Update: Are goal-kicks and long balls useless exercises?

Were goal kicks useless in the 2018 FIFA world cup as well?

This is a follow-up piece on the goal-kick post that I wrote a couple of months ago. I must admit though that this serves more as a filler than as a genuine attempt at validating the findings reported in the previous post. Blame it on writer’s block.

Data on the FIFA 2018 world cup games are once again from the publicly accessible whoscored.com website (https://www.whoscored.com). I repeated the same analyses that I had done previously. In short, I asked the following questions:

1. Are long balls common?

2. Do they adversely affect possession?

3. Does possession impact goal-scoring ability?

4. Do long balls adversely impact goal-scoring ability?

As in the 2014 world cup, long balls are common, with one long ball played for every 6 short passes. Long ball accuracy was poorer in 2018 (under 50%) compared to 58% in 2014. This was probably because there were more teams this year who were not good long ball players. Fifteen of the 32 teams this year had a long ball accuracy of less than 50% compared to only 4 of 32 in 2014. This was also reflected in a wider range of possession percent than the last time around. The ratio of long balls to short passes had a greater impact on possession in the 2018 games (fig).

However, possession explained only 11% of the variation in open-play goals scored this time on, with one-fifth of the goals scored by teams who controlled the ball less than 45% of the time (twice as likely as in 2014). However this finding is confounded by the fact that 27% fewer goals were scored in 2018 compared to 2014, and the majority of them were scored from set pieces. Open play goals accounted for only about 45% of the goals scored in 2018 compared to nearly 65% in 2014. The reasons for this change are fodder for more detailed analyses. Perhaps another slow Sunday with writer’s block! But my guess is that the unexpectedly poor performance of some countries who play high quality possession football skewed these results. For example, Germany (possession 65%) scored just one goal before exiting the tournament. Spain (possession 69%) didn’t do much better either (4 goals). Whereas, the Russians, who did more running than passing, (possession 39.4%) scored 5 goals. Be that as it may, long balls most certainly did not help teams score goals in 2018. The ratio of long balls to short passes adversely impacted goal-scoring in open play. The caveat though is that in 2018, there were several weak teams which relied heavily on long passes (and perhaps also desperate clearances). Iran (long ball to short pass ratio over 0.4), which did not score a single goal, is a good example.

The 2018 world cup finals also confirm the general sense that long balls (including goal-kicks) can adversely affect possession and goal-scoring ability. However, when teams are evenly matched, long balls are often used to open up the opponents defenses to create goal-scoring opportunities. But this strategy is likely to work only when long pass accuracy is high, and there is a more than even chance of winning aerial duels.