Knicks Morning News (2016.07.19)

  • [ESPN.com – New York Knicks] Tuesday's Knicks Links: Revamped roster excites Carmelo Anthony (Tue, 19 Jul 2016 05:24:43 EST)

    Tuesday's Knicks Links: Revamped roster excites Carmelo Anthony

  • [ESPN.com – New York Knicks] Melo: 'As a country, we have to stand united' (Tue, 19 Jul 2016 03:33:13 EST)

    Melo: 'As a country, we have to stand united'

  • [New York Times] Carmelo Anthony Is Using His Voice and Olympic Platform for Social Change (Tue, 19 Jul 2016 05:29:39 GMT)

    While training with the United States basketball team that will be headed to the Rio Olympics, Anthony spoke about his activism against gun violence.

  • [New York Times] 76ers Believe Future Could Be Bright After Dismal Seasons (Tue, 19 Jul 2016 07:21:39 GMT)

    Without playing a game, the 76ers will still spend this summer in the cellar.

  • [New York Times] Valentine’s Buzzer Beater Gives Bulls Title at Summer League (Tue, 19 Jul 2016 07:03:36 GMT)

    Denzel Valentine put a thrilling finish on the NBA Summer League.

  • Liked it? Take a second to support Mike Kurylo on Patreon!

    Mike Kurylo

    Mike Kurylo is the founder and editor of KnickerBlogger.net. His book on the 2012 Knicks, "We’ll Always Have Linsanity," is on sale now. Follow him on twitter (@KnickerBlogger).

    74 thoughts to “Knicks Morning News (2016.07.19)”

    1. Per the NY Post:

      The starting five of Rose, Lee, Anthony, Porzingis and Noah is probably as good as all but four clubs in the Eastern Conference, according to some NBA scouts. The season hinges on their health — specifically that of Rose, Noah and Jennings.

      That would be something. I would think that three of those four teams are Boston, Cleveland, and Toronto. I’m not sure who the scouts are suggesting is the fourth, maybe Atlanta?

    2. it’s hard enough having an intelligent conversation when people try to make facts fit theories rather than theories fit facts, but when some people don’t even want to accept facts as facts…

      Which is this fact that’s not acceptable as a fact ?

    3. @2

      Probably the fact that you essentially argued that per/36 and per/48 weren’t per game stats when, in fact, per/48 literally IS per game (given that, you know, games are 48 minutes long), and per/36 is simply a way of attempting to parse out per-minute production for dudes who don’t play starters minutes.

    4. Also…to clarify why Dred was telling you to find another blog was that, if you’re going to essentially say that stats, specifically advanced metrics, don’t matter to you at all then maybe you should seek out a blog that doesn’t implicitly state right from the jump that this is a forum for, in large part, statistical analysis of the Knicks and basketball as a whole.

    5. As far as i know “per game” stats are “Total stats divided by games played”.
      They are actual stats that were made in the court.

      “Per36” and “per48” are hypothetical stats that are being projected by multiplication of the stats that a player made in less minutes.

      Am i wrong ?

    6. The EC has had so much player movement and good, developing stars (KP, Stanley, Myles, Giannis, Kaminsky, Portis, Hezonja, etc.), it’s hard to predict anything. Charlotte could surprise, so could the Bucks. Atlanta could take a step back. Boston? Who knows? The only two teams that have roster stability is the Cavs and Raptors, everybody else is totally unpredictable.

      I think the EC is going to be crazy competitive, much more so than the WC, which is pretty much over on Day 1. The Cavs may still be the team to beat, but I wouldn’t be surprised if they take a step back.

    7. Knew Your Knicks, to quote from the movie Platoon, “you ain’t never been right…about nuthin‘!”

    8. Knew Your Knicks, to quote from the movie Platoon, “you ain’t never been right…about nuthin‘!”

      But I’m tryin’, Ringo. I’m tryin’ real hard to be a shepherd.
      [From Pulp Fiction]

    9. “Per36? and “per48? are hypothetical stats that are being projected by multiplication of the stats that a player made in less minutes.

      Am i wrong ?”

      When you use the word “hypothetical”, yes, you are wrong.

      But you’re not exactly wrong about the nature of per game stats, but you are very wrong if you think that somehow looking at total counting statistics divided by games played is somehow “real” whereas viewing per-minute production (especially per/48) are “hypothetical” and meaningless.

    10. When you use the word “hypothetical”, yes, you are wrong.

      You don’t like the word.ok.
      How about “Projected” ?

      “”But you’re not exactly wrong about the nature of per game stats, but you are very wrong if you think that somehow looking at total counting statistics divided by games played is somehow “real” whereas viewing per-minute production (especially per/48) are “hypothetical” and meaningless.””

      I’m not saying that per48 are meaningless stats but i’d say that many times are very misleading.
      Check out Fredette’s,Antetokoumpo’s,OQuin’s,Seraphin’s,DWilliams’,JGrant’s “per36”:
      http://www.basketball-reference.com/teams/NYK/2016.html

      Would you trust those “per36” stats to make future decisions about your team’s roster ?

    11. If I’m a salesperson who averages 8 new accounts added per week and the guy in the cubicle next to mine averages 12, you’d look at those per/week numbers and assume that my colleague is just more productive than me. But if it’s a part time job for me and I only work 2 days a week and my colleague works the traditional 5, then my per/day production is higher than my colleague.

      Sure, you can then drop the appeal to authority argument and say that if I was capable of equalling my colleague’s production over the course of a full work week then I’d obviously be getting more shifts per week but…well…do I even need to get into how flimsy that argument is?

    12. “Would you trust those “per36? stats to make future decisions about your team’s roster ?”

      Not without context, but I’d certainly use them to help me determine if player X might represent terrific value because his lack of minutes/opportunity may be because of some extenuating circumstance (dumb coach, established “stars” ahead of player X on the depth chart, youth, and things as stupid as draft position, pedigree, name recognition, etc), and not because he’s a garbage player who could never produce at a high level if afforded greater opportunity.

      Nobody on this board, not even Jowles, truly argues that stats can be used to analyze players and inform roster decisions without context.

    13. KYK, It has been proven beyond any reasonable argument that per-minute stats are more a more statistically valid basis for player comparison than per-game stats. If you want to hypothesize that players who play much fewer minutes are getting less minutes because their production is unsustainable at a higher minutes load, that’s been disproven time and time again.

      There are exceptions to any statistical rule, but those are usually outliers. I, for one, think that Wilt Chamberlain’s greatest achievement was averaging 48.5 mpg for a season while averaging over 50 points per game and 27 rebounds per game being triple-teamed every night at a much faster pace of play than in today’s game. Especially at a time when players often travelled by bus and played back-to-back-to-back-to-back-to-backs in smoky, non-climate controlled arenas wearing Converse All-Stars and having primitive training and medical technology. (Not to mention while doing his thing with the ladies in his off hours.)

    14. @BANG! Like Mike Breen

      Why did we “lost” DWilliams since his “per36” stats look pretty solid ?
      Will we miss him ?

      KOQ has 14,6pts 11,8reb 2,3blk 3,4ast “per36”
      Is he so good ?

      JFredette averages 50,4pts per36…
      ?

    15. Yeah, there’s at least a few studies that show that the vast majority of players continue their per minute production under greater minutes loads except in certain edge cases (players that play like 2 minutes a game, or have very low MP over the course of a season being the examples that immediately come to mind.) It’s pretty well established that per minute stats offer better insight into a player’s box score than counting stats.

      Jimmer and grant and d will probably would approximate their (in Jimmer’s case, career) per 36’s if they did play 36 a night, but what makes them bad is not that they’re not attaining those raw numbers, but rather that they’re either doing so inefficiently or aren’t impacting the court in other ways.

    16. Knew Your Knicks…

      I will add that I’m not trying to mock you or anything and I don’t know how familiar you are with the modern NBA but these metrics are used by nearly every NBA Front Office, along with traditional scouting (eye test shit), and, occasionally, common sense when constructing their rosters.

      Also, as previously stated, they form the essential framework for discussion on this specific blog and many of us come here to engage in intelligent debate (or, in my case, mostly observe intelligent debate) that centers on evaluating these metrics and what they might mean to the team we all root for and the game we all love (presumably).

    17. Why do you not understand that no statistic is infallible. If you only want to cherry pick the numbers of guys who look better in per-36 than they do in their actual production on the floor you can find a lot of those examples. That still doesn’t make per-36 useless or stupid.

    18. Again, I’m not claiming these stats are infallible or should be used without context.

      “Why did we “lost” DWilliams since his “per36? stats look pretty solid ?
      Will we miss him ?

      KOQ has 14,6pts 11,8reb 2,3blk 3,4ast “per36?
      Is he so good ?”

      I had high hopes for KOQ given his per-minute production in Orlando and still feel that he would probably come close to maintaining that production over a larger minutes allocation…however, this is where those extenuating circumstances and common sense elements factor into the discussion and analysis. It seems OQ is a guy who rubs coaches the wrong way (anecdotally, we’ve heard rumblings that he doesn’t really bust his ass in practice but who knows). He also makes a lot of boneheaded decisions on the court which could reduce a coaches trust in a player and put a cap on his MPG.

      Do I think we’ll miss DWill’s efficient (if one-dimensional) scoring off the bench? Well…that will depend entirely on the kind of contributions we get from Ndour/Kuz, who it seems are the most likely guys to, essentially, fill DWill’s role.

    19. I agree about Boston that who knows how they’ll play, but I think scouts see a team that was good last year and added Horford, so from that point of view they are one of the top four teams inthe East

    20. This whole per 36 debate started because KYK said that he wasn’t impressed with the Euro stats put up by Hernangomez and Kuzminskas. While I disagree with him about their stats (which I think are at least pretty good), I’m more impressed that Hernangomez is starting games for one of the best international teams in the world (a team that has 7 players with NBA experience) at age 22, that Kuzminskas is starting for another one of the best international teams in the world, and that both have played a lot of games against and with current and former NBA players. While that doesn’t guarantee that their games will translate to the NBA, I’m cautiously optimistic that they have the potential to be solid contributors.

    21. ok guys
      “no statistic is infallible” covered me.
      Sorry for spoiling/distracting the intelligent debates.

    22. Sorry for spoiling/distracting the intelligent debates.

      No apologies necessary, homie.

      But if you’re going to contribute to the discourse on a stats-based blog community, maybe at least try to gain an understanding of these metrics and why we (along with NBA front offices, to varying degrees and effect) attempt to utilize them in our analysis . Even if you disagree and put forth a valid argument against one or all of these metrics, that’d be a whole lot more productive then, essentially, saying, “all stats are meaningless unless they are of the antiquated per game counting variety”.

    23. The truth is that i’m an “old skool” bball fan/stat lover.
      I must say that i used to watch A LOT eurobasketball back in the 90’s.
      Those were the days that The truly great players like Drazen Petrovic,Sabonis,Radja,Vrankovic,Komazec,Galis and many many others were putting up “per game” numbers of 20-35pts ,10-15rebs,2-3 blocks and they were playing 30-35 min per game.

      After checking the stats of many euro leagues last night i must say that it looks that bball rotation has changed SIGNIFICANTLY and min per game have decreased for most players.
      Those myrthical stats from the 90’s look Chamberlainesque now…

      Is Europe’s bball declining ? Or are the talented Europeans going to the NBA younger than ever just like the freshmen from NCAA ?

    24. But even funnier is Mike K carping that undrafted players aren’t given a fair shake because of X Y Z when ONE YEAR EARLIER D’Antoni started a second-round rookie.

      And David Duke has a black friend, so he’s not racist.

    25. Saying D’Antoni gets credit for Lin is like calling our tramp a stock market genius. If anything D’Antoni made the situation worse the whole time by playing inferior PGs and racking up L’s.

      25 games is not very many games. Like I said in my post on the previous thread, even in the lockout shortened season it’s only the first third of the season.

      It’s obviously easy to say now that D’Antoni should have started the season with Lin but we are talking about a guy who at the time had put up one season of 9.6 pts/36 on .458 TS% with a WS/48 of .078. That doesn’t exactly scream NBA player. Since you have already conceded that practices are not a good place to evaluate talent in-season, I’m sure you can see the reasoning behind not starting a guy with no track record of NBA success until there are no other options. Further, you continue to ignore that in D’Antoni’s first game playing Lin meaningful minutes he went 0-3 in 7 minutes. He certainly would have been justified in continuing to bench Lin at that point but he didn’t. He should absolutely be credited for sticking with Lin.

      Also, your stock analogy does not work. D’Antoni did not make the “bad financial decisions” that landed him Tony Douglas, Mike Bibby, and Baron Davis to start the season. Here’s a better analogy: someone takes over a bad portfolio of stocks all in free fall, continues to stick with the three worst stocks for a period of time until eventually deciding to sell them off and cut his losses. He then decides to take a chance on a random shitty stock no one has heard of, which immediately rockets up in price.

    26. Here’s a better analogy: someone takes over a bad portfolio of stocks all in free fall, continues to stick with the three worst stocks for a period of time until eventually deciding to sell them off and cut his losses. He then decides to take a chance on a random shitty stock no one has heard of, which immediately rockets up in price.

      Sure I’d agree with that. In that case I don’t think the stock trader is some kind of genius either. Much like I don’t think D’Antoni should get credit for Lin.

      I’d like to add that while Lin’s major league stats were mediocre, he played great in summer league (outplaying Wall IIRC) and had great games in the D-League too (a triple double IIRC). So it wasn’t as much of a random stock. Just one with a puncher’s chance.

      I think were much closer on agreement than we both think. We both don’t think that coaches (I’d add GMs) are very good at spotting unproven talent. I’d add that it’s why statistical analysis is so important, to cover their obvious blind spots.

    27. Evan Turner and Jared Sullinger were big parts of what Boston did last year. It’s possible that Isaiah Thomas can be better than he was last year and that some of their other players will improve, but I doubt Boston is going to improve more than 2 games at most. I honestly think they’ll be worse because they’re likely to replace Turner’s minutes with a guy who is bad at NBA basketball like Jaylen Brown.

    28. I honestly think they’ll be worse because they’re likely to replace Turner’s minutes with a guy who is bad at NBA basketball like Jaylen Brown.

      I don’t think they’re likely to improve all that much but I think you may be underselling the addition of Horford and the impact he’ll have on both ends relative to last season’s bigs rotation (of which Sullinger was a big part).

      Also, I’m not at all high on Brown, especially what level his college numbers suggest he’ll produce at until he learns to put basketballs through baskethoops at a decent rate…But I do think it’s a bit premature to say that he’s bad at NBA basketball until we see him in action.

      Basically…I’m not counting on significant (if any) regression.

    29. Boston finished 4th in DRtg and 10th in ORtg. I like Horford and think he’s a damn good player, but he’s not an elite player. They overpaid for him which is fine, because they’re getting plenty of surplus value elsewhere and still have plenty of cap space and assets that it doesn’t hinder them, but I don’t think he elevates them to another level even if he’ll help them push over 50 regular season wins.

    30. @30 I agree, I think they’ll miss Turner, and they’ll continue to depend on Isaiah Thomas making crazy shots in tight games. Not a threat to unseat Cleveland.

      On the other hand, they do have the smartest coach in the history of the NBA, so I assume he’ll get this group of loveable overacheivers to super-overacheive.

    31. @30

      Yeah I mean I don’t disagree at all. I do think he’s enough of an improvement to mitigate what they’ve lost.

      But I don’t think they’re do for a major leap forward. Barring significant injury to one of their top 6 minutes earners, they’ll likely stay right around last year’s level

    32. Horford is a good player, but it seems Boston’s worst problems on defense were that they fouled a lot and couldn’t get defensive rebounds (5th worst in the league opponent ORB%). And Horford was one of the worst centers in the league in terms of rebounding — one of the major reasons Atlanta couldn’t take the next step. Offensively is he so much better than Olynyk and Sullinger? I guess he passes better.

    33. Yeah I’m not sold on Boston at all. I do think that Atlanta will be better this season, though. It’s TH2’s contract year and he wasn’t bad for them last year, so he should improve. Schröder should improve, too. Dwight Howard, for all his flaws, went back home to be around his support system and was still statistically superior to Al Horford as a rebounder, defender, and in scoring efficiency. They might have the best starting five in the Eastern Conference that doesn’t have LeBron James on the roster with Schröder, whoever starts between Korver/Sefalosha/Hardaway, Bazemore, Millsap, and Howard.

      I think the top three teams in the East are going to be Cleveland, Atlanta, and Toronto with Boston and Detroit 4th and 5th. If the Knicks stay healthy and if MKG comes back balling in Charlotte, the East will be the better basketball conference as I can see a scenario where all of Cleveland, Toronto, Atlanta, New York, Detroit, Boston, and Charlotte win at least 50 next season. Out west, only GSW, LAC, and SAS look like 50 win ball clubs, and all of them should be above 55 barring injuries.

    34. I think were much closer on agreement than we both think. We both don’t think that coaches (I’d add GMs) are very good at spotting unproven talent. I’d add that it’s why statistical analysis is so important, to cover their obvious blind spots.

      Agreed. I just don’t like the idea of saying that luck is the main factor in anything sports related. Obviously luck plays into everything that happens in sport, but there are so many other factors that get left out when a situation is described as “team/coach/player X lucked into situation Y.” On top of that, when you do get lucky you have to be smart enough to capitalize on it (i.e. Kristaps falling to 4 and Phil taking him instead of the many other options on the board).

      To use one of my favorite sayings: the good player is always lucky.

    35. @34 I wouldn’t overlook the Blazers either. They won 44 this year and should only improve as they continue to play together. Evan Turner will help them and maybe another one of their young guys like Vonleh takes another step. I can see them winning 50+ if everything breaks right, especially in a weakened conferenced. I also like Houston to bounce back after a down year. They’re going to give up a TON of points, but Harden running a D’Antoni offense with Gordon, Anderson, and Ariza spacing the floor is something I’d love to see.

    36. They might have the best starting five in the Eastern Conference that doesn’t have LeBron James on the roster with Schröder, whoever starts between Korver/Sefalosha/Hardaway, Bazemore, Millsap, and Howard.

      Totally agree and I’m surprised by how many people, both fans and media, seem to be writing them off.

      I’ve seen a few people say things like “the only EC teams clearly better than the Knicks are the Cavs, Raps, Celtics, and [insert non-Hawks team X]”.

    37. Atlanta is def going to be worse…. replacing teague with schroeder maybe isnt a huge step down but whoever the backup is is def worse than schroeder… bazemore im convinced will turn back into a pumpkin and I doubt anybody else steps up amongst the young wings.. at least this year…

      Boston is def better… horford is a fair bit better than the other bigs they were throwing out there since he is a more complete player… he isnt replacing sullinger directly but more that it allows the other bigs to play more minutes against bench units…

      Pelton had them at mid 50s with the horford signing and I think thats pretty accurate…

    38. Yeah, I figure Boston wins around 53 wins and then with the current core they have they have a chance against anybody in the East until they play Cleveland and get smacked. IMO their improvement because of adding Horford is one that will reap benefits in the regular season more than it will come the playoffs.

      Atlanta is weird. I think Dwight is going to have a big year, but their guard and wing play leaves a lot to be desired. I think they’ll finish in a similar range to the 48 wins they put up last year.

    39. Horford is a very good player, but both Turner and Sullinger did a lot for Boston last year and they haven’t really done much in the way of replacing them. I think Smart will take a step forward, and maybe a big one, so that combined with Horford should lead to them being better, just less so than people think. Ainge did a great job accumulating all their assets, but hasn’t handled them all that well. I mean, Jaylen Brown…yikes.

    40. Just curious to those who don’t “believe in” per 36 minute stats.

      A. How do you determine the best pitcher in MLB?
      B. The best contact hitter?
      C. If the weekday weather man is right 3 days a week, and the weekend guy is right 2 days a week, which one is more accurate?

    41. If the Hornets win 50 games, I will eat my hat. Batum would have to replicate his best season ever, they lost Courntey Lee, Jeremy Lin and Al Jefferson and replaced them with basically nothing, nobody is quite sure what MKG will bring to the table but you have to imagine he takes some of Batum’s minutes. Am I forgetting some great offseason move they made?

    42. Replacing Lin with Sessions was a nice move. I think Charlotte played a bit over their heads last year. Would say they’re closer to 45 than 50.

    43. A. How do you determine the best pitcher in MLB?
      B. The best contact hitter?
      C. If the weekday weather man is right 3 days a week, and the weekend guy is right 2 days a week, which one is more accurate?

      A. WINZZ
      B. Hits per game
      C. Weekday guy, because three is higher than two

    44. Just curious to those who don’t “believe in” per 36 minute stats.

      A. How do you determine the best pitcher in MLB?
      B. The best contact hitter?
      C. If the weekday weather man is right 3 days a week, and the weekend guy is right 2 days a week, which one is more accurate?

      Dear Mike i’m one of the “heretics” per36 non believers but unfortunately i’m not aware of baseball/baseball stats.I’m not from US so i am unable to answer the first 2 Questions.

      As for the 3rd one i’d say that 2/2 is more accurate than 3/5 but i’d like to add here that if the weekend guy was right 2 days a week for only 1 week while the other guy is correct 3/5 every week that doesn’t mean that the weekend guy is The God of Meteorology and the other guy is just…good or average.
      Accuracy w/o consistency is not trustworthy imho.

    45. Let’s extend the dumbest argument in Knickerblogger history into a third day.

      Here are some more controversial topics we can debate:
      Heliocentrism– valid science or hippie bullshit?
      Is a pigeon a bird or a mammal?
      More important musician: Beethoven or Scott Stapp?
      Is red a color?

    46. Good lord this argument will not die.

      KYK – for what it’s worth, you might not be “one of the per 36 heretics” so much as the only one here, really. Not saying you should leave, but be prepared for a long, lonely and pointless battle. If that’s the particular hill you’re picking to die on – best of luck. You do you, champ. Just positing that it might be better to let this specific argument die out rather than continue to argue against the consensus opinion of pretty much everyone else here.

    47. So let’s imagine Rose plays his first game this year scores 30 pts on 35 shots in 38 min, gets injured and is out for the rest of the year. Mindaugas scores 18 ppg in 12 mpg for 70 games. Rose has 30ppg Mindaugas has 18ppg. Did Rose have a better year than Mindaugas?

      Let’s extend the dumbest argument in Knickerblogger history into a third day.

      Here are some more controversial topics we can debate:
      Heliocentrism– valid science or hippie bullshit?
      Is a pigeon a bird or a mammal?
      More important musician: Beethoven or Scott Stapp?
      Is red a color?

      Also, red is clearly a shade, not a color, between grue and bleen.

    48. @45 – Let’s say the weekend guy was right about 85% of the time and the weekday one was right 50% of the time over the course of a year. But of course as you’re aware, the weekday guy had 250 broadcasts, while the weekend guy only did it 100 times.

      You’re starting a rival news station, and you want to steal one of these weatherman from your competition, and the goal is to get the weather right more often than not.

      Which one do you hire?

    49. So let’s imagine Rose plays his first game this year scores 30 pts on 35 shots in 38 min, gets injured and is out for the rest of the year. Mindaugas scores 18 ppg in 12 mpg for 70 games. Rose has 30ppg Mindaugas has 18ppg. Did Rose have a better year than Mindaugas?

      No.
      Rose didn’t have a year at all.

      Let’s now imagine that JFreddette played 5min and scored 7pts which means 50,4pts “per36”
      http://www.basketball-reference.com/teams/NYK/2016.html

      Will you make him a Hall of Famer ?

    50. PER36 is a decent stat to look at, but of course it does not show the whole puzzle. Pair it with MPG to get a feel for the productivity of a player.

      A guy just playing 5 min. per game is probably mostly playing in garbage time vs. scrubs. A chucker like Fredette might just put up good scoring #s in such a case.

    51. @45 – Let’s say the weekend guy was right about 85% of the time and the weekday one was right 50% of the time over the course of a year. But of course as you’re aware, the weekday guy had 250 broadcasts, while the weekend guy only did it 100 times.

      You’re starting a rival news station, and you want to steal one of these weatherman from your competition, and the goal is to get the weather right more often than not.

      Which one do you hire?

      If my goal is to get the weather right more often i’ll try to hire the most accurate one after examining ‘why the most accurate one was only for the other channel’s weekend?…

    52. @55
      Yes, the more minutes per game over time (many games), the more the player is playing against decent opponents in meaningful game situations (not garbage time). Sample size.

    53. If my goal is to get the weather right more often i’ll try to hire the most accurate one after examining ‘why the most accurate one was only for the other channel’s weekend?…

      EXACTLY!!!!

      You’ve chosen the guy that’s likely to give you the better production based on fewer minutes. And you chose him because his RATIO was better in a LOWER AMOUNT OF ATTEMPTS, despite HAVING FEW CHANCES.

      Now it doesn’t tell the whole story, maybe this guy is a alcoholic and can’t handle 5 straight days of work. Per minute stats doesn’t say that one player should get more or less minutes. IT ONLY SAYS IF YOU GAVE HIM MORE MINUTES HE WOULD PRODUCE THE SAME.

      So to circle back, let’s assume that you hire this guy. By per minute stats YOU’D EXPECT HIM TO BE RIGHT 85% OF THE TIME. Now your ratings may go down because he is unattractive, stammers, etc. But he’d be right 85% of the time.

      That’s all per-minute stats are saying.

    54. So, guys who put up good PER/36 numbers in small sample sizes are possibly alcoholics?

      Got it!

      :-)

    55. I do a research on 1840 NBA seasons, ranging from 10-25 MPG in 41+ games played per season. I found that an increase in minutes played led to the same per minute production.

      In other words 5 minutes isn’t enough. But if he kept that production up for 10 minutes per game over 41 games, then yes it would be.

      I believe you.
      It’s logical.It’s not definitive tho.
      Other players are improving their games while others lose it.

      And i definitely believe that Great Teams need role players with accurate “per36” stats.
      And if those “per36” stats are consistent enough they will definitely earn more playing time and have the chance to become starters,allstars and superstars.
      Till Role players become consistent starters allow me to see that happening, instead of predicting it by their “per36” stats.

    56. Let’s now imagine that JFreddette played 5min and scored 7pts which means 50,4pts “per36?

      I want to expand on this.

      Let’s pick a player – Al Coleman. Mr Coleman averages 5.5 ppg and 4.8 rpg, but only 13.3 minutes per game. His team, let’s call them the California Vessels, decides in the 40th game of the year to give him more than double his minutes for the first time in a season (28). What would you expect his points & rebounds to be? About 6 points & 5 rebounds, or much more than that?

      [hint: http://www.basketball-reference.com/boxscores/201601160LAC.html%5D

    57. I believe you.
      It’s logical.It’s not definitive tho.
      Other players are improving their games while others lose it.

      It’s not definitive. You can’t say it’s true for everyone. But it’s true for more players than not, and it’s true in an overall sense. Hence you have to start at the assumption that it is true with a player until it is proven not. To go back to the weatherman, if our weekend guy says it’s going to rain on Sunday, you’d start with that assumption.

      And i definitely believe that Great Teams need role players with accurate “per36? stats.
      And if those “per36? stats are consistent enough they will definitely earn more playing time and have the chance to become starters,allstars and superstars.
      Till Role players become consistent starters allow me to see that happening, instead of predicting it by their “per36? stats.

      I think you’re confusing a few things. Most notably, good per-36 minute stats doesn’t always correspond with winning, because individual stats don’t capture everything (most notably defense). However I’d say if you had to choose between a guy with good per-minute stats and efficiency, and one who doesn’t — you better have a damn good reason to go with the other guy.

      We know that per-36 minute stats and generally efficiency translates well year after year. So it would be dumb not to start the conversation of player worth using these tools.

    58. Let’s pick a player – Al Coleman. Mr Coleman averages 5.5 ppg and 4.8 rpg, but only 13.3 minutes per game. His team, let’s call them the California Vessels, decides in the 40th game of the year to give him more than double his minutes for the first time in a season (28). What would you expect his points & rebounds to be? About 6 points & 5 rebounds, or much more than that?

      Logically more than 6p & 5r w/o excluding the fact that he may has a bad game.

    59. Let’s pick a player – Al Coleman. Mr Coleman averages 5.5 ppg and 4.8 rpg, but only 13.3 minutes per game. His team, let’s call them the California Vessels, decides in the 40th game of the year to give him more than double his minutes for the first time in a season (28). What would you expect his points & rebounds to be? About 6 points & 5 rebounds, or much more than that?

      Logically more than 6p & 5r w/o excluding the fact that he may has a bad game.

      So if Cole Aldrich went 19/10 that day, it wouldn’t be a surprise.

      I mean Al Coleman.

      But you’d expect him to be higher, because in your head you are doing per-minute calculations.

    60. Till Role players become consistent starters allow me to see that happening, instead of predicting it by their “per36? stats.

      The point of per36 stats is that they give you a quantitative idea of how a player will perform if you give him a bigger role, which is important to see before deciding to give a player a bigger role.

      For example, let’s say you have two point guards: One is a short, skinny guy who scores 20pts/36 and 5assists/36 in 5 minutes per game. The other is a lighting quick, physical freak who scores 20pts/36 and 5assists/36 in 10 minutes a game. Assume they each play in all 82 games of a season. Neither guy plays much at all, so they aren’t what you define as “consistent starters” or anything like that, but each puts up a big enough sample size to determine that there wasn’t just a flukey 5 minute stretch where one guy played well.

      According to per game numbers, they look like the same player. According to the eye test, the physical freak looks like the better player. But when you look at their production relative to the actual time they spend on the court, the short, skinny guy is producing the same stats in half the time spent on the floor. Given that neither guy is playing big minutes, you can use the per36 stats to differentiate their identical per game numbers and determine who is the more efficient player, and therefore better, player.

    61. So if Cole Aldrich went 19/10 that day, it wouldn’t be a surprise.
      I mean Al Coleman.
      But you’d expect him to be higher, because in your head you are doing per-minute calculations.

      I’d expect him to be higher w/o excluding to play badly.
      But usually the most possible thing is to put up better numbers.

    62. What would you say about the Knicks 2015-16 per36 stats ?
      Are they useful ?
      If yes, in which way ?

      Per-36 minute stats are most useful for player evaluation, not team. For instance we know Kyle O’Quinn was the team’s best rebounder (11.8 reb/36), Calderon & Grant were nearly equal with assists (5.3 & 5.0 respectively), etc.

      If you want to play around with per-36 numbers, go here and choose some players:
      http://www.basketball-reference.com/play-index/pcm_finder.cgi

      To answer your question, for team evaluation, you would need the four factors.

      The four factors are shooting (eFG%), rebound percentage (reb%), turnovers (to%), and free throw rate (ftm/fga). Of course there are both offensive and defensive components, so really there are 8 factors. Also some factors are more important than others. shooting is by far the most important, and free throw rate the least.

      Go to the b-r page you linked, and look under Team Miscellaneous. You’ll see the Knicks on offense were ranked 28th in shooting (eFG%), 11th in turnovers, 18th in rebounding, 16th in free throws.

      On defense, they were ranked 9th in opponents shooting, 30th in turnovers, 18th in rebounding, and 13 in free throws. If you look at the top of the page, you’ll see

      Off Rtg: 104.6 (24th of 30) ? Def Rtg: 107.6 (18th of 30)

      That means the Knicks offense was 24th last year, and the defense was 18th. If you look at the four factors, this seems about right. We were awful with shooting and forcing turnovers, while pretty good at stopping teams from scoring and average at just about everything else.

    63. You’re starting a rival news station, and you want to steal one of these weatherman from your competition, and the goal is to get the weather right more often than not.

      Which one do you hire?

      Trick question, the answer is always Cole Aldrich.

    64. Per-36 minute stats are most useful for player evaluation, not team. For instance we know Kyle O’Quinn was the team’s best rebounder (11.8 reb/36), Calderon & Grant were nearly equal with assists (5.3 & 5.0 respectively), etc.

      Ok Mike.Thanks for explaining me those things.
      Personally i see as team’s best rebounders Melo,KP and Rolo and at assists Melo and Calderon but i can understand the other way to view the stats also.

      I think i consider ‘min per game’ as something significant earned by a player with his consistency/durability/trust by his coach that’s why i prefer to trust “stats per game” more than “per36” stats which i see as something like “what if”.

    65. Mike K – have been thinking a bit about your per-min PER study, not totally convinced by its methodology and your choice of control.

      The major issues with per-minute stats being extrapolated from “role-player minutes” to “starter minutes”, like you said in your post, is that generally speaking, only players who are expected by their coach to remain or be increasingly productive would be given more minutes to play.

      Your control group, as far as I can tell, is just a random group of players who played 10-25 minutes/game. Your test group is not necessarily related to the control group — ie. they’re not the same players with an increase in minutes like in what’s called a case-crossover design where the player is his own control. In fact, they’re specifically NOT the same players since you chose the 2 groups specifically to not overlap. So I’m not sure that the “control” group in your study is appropriate. Please let me know if I’m thinking about this wrong.

      In addition, I’m not sure about your choice of “>+5 min/game from the year before”. For someone playing 10 minutes/game to go to 30 min/game is really a 300% increase in minutes, whereas for a player playing 10 minutes –> 15 min is far less a difference, yet both would fall within your criteria. The amount of effort/fitness involved in those two situations is, of course, completely different. By your criteria, the separation between the 2 groups is probably minimal, so any differences (or lack thereof) are less meaningful.

      Overall it’s a very difficult thing to prove or disprove. You could even take single players and compare their stats in games with >20 minutes played and <15 min played — but that in itself is probably biased in that if the player is playing poorly they might get pulled quickly, and if they're playing well, they keep playing.

    66. Even if your study does have the right conclusion (there’s no correlation between minutes played and PER), it doesn’t mean that a player’s PER may not decrease given more minutes. it just means that roughly the same amount of players had their PER go up as go down (or stay the same), and so minutes in and of themselves are not the determining factor in PER.

      it doesn’t prevent or disprove the statement, for example, that Cole Aldrich is completely out of shape and I think he’ll have a heart attack if he has to play more than 17 minutes in a basketball game, and as such, I can’t justify paying him anything more than what’s appropriate for a 3rd big man.

    Comments are closed.