(If you're not up to speed, check out the series introduction here.)
It's not whether you win or lose, it's how you play the game. -Grantland Rice
With all due respect, Grantland Rice is full of it. Winning is what matters in the NBA. Winning separates the teams that ultimately matter from the teams that don't. Winning is the literal barometer of a team's level of success. If winning didn't matter, why even keep track of the score?
While we may agree that winning is important, agreeing on how valuable winning is can be tricky. After all, there are lots of variables that go into team's pursuit of the almighty W, and it can be hard to keep track of them.
But let's not forget about what we know, full stop, about winning in the NBA. Wins are a finite resource, and the one thing all teams have in common is that they use salary to pay players to go out there and win. So which teams make the best of what they have?
To answer that question effectively, we need to nail down how we're going to define "value". Again, as the annual "What does the V mean in MVP?" conversation shows us, this isn't easy to do! But at its core, value is comparing what you get to what you spent; if you spend less to get something, you got better value than someone else who spent more for the same thing. For this exercise, win shares (WS) are what we're buying, actual wins are what win shares convert to, and salary ($) is what we're spending.
So let's start with the basics. In the 2015-16 NBA season, the thirty NBA teams spent roughly $2.353 billion on salaries for 564 players. As with any regular season, those thirty teams split up 1,230 wins based on, you know, actually playing basketball against each other. League-wide, that can be converted to the statement that each win was roughly "worth" $1.913 million.
To respond to a number of comments from the introduction, I fully admit that measuring players' contributions to team wins is imprecise; win shares gets us close, but a close game can can flip a win into a loss (or vice versa) without affecting a player's WS performance. We mentioned that the league saw 564 players under contract last year, and those players amassed a total of 1219.70 win shares. A small difference from the 1,230 total wins that were won, but it's important to specify the difference nonetheless. As a result, we conclude that in 2016, 1.0 player WS = $1.93m salary.
As you can imagine, value is not static from season to season. For those who are interested, here is a table showing the same calculation of league average WS value over the past 5 seasons and a rough estimate of the next 2 seasons (using projected salary data for 2016-17 and 2017-18):
|Season||League Average WS Value|
*based on 66 total games: 2011 NBA lockout **projected
Taking it a step further, we all know that not all teams spend salary equally. Some will struggle to stay below the tax line, while others will fight to stay above the salary floor. Likewise, not all players will contribute according to their salary scale. An aging veteran with an eight-figure deal might put up the same WS as a rookie second round pick, which is either great for the rookie or terrible for the vet. With all that said, we will need to break down each team's WS and salary totals individually, and compare it to the league average that we determined earlier.
(One quick reminder, as this exercise relates to value: the less you spend for the same commodity, the better the value. As a result, the teams that spend the least on wins can be said to have gotten the best value, whereas teams that spend the most have gotten the worst value.)
I could tell you to do the math and get back to me, but I'm not like that. For your convenience, check out the table below (ranked by Team WS Value for the 2015-16 NBA season):
|Rank||Team||Wins||Total Roster WS||Total Salary Paid||Team WS Value ($)|
Some of the outliers league-wide are obvious: the San Antonio Spurs and Golden State Warriors each had high payrolls, but won so dang many games that it pushes their averages down below the league standard. Seeing Toronto, Utah, and Portland in the top five of WS value is surprising at first glance, but all three teams had a number of contributors on smaller deals. At the other end of the spectrum, the Los Angeles Lakers paid an awful lot of money for barely any production, spending nearly $10 million per WS! The Philadelphia 76ers and Brooklyn Nets were also terrible, paying nearly double the league average for a single WS relative to their W/L record, and the New Orleans Pelicans and Phoenix Suns round out the bottom five.
One season doesn't necessarily reflect how well a team has done, so we also put together each team's salary and WS data for the past 5 years, going back to the lockout-shortened 2011-12 NBA season. That data is in the next table below (sorted by Average Team WS Value from 2011-12 to 2015-16):
|Rank||Team||Avg Wins (2011-2016)||Avg Roster WS (2011-2016)||Avg Salary Paid (2011-16)||Avg Team WS Value (2011-16)|
This data isn't some great revelation. The good teams are good, the bad teams are bad, and both tend to get good/bad value from the contracts they sign their players to. (Sorry Kobe Bryant fans, but that last contract the Lakers gifted him was absolutely a bad move...but we'll get into individual player value later.) However, the fact that a correlation exists between value (as we've defined here, relating to both salary and WS produced) and team success is proof enough for me that this is worth further examination.
And since this is a Bucks site, let's take a closer look at the Bucks' overall performance since 2011. Curiously ranking dead-center at 15th, they have a pretty low average win total and terrible WS production from their roster (at 29.62/season, the Bucks would be 6th worst in the NBA!)...but because their average salary has remained so low (29th in the league, a shade higher than the Utah Jazz), they actually have gotten relatively "good" value out of the contracts they've signed players to. Part of that is the consistent youth the team has had recently, and another part is the relative lack of big money free agents the team has signed since 2011, but the fact remains that the Bucks have (at least) not vastly overpaid for poor production.
Eric: Here are some of my thoughts:
- It is very tough for your roster to have value if you don't win more games than you lose. I don't think it's impressive, but seeing the Bucks in 15th in the five-year value chart despite averaging just 31 wins was interesting to see.
- Paying a lot or a little for your roster doesn't seem to have an effect on the value of your roster.
- That being said, if you pay a lot for your roster, your players better perform. As is often discussed here on this website, Giannis Antetokounmpo and Jabari Parker are going to get paid a lot of money in the next five years. When they start receiving that money, they need to make a major impact on the floor.
- It's going to be fascinating to see where Portland ends up on the first list next season. Almost every summer, I'd assume we would be able to find a team high on the value list that didn't pay a ton of money for their roster and had young players outperform cheap contracts. Then, we'd see that same team decide to spend a ton of money on free agents and probably find themselves a little bit lower on the value list a year later.
- I tried really hard to draw some conclusions about coaching with the idea that great coaches could somehow help you get more "value" out of a roster, but I failed at finding any sort of pattern.
- If Mitchell and I continue to write on this website for the next five years, we should probably continue to do this exercise each summer. "Value", or at least that the way we defined it in this piece, is going to be of the utmost importance as the salary cap explodes. It could also be highly volatile in the coming years as the entire league tries to figure out how to build teams as the cap rises and then falls a bit after the initial bubble.