#### The mathematical review of the results of the 27th matchday

At this point, a comparison of the predicted goal expectations with the reality, as well as the comparison of the expected determination with the arrived.

Goal expectation

First of all this remark: last week an error crept into the expectation. All expectations probably corresponded to those of the previous week. It was a copying error, as you might imagine. This error is corrected here, the expectations given are those calculated by the computer (even if now no longer directly comprehensible to the reader).

All comments on the over/under games from last week in the preview therefore have no meaning.

Further, the away teams exceeded expectations and the home teams underperformed, so there was some alignment for the season statistics, where the home teams had so far clearly scored too much and conceded too little.

Freiburg was an under. The computer had expected few goals, two were scored. In the case of Bayern against Hannover, although it looked like an over, the betting market certainly had the same view, so that here it was more likely to assume an “under” tip. This would have been true to a certain extent, as the line was certainly three or more goals, so one would not have lost with an under bet (unless one had played the traditional under, over/under 2.5 goals).

Many goals also in Mainz, which the computer was also right about. Stuttgart, too, would probably have been an “under” bet – and it would have come true. On the other hand, the many expected goals in Schalke did not materialise, so that one would certainly have lost this over.

The expected goal difference was undercut, which speaks for very good predictions overall.

Here, always as a reminder, are the values from the previous week. The main point is that the expected determination is calculated here once again in a comprehensible way.

The number of surprises on this matchday is almost exactly the expected one, as the expected determination was almost exactly hit. It is probably hard to get any closer than that for a matchday.

The surprises are Hoffenheim’s win at Gladbach (with only 15.56%), Berlin’s win at Mainz (with 19.13%) and the draw between Werder and Augsburg (with 22.02%). Otherwise, the favourites prevailed, but overall confirmed the numbers (the right amount of surprises, with these selected three games).