Route Two Soccer: Replacement Level in the NWSL – Part 2

FacebooktwitterreddittumblrmailFacebooktwitterreddittumblrmail

Last week I discussed the theory of replacement level valuation, and described some general ideas about how it can be usefully employed to think about the NWSL. This week, I want to dig into things a bit more, with four observations about player value in the league, informed by the idea of replacement level.

1. Setting replacement level in a precarious league

Replacement level is not static. If overall talent levels improve, so should the replacement level. And in a league like the NWSL, with a precarious employment structure, the movement is probably more significant than in other more settled systems.

While early retirements are growing less common, it remains true that plenty of good players leave the game for reasons that have little to do with their abilities on the pitch. Christina Gibbons’ recent retirement is a good example. She wasn’t forced out by lack of quality; she left because the hassles of trying to maintain a professional career for little money and no amenities seem to have overwhelmed the desire to play.

A league which can’t necessarily compete with economic opportunities in the private sector will naturally suffer more turnover and loss of talent than one where players make hundreds of thousands as a baseline. That in turn means that marginal player availability is often determined more by the willingness to accept a lack of compensation than by a strict accounting of ability. As a result, simply showing up and turning in 90 minutes is generally worth more in the NWSL than it would be in a league with fairer compensation.

However, as time goes on and standards improve, more players are willing to stick it out. This means that replacement level is going up over time. A player who was modestly above replacement level in 2014 might not be any longer, simply because the overall tide is rising.

2. Measuring modest contributions

The concept of replacement level is a useful device to square some circles within discussions about NWSL talent. Too often, conversations exist in a framework where a player is either excellent or useless, without any clear sense of the space in between. But in fact, the league is full of players who are contributing modest value, without necessarily rising to the level of average.

In a league with nine teams, there are probably only about 50 players who could reasonably be described as average or better. But 188 players received minutes last year. How do we account for the players in the middle of the pack? Here I’m thinking of players like Adriana Leon, Joanna Lohman, Christen Westphal, Amy Rodriguez, Rebecca Quinn, Thembi Kgatlana, Brooke Elby, etc.

With any players, it’s obviously important to look at context. What did their team ask of them, how well did they fulfill their role, what alternatives were there? What sort of potential do they have? Obviously, Andi Sullivan is a different sort of player from Brooke Elby, and it wouldn’t be helpful to pretend their total contributions could be measured by one universal metric. At the same time, if you want to tell the story of the season, it’s useful to have some form of cross-contextual comparison.

So with players like this, one useful perspective might be to emphasize that they logged important minutes, and provided meaningful value, to the extent that they performed above the replacement level, while also recognizing that their contributions were probably below the average production levels in the role. This can then be supplemented with more specific evaluations.

For example, while Elby and Sullivan both contributed some value, Elby was only expected to be a role player. She was selected 23rdd in the Breakers dispersal draft—almost literally the definition of a replacement player. That she contributed real positive value helped her teams enormously. Sullivan, meanwhile, was the top pick in the draft, and Washington was counting on her to step into the pro game immediately. While she wasn’t hopeless (she did contribute real value), merely being above replacement level was extremely damaging for the Spirit who needed more.

Sullivan clearly has the higher ceiling, and likely will have many strong seasons to come. But in 2018, her performance hurt Washington a lot because they were counting on more.

3. Replacement level variations across roles (the problem of too many good forwards)

Another important feature of an analysis informed by replacement level: emphasis on the distribution of talent across roles. Specifically, the imbalance between attacking and defensive talent. Because the reality is that the overwhelming percentage of top-quality players in the league fill attacking roles. This is partly a feature of the game itself—where individual brilliance matters more in the attack, while team structure matters more in the defense—but it’s also a consequence of the developmental structure in the US system. With college still the dominant training system, players are free to continue as forwards long past when they might have been forced to switch in a world where the pipeline narrowed earlier.

Whatever the cause, it’s clear that the league is stacked with attacking talent. Unfortunately for the players below the top tier, this significantly reduces their value, because replacement level is fluid and depends on the actual distribution of talent.

For a given team, their 4th or 5thh choice striker is probably going to be close to replacement level. She is the player who can perform satisfactorily and do a job, but is well outside the top talents in the league. But look at even a team as hapless as Sky Blue and realize that they have Naho Kawasumi, Carli Lloyd, Savannah McCaskill, and Imani Dorsey. Not to mention Jen Hoy. And McKenzie Meehan. And Paige Monaghan and Kyra Carusa coming in.

These are all very good players. But when you do the same exercise across the league, you realize that every team is objectively stacked in their attack. Unfortunately, though, ‘stacked’ is ultimately a relative term. The problem for Sky Blue isn’t a lack of excellent players in the attack; the problem is that teams like North Carolina and Chicago are even more absurdly blessed.

But this means that there are dozens of genuinely great attacking players who not only can’t get a regular starting job, they’re not even particularly close to one.

Adriana Leon is particularly apposite example here. When the Boston Breakers folded, she entered the dispersal draft and fell to the 18thh pick. That felt low to many people, who pointed out her six goals for Boston in 2016. When she found no playing time in New Jersey and was eventually traded to Seattle for a 4th round pick, there was more outcry. A seasoned striker, a Canadian international no less, had to be worth more than a low draft pick.

But thinking about it in terms of replacement level can help clarify things a bit. Because the reality is: six goals in a season notwithstanding, Leon simply doesn’t have that impressive a record over her career. In 83 NWSL games, she notched 10 goals. That’s not nothing, but given the same opportunity to occupy a roster space, many other players might have found the net far more often, or contributed in other ways. Ultimately, strikers just aren’t scarce in the NWSL, while opportunities at striker very much are.

None of which is to suggest that Leon isn’t a good player. She certainly is. It’s just that she’s not clearly comparatively better than the many other good players who can fill the same role. Probably every team in the league would be happy to have her, but none of them would be willing to give up much (or any) value to do so. That’s because they’re not assessing her talent on an absolute scale; they’re looking at it comparatively.

Compare this to defense, and things look very different. Here, the replacement level is much higher as teams struggle to fill out their roster with players who can plausibly handle the job.

This is a big part of the reason why teams consistently try to shift attacking players back into more defensive roles. They’re trying to take advantage of their overabundance in one area to bolster themselves in another space. And it’s why positional flexibility is very helpful for marginal players. The ability to step into multiple roles increases their potential value over replacement.

4.Replacement level in a World Cup year

As we know, the World Cup takes place this summer. And just like in 2015, the NWSL will continue amidst the tournament, despite the removal of three dozen or so of its best players for a substantial chunk of the season. What’s more, the removal of talent is by no means balanced. Teams like North Carolina and Portland will lose most of their starting XIs, while Sky Blue, Washington, and Houston will be significantly less ravaged (as always, the wonderful NWSL roster sheets maintained by Jen Cooper are crucial here). To some extent this will have a balancing effect on the league, pushing everyone toward the middle.

But it also depends on how well teams manage their replacement markets. Team depth is always important, but especially in a season like this, when it will be significantly more tested. And it’s a reminder that ‘replacement level’ as an abstract concept is never quite the same as the actual replacement level for a specific team. Those that play the game well will assemble supporting casts that are better equipped to step into the breach. It’s a reason to pay special attention to the preseason this year, because that’s the time when the league’s replacement talent (recent college graduates, trialists, part-timers, etc.) have the chance to make their case.

Follow us on Twitter