Basic analysis of football stats (Part 2)

In the previous post, Corner-stats.com team reviewed a basic analysis of football statistics and made a prediction for corners in the match Inter Milan - Venezia. Now we will continue our reflections and reveal, in our opinion, an important logical error in this approach to the analysis of statistics. For a better understanding, we recommend that you first read the first part of our study.

So, based on the average total corners of Inter and Venezia, as well as the average totals of their opponents, we have determined that Inter should take 6.41 corners in the upcoming match, and Venezia 3.41. Accordingly, if Inter would play, for example, with Atalanta, we would get the expected corners for Inter 5.07, and for Atalanta 4.81. It seems that everything is logical: for Inter against weak Venezia, we got a difference of 3 corners, and against strong Atalanta, the difference on corners is minimal.

But let's think - if the average individual total corners for Inter is 6.19, then it is logical that on average against weak teams its total should be higher (say, somewhere around 7.5), against middle teams it will be around 6.19, and against strong it should be lower (somewhere around 4.9). Similarly, if Venezia on average allows 6.62 corners, against strong teams they should allow about 8.0 corners on average, against middle teams about 6.62 and against weak teams about 5.2 corners.

Based on the above, as well as the fact that Inter is a strong team (2nd place on corners), and Venezia is a weak team (last place on corners), then Inter in the upcoming match, of course, should take significantly more corners than its average individual total. If we take the approximate figures from the previous paragraph, then the individual total corners of Inter should definitely not be less than 7.5 (rather (7.5+8.0)/2=7.75 ), and this is significantly more than 6.41, as we did in the calculations in the first part of the study.

We make a logical conclusion that the basic analysis, how many corners a team takes on average, and how many corners an opponent allows on average, does not give us the correct results, but almost always underestimates the difference between the teams. And the greater the gap in strength between the teams, the greater will be the underestimation of the difference. We also draw your attention to the fact that the conclusion obtained is true not only for corners, but also for all other statistical indicators (goals, shots, throw-ins, etc.).

We hope that you have benefited from our research and that it will help you in your strategies. If you do not agree with us or have other comments on this matter, you can post them in our Telegram chat.

P.S. In the current article, we used approximate calculations (by eye), but how this can be calculated more accurately using formulas, the Corner-stats.com team will tell in our next post.