payitforward wrote:But, tell me, have you looked at Giannis's box score data? How about LeBron's box score data? How about, historically, Michael Jordan's box score data.
Should we also ignore their box score data, because "box score data doesn't capture everything..."? Does box score data capture nothing at all, or even just say nothing important, about those players?
To put it slightly differently, don't you think you might be downplaying box score data in the case of this comparison precisely because you have a settled opinion about the players in question? Just something to think about -- not to reject out of hand, nor to accept automatically either, but to think about.
I can't have the same argument with you over and over. Jordan and Giannis are exceptional defenders and primary creators against the focus of defensive attention. Their box score data has to be viewed in that light. Bryant needs other to set him up and furthermore, he needs guys like Bertans to space the floor for him. And he plays terrible defense. It's apples and oranges.
payitforward wrote:nate33 wrote:...Bryant has an on/off differential of -5.7 per 100 possessions. Bertans has an on/off differential of +8.1 per 100 possessions. Arguing that Bryant has been better than Bertans doesn't pass the smell test...
I don't give "the smell test" a whole lot of credence. & I don't give individual on/off numbers much credence either. Wasn't it you who pointed out that... I don't remember who... came off the bench & played against bench players? Isn't that true of Bertans? Should we ignore the fact in this instance, while we accept its relevance in the other?
Or, instead, let me reproduce your statement with a slight difference of players: "Bradley Beal has an on/off differential of -4.9 per 100 possessions. Bertans has an on/off differential of +8.1 per 100 possessions. Arguing that Beal has been better than Bertans doesn't pass the smell test."
Any problem with that analysis? Admittedly, -4.9 is a little better than -5.8. Bertans is plus 13.8 vs. Bryant & only +12.9 vs Beal.
Of course, you wouldn't suggest that the comparison has any significance whatever when it's Bertans vs. Beal. If I used it as data to support the idea that Bertans is a better player than Beal, you would scoff at the idea. But... if the value of a comparison of data can't be judged until we know who is being compared... well, then, that data -- unlike box score data -- is altogether irrelevant.
On/off data requires a good amount of on time and a good amount of off time, and preferably some time alongside starters and some time alongside the bench to be the most accurate. Bertans and Bryant satisfy all that criteria so I think the on/off data for them is pretty sound.
Beal's on/off data is concentrated against opposing starters (while alongside really bad teammates on our starting lineup due the idiosyncratic way Brooks stacks his bench unit). Furthermore, Beal has very little off time that's not also garbage time or time against the opposition's bench.