Pelton mailbag: Which playoff contenders have the best benches?

The Rockets have a top NBA bench, but is it the best? Chuck Cook/USA Today Sports

This week's mailbag features the best benches in the NBA, league-leaders in wins above replacement and more.

You can tweet your questions using the hashtag #peltonmailbag or email them to peltonmailbag@gmail.com.

"Bench" is a fluid concept, naturally, but I tried to take each team's healthy starting lineup out of the mix and looked at the wins above replacement player (WARP) for all remaining players currently on the roster and then for the top three of that group by minutes played.

Unsurprisingly, the San Antonio Spurs have the best bench by either measure. Counting Pau Gasol, the Spurs have four players who have rated as worth at least 3.0 WARP this season, with Patty Mills, Manu Ginobili and David Lee (who doesn't even count toward their top three) rounding out that group.

You probably don't think of the Milwaukee Bucks as having an elite bench, but led by Greg Monroe they've been quite good this season. The Miami Heat's success is owed in large part to their effective second unit, and the Houston Rockets have the likely top-two finishers in sixth man award voting (though, of course, they only get credit for Lou Williams' contributions since he was traded to Houston).

The Memphis Grizzlies, Boston Celtics and Atlanta Hawks are all examples of teams whose benches look better if we focus on the top three players and look worse when they go beyond that group. Playoff teams with relatively weak benches include the Cleveland Cavaliers, Washington Wizards (better with the addition of Bojan Bogdanovic and a healthy Ian Mahinmi, but still poor overall), the Los Angeles Clippers (though much better when you count Marreese Speights, who isn't in their top eight) and the Indiana Pacers (a sneaky-terrible bench).

The last team worth discussing is the Golden State Warriors, whose bench has generally been pretty good, but whose score in the top three is tanked by Shaun Livingston's minus-2.8 WARP. (It also doesn't help that new arrival Matt Barnes is in their top three in minutes per game, ahead of the superior David West and JaVale McGee.) I'm inclined to think of the Warriors' score as a limitation of this method rather than a weakness.

I posted a WARP leaderboard in Wednesday's analysis evaluating Kawhi Leonard's MVP case, and, as Krishna notes, Towns ranked fourth, ahead of both Leonard and LeBron James.

Indeed, that has much to do with Towns ranking third in the league in minutes played, a key component in any value metric like WARP. Player win percentage is the per-minute component of WARP, and Towns ranks just 11th there:

For a variety of reasons, there's an unusually large disconnect this season between the best players on a per-minute basis and those who have played the most minutes. That's one of the reasons the MVP race appears so difficult to predict.

"As a fan, I hate to see teams like the Jazz get "punished" by the salary cap by drafting well and growing most of their own talent. What do you think about the following rule for teams, fans, and players: For each year a player plays for a team, his yearly salary only counts for 100 - X *years played, where x is 1-2 percent? I think this is a win, win, win." -- Dane Johnson

I'm not sure I would say the Utah Jazz are getting punished because there's nothing in the rules stopping them from re-signing anyone who wants to re-sign; their issue, much like the Oklahoma City Thunder when they traded James Harden, is much more about willingness to pay the luxury tax over an extended period.

In general, I think a rule like this makes sense. The idea of some sort of designated-player spot making it easier to re-sign long-tenured veterans like Kobe Bryant and Dwyane Wade has been mooted in the past; this is probably a more logical version of that.

The downside is you'd probably have less player movement, which the league had generally been trying to facilitate before implementing the designated-veteran rule in the new collective bargaining agreement. But for most players, this probably wouldn't make a huge difference.

Kudos to Tom Ziller of SB Nation for being ahead of the curve on this trend. Ziller predicted in November that this would be the first season in years -- 46, to be exact, per a December column by ESPN's Marc Stein -- without an in-season coaching change. With a month to go in the season, that's looking likely.

As to why, I'm inclined to start by invoking my razor: Never attribute to causal explanations that which can be adequately explained by random chance. As stark as "never in 46 years" sounds, there have been five seasons (including 2013-14) since the ABA-NBA merger with just one in-season coaching change. One isn't so different from none.

Beyond that, I'd note that few teams have dramatically underachieved based on preseason expectations. A few years ago, Nate Silver studied all coach firings (not just in-season) on FiveThirtyEight and found that more than two-thirds of coaches whose teams finish at least 10 games worse than their preseason over-under total either are fired or resign by the start of the following season.

Only one team is on pace to come close to that mark this year: the Portland Trail Blazers, currently on pace to finish with 37 wins after being predicted for 46.5. Since Terry Stotts is in the first year of a new contract after finishing runner-up to Steve Kerr for coach of the year last season, firing Stotts midseason makes little sense. The next two teams underperforming their preseason line, the Minnesota Timberwolves and Orlando Magic, are doing so by more modest amounts (each by about seven wins), and both have first-year head coaches, whom Silver naturally found are less likely to get fired.