The Pride’s Offense is Alive but Recently Has Not Thrived

The Pride keep winning, and former New York Jets coach Herm Edwards said it best: “you play to win the game. Hello!” Whether the result looks like free-flowing and beautiful soccer or brute force, rough-and-tumble soccer that looks more like (American) football, in the end all that matters is the scoreboard, and the Pride are still undefeated and in the midst of their second six-game winning streak of the season.

That said, since the Pride came back to NWSL play after the Olympic break, they have not quite looked as dominant offensively as they did pre-Olympics. The defense remains rock solid, with zero goals given up in all three games since the return (note: zero goals in the last three games and only one goal given up in the last seven NWSL games, wow!), but the offense looks slightly off, or as off as a team can be that scores in and wins three straight games. Are these champagne problems, or is there something bubbling up that will need ironing out as the calendar turns to autumn and playoff season nears? I do not want to be the male version of Debbie Downer (Andrew Alarmist? Hopefully there will be no need to workshop this any further.), but I think it is worth exploring.

Let’s take a look at some numbers for the Pride that illustrate how the offensive performance during the most recent three games looks as compared to prior games. In the chart below, taken from data provided on fbref.com and excluding opposition own goals, are the totals for the first 16 games of the season in the first row, the last three of those first 16 games in the second row, and then the most recent three games in the third row.

With the exception of the percentage of shots the Pride put on target, games 17-19 were worse across the board in every other category when compared to the three games immediately prior to the Olympic break and to the first 16 games overall. Worse is a relative term, because if you compare the “worse” performances in those most recent three games to the rest of the NWSL throughout the full season, the Pride’s performance compares favorably in many categories, even though when compared to their own earlier season performances it seems like a downturn.

Compared to every other NWSL team’s full season metrics, the Pride’s last three games are demonstrably better (blue highlighted cells are where the Pride’s output is better) than most other teams in goals, shots on target, and expected goals. The Pride are worse in the conversion stats, goals as a percentage of shots and as a percentage of shots on target, and the number of shots they are expected to take per goal (based on expected goals scored). This is the part in every article where I add my view on expected goals; I think this data is useful in terms of looking at whether the Pride are taking shots from dangerous locations, but I do not actually put a lot of faith into whether a goal really should or should not be “expected” from that location.

If you do the PhD-level mathematics of counting the blue cells as a percentage of the total cells, the Pride are better in 47.4% of the cells above, which ranks them seventh among all teams. This is, of course, not an even comparison, as it is three games for the Pride as compared to every other team’s full 19 games, but the point is that even during a recent “down” period for the Pride’s offense the team is still in the top half of the 14 team NWSL and is performing better than all but three teams in the offensive category that matters most — goals per game. And the Pride are performing better than every team in the league in another somewhat significant measure, the league standings. Take that, Andrew Alarmist.

Alas, we cannot abrogate (SAT word alert) Andrew Alarmist just yet. We still need to take a look at Barbra Banda and what has been going on with her recently, as she has zero goals or assists in each of the last three games. Back in July I wrote an article about the historic pace Banda was on, and it is important to preface a look at the recent “poor performance” by Banda by remembering that with 17 goal contributions (12 goals, 5 assists) she is already tied for the 11th best single season in NWSL history, and with her next goal or assist she will move into a tie for 10th. If we look at the same data for Banda as we did for the full team earlier, we can see that her performance mirrors the Pride’s, which should not be surprising, considering of how much of a focal point she is for their offense.

Clearly the zeroes in the goals categories stand out as compared to the other numbers, but the big takeaway for me was really was around the quality of her shots. This is where I think expected goals really shines, because like a standardized test, it is the same for everyone, and so you can see that the locations of Banda’s shots pre-Olympics (918 minutes over 12 games) were much better than they have been recently (257 minutes over 3 games).

Banda is averaging nearly the same number of shots per 90 minutes as she was pre-Olympics (5.3 shots per 90 mins post-Olympics vs. 5.8 shots per 90 mins pre-Olympics), so the fact that her expected goals per game in the post-Olympics games is half as high indicates that the shots she has taken recently have been from more acute angles and/or farther away from the goal. Correspondingly, her shots on target percentage have dropped, which should not be a surprise based on the expected goals evaluation that the shots she has been taking are from more difficult locations.

On the flipside of the lack of goal contributions by Banda though is, once again, that the team has won all three of its recent games, even with her not scoring. Marta has now scored six goals on the season and Adriana and Summer Yates both have five, making the Pride the only team in the NWSL with four players with at least five goals scored this season. It is always better to have multiple threats, and the Pride are now rolling out four players that force the opposition to game plan against.

To go back to my question earlier, I think that it is indeed a champagne problem for the Pride that since the Olympic break they are winning games by close scores as opposed to walloping their opponents. Bill James, one of the originators of analytics in baseball, felt that run differential was more important than a team’s actual record in terms of determining who the best team in baseball really was, and if we apply this same concept to soccer, we would look at goal differential as the corresponding measure. The Pride are +25 thus far this season, 39% better than the team ranked next behind them and on track for one of the best seasons in league history, if not the best.

To give myself a little bit of credit, I think it was fair to look at the recent games with a discerning eye, but if Julie Doyle was not offside by inches against Gotham, that game is 3-0 instead of 2-0, and Banda also had multiple shots that were cleared off the line against the Red Stars, so that game could have been a multiple-goal victory as well.

To go back to the baseball well one more time, sometimes the Moneyball idea that you can evaluate sports by only looking at the numbers can be deceiving. You have to watch the games too. The Pride have clearly been the better team during all three of their recent games and Banda has been a force to be reckoned with as well. They just have been inches away from showing it both on the field and in the box scores.

I expect that a positive regression to the mean is coming for the Pride’s offense, hopefully as soon as Friday night in the Pride’s rematch with the Current!