An Early Look at the 2019 USWNT World Cup Roster

By Brendan Kent

We’ve got the three stars. We’ve had ourselves a ticker tape parade. And while I don’t think the celebration should end anytime soon, it’s been long enough since the final whistle blew in Vancouver to start thinking about the future of the USWNT.

Using logistic regression on data from all previous WWCs and factoring in position, age, and whether or not the player started the final competitive match in the previous World Cup*, I constructed a model that produces a very rough probability that a player that made the 2015 roster will make the cut for France 2019. Obviously, four years is a long time and a lot can change, so these probabilities are hardly precise and should be taken with a large grain of salt. I have not included goalkeepers, as the very small sample size makes things difficult to predict, and have also left out players that we already know will not play in 2019. Furthermore, I have included forwards but the sample size is also very small, making the probabilities even less precise than that of defenders and midfielders. So at the risk of getting way ahead of ourselves, here are the very rough probabilities that the members of the current squad will be back to defend the Cup in France:

2015 WWC Player 2019 WWC Roster Probability
Morgan Brian 0.87
Alex Morgan 0.8
Tobin Heath 0.73
Julie Johnston 0.71
Meghan Klingenberg 0.66
Megan Rapinoe 0.66
Becky Sauerbrunn 0.58
Ali Krieger 0.58
Carli Lloyd 0.53
Kelley O’Hara 0.22
Whitney Engen 0.21
Heather O’Reilly 0.19
Sydney Leroux 0.19
Lori Chalupny 0.16
Christen Press 0.16
Amy Rodriguez 0.11

Unsurprisingly, Morgan Brian, who impressed as the youngest player on the 2015 team, is the most likely to be back in 2019. In fact, the top part of the table seems reasonable. Where I begin to disagree with my model is Sydney Leroux, who according to the model only has a 19% chance of making the 2019 roster. This seems like a significant underestimation of Leroux’s chances. The same goes for Christen Press. This issue though, is probably due to the small sample size of forwards.

As I noted, this is way too early to start making any serious predictions and the probabilities above are very, very rough. It’ll certainly be interesting to revisit these predictions in the coming years, but for now, here’s to our third star.

Article as appeared on American Soccer Analysis

*This is a proxy for “importance” to the team. The last competitive match is the final, unless the US was eliminated in the semi-finals, in which case the last competitive match is the semi-final, not the third place match!

The data used for this analysis can be found here.

About the author


View all posts

Leave a Reply

Your email address will not be published. Required fields are marked *