The Pressure of Passing in the NWSL


On average, a team has 350 pass attempts, 133 defensive actions, 40 set pieces [free kicks, corner kicks, and throw-ins], 10 shots, and 1 goal per game. Yet, the only thing remembered and rewarded are the goal scorers. I cannot fault anyone for this. There is nothing like watching a rocket into the upper 90. I mean, I can watch Christen Press goal replays for days. However, goals alone cannot explain how a game unfolded. Everything else I mentioned can explain the state of a game or a team or a season but it requires a lot more memory and the ability to compare objectively. Thus, for the next few months, I will deconstruct the 2016 NWSL and USWNT season using WoSo Stats. This will be a series split between Backline Soccer and Positives and Negatives. Please note, the data in this post represents 78 games from the 2016 NWSL season (only Portland has a complete set of data). For the visuals that illustrate player position, the data points are the average of the players in said position for their team. In addition, only players who have at least 180 minutes logged are included. For now, we start with basics!

Figure 1. On average, a team has 350 pass attempts, 133 defensive actions, 40 set pieces [free kicks, corner kicks, and throw-ins], 10 shots, and 1 goal per game. Here, you can see how these stats break down for each team.

As mentioned before, on average, there are 350 pass attempts for a team per game. Even from the simple bar graph (Figure 1), there is a clear indication that Seattle has the most attempts per game. However, we can get a better indication of the quality of these passes and who makes them if we compare pass attempts per 90 minutes vs. pass completion percentage. There is a linear trend, which is especially notable if we ignore the positions and just compare by teams (not shown here), with the only outlier being Seattle. This indicates that as pass attempts increase, the percentage of these passes completed also increases and this is very favorable.

Figure 2. As the number of pass attempts increases, the completion rate increases. The bold line is the overall trend and the curved lines indicates the 95% confidence interval (there is a 95% chances a data point lies between these lines).

A bigger part of this visual is deconstructing pasted on player position – forward, midfielder, defender. There is a clear indication that midfielders (green) and defenders (blue) are more involved in the game than the forwards (orange). This makes sense because a forward needs to finish the sequence with the ball into the back of the net. However, we can also examine each team individually. For instance, the Spirit’s defenders and midfielders have an almost equivalent number of pass attempts per 90 and with one of the highest completion rates, while their forwards make fewer passes with a lower completion rate (Figure 3A). Other teams that follow this same pattern include Seattle, Kansas City, and Orlando.

Figure 3. Click the image for full size. A) Washington Spirit has a similar pass attempts and completion rate for their defenders and midfielders in comparison to their forwards. B) Portland Thorns has a similar completion rate for their defenders and midfielders but midfielders have more pass attempts. C) Chicago Red Stars have a similar completion rate among all positions with different pass attempts.

In comparison, there is a completely different pattern if you examine a team like Portland Thorns, whose midfielders make more passes than the defenders and the same completion rate (Figure 3B). Interestingly, Boston follows a similar pattern and – to a lesser extent – the Western New York Flash (now the North Carolina Courage). In fact, the Flash make the least number of open play passes, passes that do not include set pieces, and we will investigate this further on a later date. Houston and Sky Blue also illustrate higher passing in the midfield, but these teams follow the actual trend. Chicago is the only team that maintains the same pass completion percentage for all of their positions with the number of pass attempts relatively similar (Figure 3C).

Figure 4. Typically, forwards and midfielders are under higher pressure than defenders (as expected). Therefore, forwards are less likely to complete their pass in comparison to the other two position. Midfielders have the largest discrepancy between teams. Note the scale difference on each axis.

A good portion of passes are taken under pressure. On average, 68% of passes from forwards, 61% of passes from midfielders, and 50% of passes from defenders are taken under pressure, with a completion rate of 80%, 84%, and 83% respectively. While the forwards and defenders form a cluster, the midfielders are a bit more spread in their range (Figure 4). This was also the case in an interactive on defense I built last year. On the spectrum, Washington Spirit midfielders completed more passes under pressure in comparison to Western New York Flash midfielders. At the same time, Orlando Pride midfielders faced less passes under pressure compared to Seattle and Kansas City. In fact, Orlando defenders had the lowest percentage of passes under pressure compared to anyone, though as I previously tweeted, Orlando also had the least defensive action per game as well. Clearly, this layer of analysis is not enough to understand the consequences of less pressure. If I were to venture a guess, many of these averages and any future data is a consequence of the current game structure, in which certain teams face a ‘rival’ four times rather than twice. In Orlando’s case, they faced Houston four times. For now, I will save the analysis of passes under pressure and defensive action for next time.

Part II will appear on Positives and Negatives in 1-2 weeks. Make sure to follow now so you do not miss a post!

Make sure to check out WoSo Stats on Twitter, how you can help, the stats database, definitions of these stats, and some visualizations courtesy of WoSo Stats. Special thanks to the fans who dedicate their time logging matches. The visualization presented here are my own and will made public once all matches are logged. I often tweet out random visuals while exploring so make sure to follow my Twitter!

Follow us on Twitter