Just now I had dinner with some relatives, and during the meal the talk turned to the World Cup — England 🏴 vs France 🇫🇷

The elders went on and on for ages, talking at length about n-things, analyzing all kinds of pros and cons bla bla — and then they said that the odds are the bookmaker’s “harvesting” tool; whatever odds you see are set by some boss, who heard it from xxx.... And then I chimed in: actually, this can be calculated.

Then came the classic “you know what” part — like “simple math can’t do it,” and stuff like that.

Well, today I’m “breaking new ground” and telling you that this probability can indeed be worked out if you put some thought into it.

Disclaimer: pure algorithmic discussion; no responsibility for the outcome.

1. Calculate the World Cup teams’ average goals scored/conceded value (you wouldn’t directly ask an AI)
102 matches, 90-minute total goals 290, average goals per match 2.843, standard deviation 1.716. Then remove extreme values (Germany 7-1 against Curaçao that match). After removing:
Average per match 2.792 goals/match → single-team baseline μ = 1.396 goals/team/match

2. Calculate the data for the two teams in this World Cup (just 90-minute match records)
France: 7 matches, 16 scored, 4 conceded; average scored 2.286, conceded 0.571
England: 7 matches, 13 scored, 8 conceded; average scored 1.857, conceded 1.143

3. Compute the teams’ strength
Attack/defense intensity coefficient = team mean ÷ tournament baseline; 1.0 means average level:
France attack = 2.286 ÷ 1.396 = 1.637 (64% more goals than average team)
France defense = 0.571 ÷ 1.396 = 0.409 (only concedes 41% of the average team’s goals)
England attack = 1.857 ÷ 1.396 = 1.330
England defense = 1.143 ÷ 1.396 = 0.819

4. Expected goals λ
Our team expected goals = our attack strength × opponent defense strength × μ:
λ(France)= 1.637 × 0.819 × 1.396 = 1.871
λ(England)= 1.330 × 0.409 × 1.396 = 0.760

5. Poisson distribution (let the AI run it; get the marginal distribution of each team’s goals)
The distribution looks roughly like this:
P(France goals - England goals)
P(1-0) = 28.8% × 46.8% = 13.5%
P(2-0) = 26.9% × 46.8% = 12.6%
P(1-1) = 28.8% × 35.6% = 10.2%
P(2-1) = 26.9% × 35.6% = 9.6%
P(0-0) = 15.4% × 46.8% = 7.2%

6. Matrix aggregation
Sum the whole matrix by regions:
France win = 63.7% → odds 1 ÷ 0.637 = 1.57
Draw = 21.7% → 4.61
England win = 14.6% → 6.87
Total goals ≤ 2 = 51.1%, both teams score = 45.0%

The above is the probability analysis for a 90-minute match.
One more reminder: this is purely academic discussion; it doesn’t take responsibility for the match result.

What I want to tell everyone is that this is also math—an algorithm; there isn’t all that much “manipulation” by retail traders and bookmakers (though, of course, there is).