My theory on the lineups that play worse is that they have two guys who can’t hit the broad side of a barn. Like, Noah and Thomas, their effective field goal percentage is like 30 percent. Dampier and George, those two have no jump shot. I’m not positive, this would take a lot of statistical analysis, but I think that’s what we’re seeing. That’s the key.I'm not sure why he includes Noah there because his effective field goal percentage ranks up there with the very best in the league. While I appreciate the Theorem's simplicity, I do not find it all that ground-breaking. Each team has only five players on the court. If two of them are horrid shooters and offer little else, that's going to case a problem.
As a reminder, a poor shooter is different than a low scorer. A poor shooter wastes valuable possessions whereas a low scorer could be a high-percentage shooter who picks his spots in the offense. Considering a possession is the very lifeblood of a basketball team, a player who throws away possessions will directly hurt their team's chances of winning. Imagine a baseball team hat only had five hitters in the lineup and two of them made outs 80 percent of the time. That's the sort of impact we're talking about.
Let's see if we can find some validity to the Winston Theorem. The first step is to find the most terrible shooters in the league. Using the Basketball-Reference.com Full Court tool we can do just that. We'll go along with Winston and use effective shooting percentage (eFG%) as the measuring stick for poor shooting. One could use other measures but I think this will serve our purpose. eFG% differs from raw FG% in that it adjusts for the marginal point value of a 3-pointer rather than treating all field goals the same. I don't want Gilbert Arenas or any others who played only a handful of games on this list so I'll (somewhat arbitrarily) filter the query to those who played at least 20 games which is equivalent to a fourth of the season. I also want players who not only get on the floor but stay there too so let's cut it players who play at least 15 minutes per game. Here are the thirty worst shooters in the NBA last year.
The Celtics took a big gamble on Marbury and it was clear that some combination of the extended time off the court coupled with the unfamiliar reduced role ate away at his effectiveness. Consequently, his playing time in the playoffs was sliced just about in half to only 11 minutes per contest. You'll notice teammate Glen Davis and his .444 eFG% finds himself on this list as well. According to 82games.com, they shared court time for a total of 212 minutes last season and surprisingly, the Celtics outscored their opponents by 11 points over that period. Translated to a 48 minute interval, they beat their opponents by roughly 2 and half points on average. The +/- numbers cited here and henceforth are not opponent or teammate adjusted but it's fair to assume this poor shooting combo was not as destructive as Winston would imagine.
There are quite a few teammate pairings on this list so we have some data to work with even if the shared playing time isn't as much as we'd like. Amazingly, the Clippers and Thunder each have four players who got big minutes despite their struggles to make baskets. Not coincidentally these two teams ranked dead last in Team Offensive Rating last year (ORtg = points per 100 possessions). The Wizards have a three poor shooters who got lots of court time so it's no wonder they also find themselves with one of the worst team offensive efficiencies in the NBA. Outside of the Clippers, Thunder, and Wizards there are four other squads who have pairings on this list: the Grizzlies, Timberwolves, Bucks, and 76ers. Will these pairings show the same net positive relationship as Glen Davis and Stephon Marbury? I doubt it.
It's worth saying that not every good player will post a positive plus/minus. In fact, no players on the Wizards, Clippers, and Thunder had enough positive influence to cause their team to outscore their opponent. So for this exercise, we're not particularly interested in the individual players plus/minus. Instead, the real focus should be centered on the pair's plus/minus as it relates to the two individuals by themselves. For example, despite their inability to shoot very well, the on-floor tandem of Darrell Arthur and Quinton Ross actually raises the Grizzlies quality of play to beat their opponents by about 5 points every 48 minutes on average. Winston's Theorem, as I understand it, would suggest otherwise.
It's interesting that Russell Westbrook and his miss-inclined counterparts aren't noticeably worse when they're on the floor together. If anything, the Thunder are better off when Westbrook gets paired up with Desmond Mason, another player can't shoot a lick.
As for the Clippers, Baron and Ricky Davis each had their worst shooting performances of their career last season but the 389 minutes they shared on the floor were not, to my surprise, 389 too many. Each saw their shooting percentages increase with the presence of their Davis counterpart. Baron's responded by shooting more while Ricky shot less. The most jarring tandem in this sample has to be Ricky Davis with his backup big Brian Skinner. To see this effect, let's contrast Ricky's production with Brian Skinner on the floor as compared to his other 6'9" 250 lb bruiser Zach Randolph (stats are per 40 min unless otherwise noted).
With the limited information at hand, we can't know for sure why Skinner has this effect on Davis but it's certainly something a coach would want to know about. A possible reason for the difference could be situation-related. The game could be already out of hand when the two share the court and Ricky takes it upon himself to take risky, low percentage shots. Conversely, Davis may not experience the same temptation with Zach Randolph from a game situation perspective. Can't know for sure.
The data here does not provide much evidence that the Winston Theorem is an airtight assertion. A clear exacerbating effect of two poor shooters on the floor does not appear to afflict these teams when we look at the player pairs plus/minus numbers. Playing two players who can't shoot could have devastating effects, all things equal. But if a player can't shoot, he better be able to contribute in other ways or else he'll never receive the the opportunity to contribute. Russell Westbrook can still offer dynamic skills that mitigate his shooting woes even if another bad shooter comes in.
In the end, I don't believe two poor shooters will cripple a team. A pair of one dimensional poor shooters, on the other hand, will.
Thanks to Basketball-Reference and 82games for the data.
0 comments:
Post a Comment