Question regarding player comparisons

[Biznit]

Thanks to anyone who helps me out on this one. I'm sure to you guys it's a simple answer.
I've gone to 82games and basketball-reference and I see the numbers but I can't figure out how it translates when comparing players.

Player X is avg 15mins more per game than player Y and obviously more ppg and rpg as a result.

But when I compare their per 36 numbers they are basically the same. Does that mean that player Y would be equally as good, if not better than player X if given an equal amount or more mins? If not then what is a better way to compare the two to make an accurate assesment?

Thanks

[Doctor MJ]

With all these things the interpretation is up to you. There are no absolute rules.

Some things I believe though:

-If a player is playing very small minutes, then don't put much stock in the Per 36. Way too much uncertainty.

-If a player is playing 20-30 minutes and putting up great Per 36 numbers, then the big question becomes Why isn't he playing more? If there's an explanation that makes sense that has nothing to do with that player (star teammate at the same position, etc), then that player is a prime candidate for playing more - but you still don't know if he can put up the big minutes until he does it.

-If a player is well into the 30 minute range, then Per 36 numbers are a great way to understand per minute impact when evaluating against another big minute guy. Still though, in terms of actual value contributed, playing more minutes is a good thing.

[Biznit]

-If a player is playing very small minutes, then don't put much stock in the Per 36. Way too much uncertainty.

Very interesting, I totally agree with this.

-If a player is playing 20-30 minutes and putting up great Per 36 numbers, then the big question becomes Why isn't he playing more? If there's an explanation that makes sense that has nothing to do with that player (star teammate at the same position, etc), then that player is a prime candidate for playing more - but you still don't know if he can put up the big minutes until he does it.

Based on your belief then, If a player is in the 30 mpg range and let's say his per 36 ppg is only 2-3 points higher than his per game scoring avg, compared to a guy that plays let's say 15 mpg but whose per 36 almost doubles his per game avg in scoring, would you consider the 15 mpg player irrelivent in comparing the two because he doesnt fall under the 20-30 mpg? Or is that enough to legitimately compare the two to determine who is over all the better player?

[Biznit]

To add....Would the % of shots both players are assisted on versus not assisted have any effect in your mind in comparing two players? If so, is it something people should look at when comparing?

[Doctor MJ]

Based on your belief then, If a player is in the 30 mpg range and let's say his per 36 ppg is only 2-3 points higher than his per game scoring avg, compared to a guy that plays let's say 15 mpg but whose per 36 almost doubles his per game avg in scoring, would you consider the 15 mpg player irrelivent in comparing the two because he doesnt fall under the 20-30 mpg? Or is that enough to legitimately compare the two to determine who is over all the better player?

There's nothing that totally gets you away from "why isn't this guy playing more" when it's that extreme. If I'm coach and a guy is putting up great per minute numbers and I don't know of a big negative intangible keeping him from good net impact, then I give him a chance to play more. As an analyst without that kind of power, I tend to withhold definitive judgment.

Not that I don't think coaches make mistakes, I'm just not confident enough in my ability to say they're wrong very often.

I will give one example though that's coming to mind where more confidence seems warranted: Kevin Love on the Timberwolves. In limited minutes, still putting up big numbers with reasonable +/-. Seemed like he should be playing more, and of course the Wolves have since traded their nominal star Jefferson which should get him more minutes. So the satisfying answer was that they were limiting Love so that they could show off the guy they wanted to trade. With that said, it's still pretty fishy that it was necessary for Love to lose playing time to Milicic and Hollins, and we just saw the FIBA championships where literally everyone online was wonder "Why the hell isn't Love playing more?" because his per minute performance seems to indicate he was best 4/5 on the entire team. So we're seeing multiple coaches find reasons to keep him on the bench for unexplainable reasons, what's going on? I really don't know.

[Chicago76]

Things I look at when trying to determine if a lower minute player is comparable to a higher minute player, other than what has been said:

-usage or % of shots assisted: guys who can't create their own shot often have limited roles on a team for a very good reason.

-defense: the counting stats may look good on a per minute basis, but what if the guy is a horrible defender?

-age and injury history: a great example of this is Rik Smits. Over the last 9 years of his career, Smits consistently averaged between 20 and 22 pts / 36, and this was on a team playing at a slower pace. In terms of scoring ability, his three closest peers over the last 20 years are probably Ewing, Olajuwon, and Mourning. David Robinson at his peak was only averaging a pt or two more per 36 than Smits over the same period when you also adjust for pace. The problem with projecting is that Smits could never log heavy minutes. His feet were horrible and he just wasn't as athletic as the the elite centers of his time. He had a couple of years where he averaged 30-31 minutes, but he generally played around 26-28 while the elite bigs generally played 34 to 40 minutes a night in their prime.

One thing I've done when comparing values of players is to compute their min-adjusted value using a replacement level backup. Given two guys with the same PER (22), one might play 36 a night while the other plays 24. Replacement PER of a decent backup that needs to fill the remaining minutes might be roughly 12. The 36 minute player's value over replacement is 36 minutes / 48 minutes x (22 PER - 12 Replacement PER) or 7.5, while the 24 minute guy's value is 5.