Per 36 minutes and it's significance

Archaeon · Post #1 » by **Archaeon** » Sat Jun 15, 2013 3:53 am

I went to Basketball-Reference and there this 36 per minute stats. What is it about and how is it difference from per game stats? How would a coach evaluate a player from those stats?

Post #2 » by **bondom34** » Sat Jun 15, 2013 5:55 am

Archaeon wrote:I went to Basketball-Reference and there this 36 per minute stats. What is it about and how is it difference from per game stats? How would a coach evaluate a player from those stats?

In general guys production stays consistent if their minutes change, so it adjusts for guys that play more/less than others and gives them a level playing time. Numbers might not stay at those levels if a guy goes from say 5 minutes to 36, but smaller changes would show much smaller differences. Its a better evaluation than per game for situations where you are comparing players and say for example one plays 30 min per game, the other plays 40, the second guy would have better stats per game, but may not per 36.

Archaeon · Post #3 » by **Archaeon** » Sun Jun 16, 2013 3:50 am

Which stats is more reliable in evaluating a player? per 36 minutes or per game or both?

Post #4 » by **bondom34** » Sun Jun 16, 2013 5:31 am

A bit of both.....for example (this was just the first example I could find, and I'm sure there are better one's but it works)....take a guy like Reggie Evans for BKN. He averages 11.1 rebounds per game. Compare that to David Lee who averaged 11.2 per game this year. From that you would say Lee is the superior rebounder. Looking at per 36 numbers corrects that Evans didn't play nearly as many minutes per game (around 12 less), and shows Evans is the superior rebounder really, with 16.3 per 36 to Lee's 11.0. It really depends what you are looking at, and both can be useful, per 36 just corrects for instances like this.

islamovic · Post #5 » by **islamovic** » Thu Jun 20, 2013 7:16 am

per game is much more reliable than per 36.

per game is players actual production, while per 36 is an estimation.

now heres the problem with per 36. we are to assume that players production remains consistent regardless of minutes played. which most of the time is not true.

theres a reason why some players play 35+ minutes while other play under 20.

Post #6 » by **bondom34** » Thu Jun 20, 2013 2:58 pm

Actually, studies have been done that show generally numbers stay consistent with per 36, barring huge changes in playing time. For the most part, contributions will stay the same per minute, so per 36 adjusts for time played.

islamovic · Post #7 » by **islamovic** » Thu Jun 20, 2013 6:33 pm

but if every player could keep up with his production with increase in minutes, then they woulds all play same amount of minutes, which is not the case.

look at duncan for example, his per 36 are more or less same his entire career, however hes playing 10 minutes less than 10 years ago.

you seriously think tim duncan at age 36 could still post 24/13 with 3 bpg if he played 40 mpg?

Post #8 » by **bondom34** » Fri Jun 21, 2013 12:30 am

No, however it can be used in the opposite direction, to project younger players who can increase playing time. Duncan's decline is due not to skill loss, but age. For what I was meaning, look at a guy like Steph Curry http://www.basketball-reference.com/pla ... yst01.html
His per game numbers are huge this year compared to last, but per 36 he's the same guy. I've seen it referred to as the "Millsap Doctrine" for Paul Millsap, you can google it too. This has been shown to generally be true for players who are put into a starting role when forced to due to a trade/FA loss/injury. As well, its a good stat for simply comparing guys on a per minute basis. It doesn't really work the other way as islamovic is saying though, older guys generally wouldn't be expected to keep a pace, but younger guys can.

Archaeon · Post #9 » by **Archaeon** » Mon Jun 24, 2013 12:43 pm

Why is it that some player play more minutes than others?

islamovic · Post #10 » by **islamovic** » Thu Jun 27, 2013 5:27 am

its mostly due to the ability to produce.

some players can keep up their production with big minutes, others cant.

its up to the coach to determine how many minutes each player can play n still be productive.

giberish · Post #11 » by **giberish** » Thu Jun 27, 2013 5:58 am

bondom34 wrote:No, however it can be used in the opposite direction, to project younger players who can increase playing time. Duncan's decline is due not to skill loss, but age. For what I was meaning, look at a guy like Steph Curry http://www.basketball-reference.com/pla ... yst01.html
His per game numbers are huge this year compared to last, but per 36 he's the same guy. I've seen it referred to as the "Millsap Doctrine" for Paul Millsap, you can google it too. This has been shown to generally be true for players who are put into a starting role when forced to due to a trade/FA loss/injury. As well, its a good stat for simply comparing guys on a per minute basis. It doesn't really work the other way as islamovic is saying though, older guys generally wouldn't be expected to keep a pace, but younger guys can.

Contrary to popular opinion, the default standard is definitely that players produce at the same rates even as their minutes are varied - with some exceptions.

If a guy only plays limited, mostly garbage time minutes his rate stats aren't that reliable at all.

Players with a very high foul rate usually struggle when placed into higher minute roles (trying too hard not to foul out messes with their game).

Fitness issues can limit players effective minutes. For a few players this hits at relatively low minute levels due to being out of shape or nagging injury issues. Most players can go 30 mpg with no drop-off though, but going 35+ mpg or especially up to 40 mpg causes wear-down over a full season though not much on a short stretch of games.

Still, it's generally a safe bet that players who are effective in 15-20 mpg reserve roles can be just as effective in 30 mpg starting roles.

turk3d · Post #12 » by **turk3d** » Wed Jul 17, 2013 1:57 pm

To add a little to what's already been said, it's a way to normalize when comparing players and it's actually and indicator that a guy should be getting more minutes. Unfortunately, a coach can play anyone he wants as many minutes as he wants so even if a guy's per looks good, he may not increase his playing regardless.

The other factor in per stats (which was aolready alluded to) is that the guy getting the less minutes is in 90% of the case a non-starter, so his stats may look good, but they are likely being played against lesser caliber players (bench players as themselves) so if for example, they got moved into a starting role, they might very well go down.

It certainly is something that you might want to try and see how it works (see if the maintain with more minutes). As an example of a player that has had good per36 numbers but has not gotten much for it, see Anthony Randolph numbers. Apparently he runs his coaches the wrong way for whatever reason. He actually did get some chances to start in his career (during a few stretches) and did quite well, His numbers are very good when playing 30 or more minutes.

http://www.basketball-reference.com/pla ... oan01.html

All-in-all, for the most part, I don't think people pay too much attention to that stat (especially the stat people, they have mire fancier stuff they like to use).

Post #13 » by **NO-KG-AI** » Wed Jul 24, 2013 12:24 am

Archaeon wrote:Why is it that some player play more minutes than others?

Not everyone is the same per 36. For the most part, guys that produce the best per 36, are the ones that get the most minutes.

People are quick to dismiss guys that play low minutes because "if he were that good, he would get bigger minutes!" and it's not always that simple.