With the European Championship around the corner, there are millions of fans around the world trying to predict the results and the score in a game and the number of goals scored is a central part in these predictions.
A study previously carried out by The Stats Zone showed that despite the fact that it is unique for each respective league, all top European club competitions in football have goal scoring seasonality. Regardless of the difference in seasonality there are a few key characteristics that correspond for all leagues. Two of these characteristics are pitch conditions and the fitness of players. However, the single most important factor for the number of goals scored in a match proved to be the importance of the game.
To put this more simply, it’s essentially whether the teams have anything to play for. The study proved that the number of goals scored in a match increased towards the final stages of the season when titles had already been won, European places settled and relegation decided, making games less important. Due to the cup nature of the European Championship we can certainly say that the average game means more compared to a regular league game. The Stats Zone has decided to take a closer look at goal scoring patterns, taking into consideration the importance of games in the European Championships, to see if shorter competitions also contain trends of seasonality.
There were a few restrictions in this analysis. Firstly we wanted to look at the goal scoring level for each respective round in the tournament (group, quarter, semi & final); to provide a clear data set we have used the last five tournaments in the European Championship (1996 to 2012) as they have the same format. The format for the European Championship in 2016 has increased to 24 teams; whilst we acknowledge this, we believe that the change isn’t significant enough to influence the analysis. Secondly, we have decided to only look at the score at full time and goals scored on penalties or in extra time have therefore been omitted from our data.
To enable a comparison we have decided to look at the average number of goals scored in each round (for all championships combined) and visualize this through a bar graph. The lines in the second graph illustrate the average percentage of goals scored for each year in a given period and if this is above or below the total tournament average (normalised to 0). The bar graph within illustrates the total average and whether the number of goals scored in a certain period is higher (green) or lower (red) than this average. For example, if the average number of goals scored in round two in 1996 was 2.5 and the joint average for all tournaments was 2.5, the line graph would sit at 0%.
The table above illustrates the average goal scoring distribution for the European Championships. We can see that round three has the highest average at 2.8 goals scored per match. The most likely reason for this is that teams at this stage are either knocked out or have qualified for the next round and so have nothing to play for.
This makes matches less important and with support from our previous goal scoring study, we know this correlates to an increased average number of goals scored. Moving on to the final stages there is a significant decrease in the average number of goals scored compared to the group stage, with the quarter finals sitting at 2.1 and the semi finals at 1.8. This makes sense since at this stage of the competition teams are more evenly matched leading to a lower average number of goals scored. Interestingly, the final has an average of 2.0 goals scored which is 0.2 goals higher than the semi finals. The next graph investigates this in order to find an explanation.
From the graph above, it’s clear that round two and three see 26.1% and 85.6% more goals scored than the average. They are also the only two games that have a higher than average number of goals scored. As previously mentioned, on average, the final sees more goals scored than the semi-finals. A closer look into the line graph shows that 2012 was the only year where the score was significantly above the average. Spain won this Championship after a convincing 4-0 victory against Italy in the final which would have pushed the total average significantly higher. We can deduce that this is the reason why the average goals scored in the final is sitting above the semi final value.
Looking closer at the individual line graphs there is not obvious pattern; regardless of the period we are looking at, there are some years that sit above or below the average. However, with the support from the first bar graph it possibly makes more sense to look at the total average. Here we can see that the number of goals increase until the third round, after which they start decreasing as the tournament approaches its final stage.
To conclude, we have looked at the seasonality and goal scoring distribution for the last five European Championships. This analysis shows that the importance of the game has an influence on the number of goals scored, with high profile games seeing fewer goals per match. On the flip side, when we look at the third round -which on average is the least important game, with teams often already knocked out or qualified- we see a higher number of goals scored. With this in mind, don’t be surprised if your team is not scoring as many goals as you might expect in the early stages. To finish this off, I think it would be very interesting to look at a similar analysis for the World Cup. This would open up an opportunity to confirm or reject if the seasonality tend to be similar between different football tournaments.