The topic of advanced stats and the corresponding movement that has taken the basketball world by storm has been highly discussed over the last two or three seasons. Many stats exist that seek to discover the value each player brings to the court, but one that maybe hasn't been discussed or cited as frequently as some of the others is win shares.
Win shares, created by the fine folks over at Basketball Reference, is an advanced stat that aims to estimate the value of a player based on the amount of wins he brings to the team. The total amount of win shares each player on a team accumulates should equal (or at least be close to) the amount of total wins that the team has. For example, as of writing this piece, the Minnesota Timberwolves have accumulated 19 wins this season and their total amount of win shares sits at 24.1 (to compare, the Golden State Warriors have 54 wins and their win shares are equal to 49.6)
Additionally, win shares can be divided into offensive and defensive win shares; these two stats act functionally the same as win shares, but instead, specifically look at the value each player brings on offense and defense.
It probably comes to no surprise to anyone that rookie sensation Karl-Anthony Towns (6.0) and the oft underrated Ricky Rubio (4.3) are the top two Wolves in terms of win shares. Essentially, these two alone account for nearly half of the team's total win shares (24.1). What is maybe a little shocking is that Gorgui Dieng has accrued the team's third highest win shares total (3.9). Much of this can be attributed to his improved defense this season, as his defensive win shares currently sit second on the team (1.7), behind only Towns (2.1).
Below is a table (retrieved from Basketball Reference) that lays out the win shares of all the players to don a Wolves jersey this season:
So, what does this mean? It may be difficult to draw many conclusions from this one stat (which is something that shouldn't be done with any one stat, anyway) because of how reliant it is on the amount of wins a team accumulates; basically, teams that win more often will naturally have players with higher win share totals. And, using a player's win shares as a proportion of the amount of wins the team has may overvalue good players on bad teams and undervalue good players on great teams. Take Towns and Kawhi Leonard of the San Antonio Spurs, for example. Towns has a value of 6.0 win shares while Leonard has a value of 10.8 win shares, which is the highest on the Spurs. According to these numbers, Towns would be "responsible" for 31.2% of the Wolves victories (6.0 ws/19 wins), while Leonard would only be "responsible" for 21.1% of the Spurs victories (10.8 ws/51 wins).
But what the stat does do well is provide a decent lens in which player value can be viewed within the context of a single team. Comparing the win shares of two players from different teams is potentially dangerous, especially if the two teams aren't all that close in win total. However, comparing the win shares of two players on the same team can help paint the picture of who is the more valuable player. It is not an all encompassing stat, but win shares can be a valuable tool to use in addition to other stats such as offensive and defensive rating, real plus/minus, and other stats of that ilk to determine a player's value.
Speaking specifically about the Wolves, win shares can back up what the eye has been wrestling with all year: Towns is the Wolves most valuable (and possibly best) player on both offense and defense; despite his shooting woes, Rubio is invaluable to the Wolves offense and is a center piece for the team's defense; Dieng is much improved; and Andrew Wiggins is solid offensively, but still has some work to do on defense.
So, while wins shares is not an all-encompassing, perfect stat, it does do a decent job at placing a value on specific players within the dynamic of their team. With the Wolves, looking at win shares doesn't provide any surprises, but it does help reinforce what Wolves' fans have been seeing and feeling.