I know many dislike Hollinger, but he makes the case, though admits its quite arguable.
From ESPN Insider:
Saturday's tussle between Houston Rockets and Toronto Raptors was an unusual game on multiple levels. Despite injuries, the Raptors' starters came out and smoked the Rockets in the first quarter. But in the second, Houston's second unit thrashed the Toronto subs to the tune of 42-21, and the Rockets kept that up the rest of the game.
So well did Houston's subs play, in fact, that they stayed on the court as a unit for much of the 114-105 win. All five Houston subs played at least 23 minutes, and they rolled as a unit for the first nine minutes of the fourth quarter. The Rockets finished with a 65-23 edge in bench points and a 19-5 advantage in bench assists.
And, as it turns out, this win was surprising on another level, too. It was one of the few times this season the league's most productive bench has been outplayed. Toronto might not be awash in glory this season, but the Raptors can still gloat in at least one respect.
Toronto is the NBA leader in a category I created called "bench rating," which essentially estimates the PER (player efficiency rating) a team receives from its subs. It works pretty simply: Divide starts by total games and multiply by minutes, and you have a coarse estimate of a player's "bench minutes." Multiply that by PER, sum across all the players, and divide by team bench minutes, and you have the team's bench PER for the season.
Although this gives us a handy thumbnail, it still leaves ample room for argument over which bench is the absolute best.
The chart shows every team's bench rating, and, as you can see, the Rockets are no slouches, either. They made a case for themselves last week against the Raptors, and they might see even more production in the future. Rookie Patrick Patterson has hardly played but looked monstrous in that win over the Raptors. Additionally, spark-plug point guard Aaron Brooks just returned from an ankle injury and should boost the rating of Houston's second unit further.
Because the Rockets already have several productive subs -- Brad Miller made the heaviest contribution to the team's lofty rating, but Chase Budinger, Courtney Lee and Jordan Hill also have been pluses -- one can make a strong argument that Houston employs the league's top subs.
As for Toronto, its second unit gets a big boost from the production of Leandro Barbosa, a dynamite scoring threat as a sixth man who plays virtually no defense. Additionally, Jose Calderon and Amir Johnson helped that rating with early-season flourishes off the pine, but both players now start and are likely to continue starting even when the Raptors who are walking wounded return.
It's still a strong second unit going forward. Jerryd Bayless, Sonny Weems, improving rookie Ed Davis and emerging defensive force Joey Dorsey combine with Barbosa to form a solid second five. It's the lack of a star in the first unit that keeps the Raptors lottery-bound.
San Antonio 13.37
L.A. Lakers 13.32
Oklahoma City 12.98
New York 12.10
Golden State 10.96
New Jersey 10.50
L.A. Clippers 10.46
New Orleans 10.14
It's a decent bench but I think the stats are flawed. This team relies on the bench to play well because, for the most part, the players on the bench are of equal talent to their starting counterparts.
It's a strange situation, our bench is above average for a bench, but our starting 5 is below average for a starting 5.
PER is a good measure if you are looking at the overall efficiency of a player, especially on offense. You just never want to be using PER alone as if it's some magic number that encompasses everything.
The Raptors good bench could very well be what's preventing this team from properly tanking. Players like Calderon and Barbosa need to be jettisoned as soon as it becomes a possibility.
It's not PER in and of itself that I have a problem with, but rather the way he uses it. He used PER a few years back to say that Lebron was on par with prime MJ and wasn't being given his due, he's used PER as a way to determine who is supposedly "inarguably" the best player in the league (in which case Chris Paul would be the best player in the league right now)... He circumvents the rules for all these formula's and uses them in places that aren't really effective sometimes. Such as creating this "bench rating" of the overall PER and using it to conclude that we have the most productive bench in the NBA. Now that's all fine and good, and I guess as Chambs said... we do have a pretty good bench. But, again a PER formula doesn't take into account so many different things. What about what the opposing team's PER rating is against our bench? In fact, it doesn't take into account our DEFENSE at all? So that's why I say you can't just plug it in here and tout us as the "best bench in the league" just like that.