherlands-France | 2 (2.31) | Over (2.00) |
| 22-06-24 | Georgia-Czechia | X (3.85) | Over (1.94) |
| 22-06-24 | Türki̇ye-Portugal | 2 (1.58) | Over (1.72) |
| 22-06-24 | Belgium-Romania | X (4.49) | Over (1.77) |
As the tournament goes on, and we incorporate the data of the matches the model should give more accurate predictions. So I expect this matchday to give better results than the first one.
If you are interested in learning more about how to build a betting model for Euro 2024 and building a model to predict football results, I have written a few books where I go into the details of how to get the data, visualize and train a model, complete with code examples.
Your Ad Here
