Seattle Reign FC saw significant roster turnover this off-season, losing a number of fan favorites and long-time players through trades and retirements. This led to debate online about whether and why the team was blowing up the core and rebuilding.
To consider this, we first need to take a step back and determine the parameters of our discussion. What defines a “core player”? Is it someone who has been with the club for multiple years? Is it someone who plays more than 50 percent of the available minutes in a season? For several seasons? Do they need to start at least 80 percent of matches when they’re healthy and available? Some combination of those criteria? This is subjective, so each person probably has a different answer and perhaps we’re not considering the proper question.
A more objective way to consider roster turnover is by looking at minutes played and yearly turnover for each team as a proxy for “core”. This has some shortcomings when important players miss time due to injury or national team commitments (e.g., Megan Rapinoe would not be considered a key player for Seattle in 2015 or 2016) or when someone is signed/traded mid-season, but it should work reasonably well at the level of abstraction we are looking at here.
For simplicity in this analysis, I consider all moves both during a season and in the ensuing off-season while making comparisons between seasons. Thus, a player waived or traded during the season counts the same as one who moves during an off-season, since regardless of the timing her minutes played are no longer on that team the following year. I also treat the WNY Flash and North Carolina Courage as one club with regards to roster continuity, and likewise for FC Kansas City and Utah Royals FC. I obtained all of this data from past game logs and Wikipedia, but there could be some transcription errors or omissions, particularly for players who missed a full season due to injury.
The underlying data is available for further analysis (Google Sheets link) if you would like to delve deeper into these topics or spot any errors. There are undoubtedly other methods that you could use to look at the data — factoring in starts, appearances, or making game day rosters; giving additional weight to long-rostered players; touches on the ball; etc.
NWSL clubs and roster turnover
First, let’s look at each team’s roster stability by season, without worrying about the reasons for turnover. The following chart shows the percentage of player-minutes each team retained for the following season. For example, if Seattle played 10,000 minutes in 2013 and kept players who contributed 5,000 of those minutes on their 2014 roster, they would have 50% retention in the 2013-14 column.
A few things stand out here:
- As might be expected for a new league still figuring things out, every team had significant turnover between 2013 and 2014, with the exception of Sky Blue FC.
- The WNY Flash were huge outliers for turnover leading up to the 2015 season, keeping only four players who logged minutes for them in 2014 and losing nearly 80% of their minutes from the prior season. They made a number of big trades for draft picks and young players which were questionable at the time, but which set them up for the success and roster stability the North Carolina Courage are now enjoying.
- Seattle was both the best team in the league and enjoyed the most stability from 2014 through 2016, keeping players who contributed over 80% of their minutes during those years. We’ll look into that more deeply in a little bit.
- In contrast, the team that stymied the Reign in 2014 and again in 2015, FC Kansas City, had close to league-average turnover throughout its existence.
- Average roster stability has trended upward, and likely would have continued that trend this season if not for the unexpected dissolution of Boston and distribution draft. That doesn’t mean there isn’t still roster turnover, but indicates that more of the turnover is at the bottom end of rosters where players don’t see as many minutes. The implication for expansion is that there should be a deep enough talent pool that teams will not be significantly impacted by losing one or two significant contributors to an expansion draft.
Turnover and performance
From the chart above, you might guess that team performance correlates with roster turnover, since clubs like Houston and Boston tend to have the highest turnover each year. We can look at that in another way and plot a club’s points per game in a given season versus their turnover either the off-season before or after that season. Doing that, we see that teams that do well tend to have lower roster turnover the following year compared to teams that perform poorly.
As the plot above shows, approximately 36% of roster turnover can be explained by a team’s performance in the prior season. That’s to be expected, since good teams want to keep doing well but players depart clubs for many reasons and only some are connected with performance.
In contrast, there is essentially no correlation between roster turnover and team performance the following season.
Again, this is unsurprising, since there is a plethora of high-level talent available outside of NWSL which teams can sign, so roster stability is less important in the NWSL than it is in sports where there is one indisputable top league, such as the NBA.
Reign roster stability and methods of player loss
Now let’s turn our attention to Reign FC and consider the initial question of whether the team went through a rebuilding or retained its core, going all the way back to its first off-season, as well as how players departed the club each year.
There are many ways a player could depart a team, some of which are player initiated, and some of which are team initiated. Player-initiated departures includes things like retirement or opting to sign in a different league once an NWSL contract is finished. On the team-initiated side we have moves such as trades, contract options which aren’t picked up, or players being waived. There are also a few methods that don’t fit into either of those categories, including losing players in an expansion draft or having them reallocated to another team back when the league tried to balance the number of subsidized players per club.
I will not attempt to parse out the rationale for each trade or whether any of the players involved wanted to leave one club or join a different club; nor will I consider situations where a player did not play for the Reign before being traded again (e.g., unsigned draft picks, or Kristie Mewis when she was with Reign FC for 10 days back in November 2013.)
The following chart looks at the percentage of player-minutes that were retained, traded, retired, or otherwise no longer with the team the following year. For example, players contributing approximately 55% of minutes in 2013 were retained on the 2014 roster, as shown by the dark blue portion of the bar in the 2013-14 column.
Seattle had 17 players on their roster at some point in 2013 who were no longer with the club in 2014. The vast majority of those players (10) were traded either during 2013 or after the season, and included three players who logged at least 1000 minutes – Christine Nairn, Kaylyn Kyle, and Liz Bogus. Five players were waived or not re-signed for 2014, and two more were deallocated.
Most observers would agree that this was a major rebuild, and since the team was only one year old there wasn’t much of a “core” per se.
After a hectic first year, Reign FC led the league both in wins and in roster stability, keeping 14 players after 2014 who contributed nearly 82% of minutes. Of the remaining 18% of minutes, 12% was lost in trades (Sydney Leroux and Kate Deines), while Naho Kawasumi and her 6% returned to Japan after her loan ended.
The club lost four additional players from the 2014 roster – Amanda Frisbie was traded with Leroux but played zero minutes for Seattle due to injury, Megan Brigman and Holly Hein were waived but played only 17 minutes between them, and Carmelia Moscato was deallocated by Canada and not re-signed after playing only 12 minutes and suffering a significant injury.
The 2015-16 off-season saw more of the same, with Seattle continuing to set the standard for roster stability. Reign FC retained 15 players who contributed 82% of the minutes played in the prior year. Seattle traded only one player who played for them in 2015, sending Amber Brooks and her 421 minutes to Houston as part of an expansion draft gambit which protected the rest of the roster.
The biggest losses were due to retirements, with the club losing Steph Cox (1800 minutes), Mariah Bullock (451 minutes) and Danielle Foxhoven (359 minutes), who combined for just over 13% of minutes in the prior season. The two other departures were Katrine Veje (421 minutes), who left the league almost as quickly as she arrived; and Caroline Stanley, who was released after making one appearance for 90 minutes.
Unfortunately, all good things must come to an end, and Seattle missed the playoffs in 2016 and had significant roster turnover between 2016 and 2017, losing over 38% of minutes from 2016. Only one rostered player was traded – Havana Solaun to the Spirit – and she played only 272 minutes for Seattle in 2016. Three players were waived either during or after the season (Andi Tostanoski, Michelle Cruz, Paige Nielsen), but combined they only accounted for 215 minutes (1%) of playing time.
The vast majority of departures were due to retirements or players taking a leave of absence (Keelin Winters, Manon Melis, Hope Solo), or players whose contract was up and who opted to sign with a team in another league (Kim Little, Kendall Fletcher). Each of those categories accounted for nearly 18% of minutes played in 2016.
Thus, although the team retained nearly as many players (13) as they did in the prior two off-seasons, the types of players who were lost were significantly different and were much bigger contributors.
This off-season was a flashback to the original rebuild back in 2013-14. Reign FC retained only 10 players from 2017, while 13 players who contributed nearly 49% of minutes last year are no longer on the roster. Although this is not too much more turnover in terms of minutes than 2016-17, the reasons for those departures were vastly different.
Seattle traded eight players who logged over 10,000 minutes in 2017 (43% of all minutes played), and waived an additional three players who contributed 1175 minutes. There were two retirements, but Elli Reed and Madalyn Schiffel saw only 191 minutes of play during 2017.
So while this off-season wasn’t much different from 2013-14 or 2016-17 in terms of rebuilding, it felt much different to fans, in large part because so many of the moves were trades of players who had been with the team for multiple seasons.
However, much of this is a reflection of unusual roster stability in past seasons, combined with a club preference for older, more experienced players, which meant that Seattle faced more turnover than typical once that long-term core aged and broke up. By other metrics, Seattle’s roster remains one of the the most stable long-term in the league.
Seattle has the most 2013 originals still on their roster (Lauren Barnes, Kiersten Dallstream, Jess Fishlock, and Megan Rapinoe, plus returnee Michelle Betos). The next closest club is Utah, who has three FCKC originals (Nicole Barnhart, Becky Sauerbrunn, and Erika Tymrak) plus returnee Desiree Scott. Portland has two originals, Chicago and Washington have one each, North Carolina has McCall Zerboni — who played for Portland in 2015 and Boston for part of 2016 before returning to WNY, and Sky Blue has no original players remaining.
Another way to look at roster churn is to consider average number of seasons each current player has been on a team’s roster. This metric is complicated somewhat by the dissolution of Boston and unbalanced dispersal of their players onto other rosters, but Seattle ranks highly in terms of tenure if you do not take those players into consideration. Current Reign FC players have been with the club for an average tenure of 2.94 seasons, second only to Chicago. Adding Reign FC’s four former Breakers drops the team average quite a bit, but they are still near the middle of the league in terms of player longevity.
Average player tenure (seasons) with or without Boston dispersal signings
|Club||with Boston||without Boston|
|Club||with Boston||without Boston|
In summary, there’s little question that this was a rebuilding year for Seattle Reign FC. A new coach and 14 new players combined with a core of only 10 returning was a seismic shift from what fans were accustomed to for several years, but some of the changes were probably overdue. The 2013-14 Seattle rebuild, as well as the 2014-17 WNY/NC rebuild, show that there are multiple ways to be successful in the NWSL, and should give Reign fans reason for optimism for 2018 and beyond.