Statistical Analysis. Humor. Knicks.

Tuesday, October 21, 2014

Is Gallinari Inconsistent?

It’s been asserted by Knick fans in the past, most recently by NYK Ewing, that Danilo Gallinari is inconsistent. This opinion of him, like many by NBA pundits, are made from an observational standpoint. However, observations are prone to cognitive bias which may be false. Hence it’s always good to have an objective method to test such a claim.

So the question we should ask is a more open-minded “is Gallo inconsistent?” How can we verify such a statement? Well in this instance I think we can measure consistency by using standard deviation from game-by-game performance. To illustrate what this entails, I’ll take points scored per game for two fictional players:

Player A: 5,5,5,5,5,5,5,5
Player B: 2,4,4,4,5,5,7,9

Both players average 5 points per game, but Player A has a standard deviation of 0, while Player B has a standard deviation of 2. Therefore the player with the lower standard deviation is more consistent. Since we don’t have a baseline on standard deviation of points scored, we need to create one. However, for basketball players points scored per game is dependent on how many minutes per game they are given, their skill level, and their method of scoring, among other things. Hence we should find players similar in these respects to Gallinari to minimize the effect of these factors. To do this, I used this query from basketball-reference to find players who are close to Gallo in minutes averaged, points scored, and three pointers attempted. The five players that returned were: Wilson Chandler, Joe Johnson, Gilbert Arenas, Ray Allen, and Dorell Wright.

The results?

Name Pts StdDev Less than 10 pts 3p% StdDev GmScr StdDev
Gallo 8.04 6 0.263 8.12
Chandler 5.65 1 0.243 6.24
Johnson 7.07 2 0.199 5.7
Arenas 8.00 4 0.202 7.36
Allen 6.49 3 0.237 5.39
Wright 7.06 4 0.228 6.92

Since Gallinari has the highest standard deviation of points, three pointers made, and game score it means that he has been inconsistent when compared to similar players. Gilbert Arenas’ points fluctuates nearly as much, however he’s more stable with regards to three pointers and game score. On the bright side, the Rooster’s instability also means that on certain days he’s better than some of those other players. For instance he’s had 5 games this year where he’s made 9 or more free throws, something teammate Wilson Chandler has yet to do. Conversely, it means that something is preventing Gallo from having more great nights, or from contributing on the stat sheet when he’s having a bad day. Unfortunately whatever that is, it will restrict him from reaching his full potential unless he can overcome it.

151 comments on “Is Gallinari Inconsistent?

  1. David Crockett

    This is one of those instances where the relevant stat backs up intuition and observation.

    IMHO, the “whatever that is” that leads to Gallo’s inconsistency is his weakside movement. As has been debated here to death, Gallo camps out on the weakside and doesn’t move. The debate is about why he does that. Regardless of why, he can make it easy for defenses to deny him the ball (but at least he takes a defender with him).

    The difference between he and Landry Fields in this regard is striking. Gallo is eons better as a shooter, but Fields moves so much better on the weakside. Fields is a better offensive rebounder and passer based almost solely on his activity. (Now, if that kid ever develops any range on his shot…)

    Gallo is not a poor rebounder or passer either. He lacks a feel for how to move against screen roll defenses. I also think he’ll get better at it. Like Fields, this is Gallo’s age 22 season.

    One area that may help Gallo with his consistency is NY’s newfound running game. Gallo has a *very* good feel for running on the break–whether to fill the lane, fade to the corner, and when to pass.

  2. TDM

    The games where Gallo was his most effective, IMO, were those where he was actively trying to draw fouls and get to the line where he shoots at an incredible, Houston-like rate. I enjoyed watching him play in thosegames because it showed he understood what his strengths are. Often, unfortunately, Gallo will play a game and only get to the stripe 5 or less times and instead try to survive behind the arc. He is still young so I am hopeful, but I would have thought he would have been more consistent by now. When he is on his game however, he is one of my favorite players to watch in the NBA.

  3. Nick C.

    It’s nice when observation is backed up by data. I think one issue is that FTs are a bit like BB in baseball. You don’t really notice them quite as much as FG or hits but they count just the same. So last night I was surprised when they said Gallo had 20 because it seemed all he was doing was rimming out which a check of the box confirms 5-13 FG (2 3pters) but 8-9 FT.

  4. d-mar

    I know this is a Gallo thread, but just read this in the NY Times and it makes me feel even warmer and fuzzier about Stat:

    “He is proving to be the franchise player the Knicks anticipated when they signed him to a five-year, $99.7 million deal. He has assumed leadership of a previously rudderless franchise. While he stretched and watched film of Sunday’s win over Toronto before the game, Stoudemire beckoned the reserve center Timofey Mozgov to join him. At one point, Stoudemire rewound the tape and tutored Mozgov on catching the basketball and finishing in traffic.”

    Now that’s leadership!

  5. Frank

    David Crockett: This is one of those instances where the relevant stat backs up intuition and observation.IMHO, the “whatever that is” that leads to Gallo’s inconsistency is his weakside movement. As has been debated here to death, Gallo camps out on the weakside and doesn’t move. The debate is about why he does that. Regardless of why, he can make it easy for defenses to deny him the ball (but at least he takes a defender with him).
    The difference between he and Landry Fields in this regard is striking. Gallo is eons better as a shooter, but Fields moves so much better on the weakside. Fields is a better offensive rebounder and passer based almost solely on his activity. (Now, if that kid ever develops any range on his shot…)Gallo is not a poor rebounder or passer either. He lacks a feel for how to move against screen roll defenses. I also think he’ll get better at it. Like Fields, this is Gallo’s age 22 season.

    Alan Hahn had an interesting comment about this – can’t remember if it was on twitter or on his blog — but he basically said that Gallo’s role on many of the plays is to be the outlet 3 point guy — meaning that he’s not supposed to move — he’s supposed to be where he’s supposed to be, so that someone dribbling into trouble can know with a high degree of confidence that he will be standing at that specific spot. Landry is NOT that guy, and so is free to DWLFD. Unfortunately when Gallo DWDGD it looks a lot like standing around.

    What Gallo does need to do is knock down those wide open 3’s. I hope this is just some sort of shooting “slump” but he is not the knock-down shooter he was last year. I know his percentage is improving but it sure looks like he is missing a lot of shots that guys like Ray Allen, Kapono, Korver, etc. don’t miss.

  6. Bruno Almeida

    @4

    yeah, that’s definitely the kind of attitude we expected, Amare has been nothing short of amazing, that’s a fact.

    I think Gallinari will get better, he’s got a lot of talent and he’s still very young, but I agree he needs to get more active on the offensive end.

    he’s such a great free throw shooter, but the problem I see is that everytime he drives he’s looking for the foul… so sometimes he forces things and puts up a totally out of control shot, which most of the times doesn’t come closer to an and one, and he only gets the foul if the referee bails him out.

    a little off-topic here, Ken Berger of CBS posted today a piece on Melo, saying everything we’ve read for a while now, but he said that “rival execs agree that New York would be a logical landing spot for Andre Iguodala, and they believe the Sixers will be more than open to discussing trades for the dynamic but high-priced swingman as the Feb. 24 deadline approaches.”

    I don’t know, his salary is sure terrible and the Sixers would ask for considerable compensation, but man, he’d instantly become the best defender on this team and he could really flourish on offense with Amare commanding most of the attention.

  7. david

    Does anyone know why we never seem to have Gallo run cuts or even run off off-the-ball screen while the pick and roll is happening on the other side of the court?f Phoenix ran this play last season that lead to an open three for Jason Richardson over and over again which involved a very high pick-and-roll at the top, grant hill flashing to the free throw line (he was on the block) and richardson moving from the corner to the elbow three. Richardson’s defender would always shade over, meaning when he made his cut he was a couple of steps behind. (Not sure I explained this perfectly but here is NBA playbook on it http://nbaplaybook.com/2010/05/04/the-suns-get-jason-richardson-open-for-the-dagger/).

  8. Abasi

    with all due respect to Mike Kurylo and all other guys who know and understand stats WAY beyond what i will ever know or understand, I’ll say this:

    You don’t need to compare Gallow to anyoone ir order to determine whether he is consistent or not. you only need to compare his production game to game. comparing him to other players determines is he is consistent WITH the average production of similar players. The ned result is that he may very well be CONSISTENT with those players, but still be INCONSISTENT in his production for the Knicks.

    to me, consistency is having output within a certain range. Inconsistency is wild changes in output: 2 points today, 30 tomorrow, one week of good production, three weeks of sucking follwed by one week pof average play and 2 weeks of great play. you can be inconsistent in your daily production: David Lee looked terrific at the beginning of every game- he couldn’t miss a jumper. But somehow he couldn’t be equally terrific in every fourth quarter (not a knock on Lee- it may be that by then the game was usually lost so his effort was not noticed or meaningful) .

    I personally think Gallo is VERY inconsistent, because he mostly plays well below what we expect of him, but he also maintains deccent production numbers, which means he is playing very well for short spurts.

  9. Mike Kurylo Post author

    Abasi: You don’t need to compare Gallow to anyoone ir order to determine whether he is consistent or not. you only need to compare his production game to game. comparing him to other players determines is he is consistent WITH the average production of similar players. The ned result is that he may very well be CONSISTENT with those players, but still be INCONSISTENT in his production for the Knicks.

    This makes zero sense. Of course you need something to compare it to, else the comparison is invalid.

    For instance take how many miles per gallon a car gets. Person A says he gets 30 mpg, while Person B gets 20 mpg. Does that mean that Person A’s car is less of a gas guzzler. The answer is … you don’t know! Because you haven’t accounted for the other variables in play. Are they driving in straight highways, curvy mountainous roads, or city traffic? Does the driver jack-rabbit? Are they carrying extra cargo?

    If you want to compare two things you have to eliminate the other variables for a fair comparison. Heck this isn’t even complicated science, it’s plain old common sense.

  10. Jafa

    d-mar: I know this is a Gallo thread, but just read this in the NY Times and it makes me feel even warmer and fuzzier about Stat:“He is proving to be the franchise player the Knicks anticipated when they signed him to a five-year, $99.7 million deal. He has assumed leadership of a previously rudderless franchise. While he stretched and watched film of Sunday’s win over Toronto before the game, Stoudemire beckoned the reserve center Timofey Mozgov to join him. At one point, Stoudemire rewound the tape and tutored Mozgov on catching the basketball and finishing in traffic.”Now that’s leadership!  

    Hey d-mar, can you link the article so the rest of us can enjoy reading it?

  11. NYK Ewing

    Good write-up Mike.

    My analysis might have been a bit biased in terms of defining consistency. When we got him initially, he excelled at hitting open 3’s, but this year he’s definitely not a sure thing from downtown (although fortunately not in big moments – ie: TWolves game, OT with Pistons). However, as Mike pointed out, he has still found other ways to score (most notably off getting fouled, which is doubly helpful because it gets them closer to the limit). He’s gotten better at this area of the game, although I contend that he hasn’t looked convincing doing it. That’s not something I can easily prove with stats – maybe the closest thing I can get is number of and-ones per foul – but just looking at his game shows that he consistently doesn’t go up strong enough to the basket. So he’s been putting up points more consistently than I admitted before, but he’s been inconsistent in terms of the game I’m expecting him to play (and hence my bias).

    I should add the caveat that I was more impressed with him last night. He went for boards aggressively (one off from a d/d I believe), and he drove much harder to the basket. He still missed open threes, but let’s see if he can continue that type of play.

  12. Mike Kurylo Post author

    Mike Kurylo: Abasi: You don’t need to compare Gallow to anyoone ir order to determine whether he is consistent or not. you only need to compare his production game to game.

    Actually a better analogy to your situation would be to take the car and measure the miles per gallon. You’d find that some days your car is awesome while other days it’s pretty bad. If you don’t account for traffic, highways, streets, cargo, etc. then you can’t conclude much about your car’s mileage, since it’s dependent on these things.

  13. Mike Kurylo Post author

    Ok last comment on this because it’s really bugging me: You expect most players to do better against bad defense and worse against good defensive teams. This is pretty much by definition since better defensive teams are measured by their opponents’ stats. Hence looking at a single player won’t tell you much until you put it in context.

  14. NYK Ewing

    @ Jafa

    I think you have to pay for the NYT online.

    But here’s one of my favorite Amare quotes so far:

    “Raymond Felton’s shot bounced on the rim so long that he had time for a prayer.

    Not that it was needed. Had it missed, Amare Stoudemire said he would have put the rebound in.”

    (
    http://scores.espn.go.com/nba/recap?gameId=301208018 )

    The guy’s playing hungry, and is living up to and surpassing our expectations of him so far.

  15. Peter87

    I think the analysis is ok, as far as it goes. However, “consistency” meaning “consistency in points scored per game” is only one possible measure. If a player decides to score less in a game and go more for boards and defense because that’s what’s needed in a given game, do we really want to label that “inconsistent”? I know there’s no one number which adequately measures a player’s contribution in a game, but it would be interesting to try; either by using “+/-“, or by creating some other number which includes points scored, boards, steals, blocks, assists, to’s, etc. (probably most fantasy leagues have some sort of overall measure of this kind). I also think it would be better to use “per minute” or “per 36″ numbers.

  16. Abasi

    Mike Kurylo: Actually a better analogy to your situation would be to take the car and measure the miles per gallon. You’d find that some days your car is awesome while other days it’s pretty bad. If you don’t account for traffic, highways, streets, cargo, etc. then you can’t conclude much about your car’s mileage, since it’s dependent on these things.  (Quote)

    Thios is a better example than the first one – but it suffers from the same flaw in your reasoning: if you want to determine how consistent the car is, you don’t measure its porduction agiant the production of other cars. You measure its production against the SAME conditions every day. If you measure the car’s prosduction against other cars under the same conditions, you’re measuring something. If you eliminate other cars and only measure the day-to-day output (while keeping track of changes in conditions, etc.) then you can determine how consistent that output is. It should be noted, though, that if the conditions (traffic, weather, etc.) change dramtically, consistency can NOT be determined.

    That was my point.

  17. tastycakes

    The NYTimes is free, here’s the article being discussed:

    http://www.nytimes.com/2010/12/09/sports/basketball/09knicks.html?_r=1&ref=sports

    On the Gallo topic: when I watch the man play, I actually think he looks pretty good when he drives, and so there is this frustration because it seems like he will be more effective as a mid-range player, guy who attacks the hoop, unless he can get that 3P% up over 40%. But as observed previously in this thread, he may just be playing his role, and the Knicks are winning, so I don’t see the big problem.

    There is this expectation that he’s supposed to be our #2 guy or something, and he’s clearly not that so far, but he is just a young pup, and considering he and Fields are the same age, well, I’d like to see those 2 guys on the Knicks 3 years from now, based on what they’re doing at 22.

  18. Abasi

    to make my point much easier to understand: if Gallo produces 31 points with 10 rebounds today and goes for 6 points and 0 rebounds tomorrow, folloed by 24 points and 14 rebounds the next day and 2 points with 18 rebounds the following night, he is NOT consistent. You don’t really need stats or other players to come to that conclusion, do you?

  19. villainx

    “just read this in the NY Times and it makes me feel even warmer and fuzzier about Stat:”

    Not to go I told you so, but that’s partly why I think Stoudemire’s addition was a big part of turning the Knicks into a more aggressive shot blocking team.

    I have one request, please have more game wrap up threads. I don’t get to watch a lot of games (hardly any), and the post game posts and comments is where and how I most get a sense of what’s the games been like.

  20. Abasi

    Mike Kurylo: Actually a better analogy to your situation would be to take the car and measure the miles per gallon. You’d find that some days your car is awesome while other days it’s pretty bad. If you don’t account for traffic, highways, streets, cargo, etc. then you can’t conclude much about your car’s mileage, since it’s dependent on these things.  (Quote)

    And now you’re making my case: you don’t need to know about other cars on that road: you excluded them, the same way I’m asking you to exclude the other ‘cars’ (players) in the Gallo analysis. you only need to know the “weather, cargo, etc”, not what other cars are doing. I’m glad you posted this example, because it better explains what I have been saying.

  21. njasdjdh

    [20] Yes, you do, because without knowing whether every player of a similar profile plays that way we have no idea what is and is not consistent.

  22. Nick C.

    As I understand it consistency would be by the game to game fluctuations (standanr deviation?) but you compare him to others to get an idea of whether or not this is normal.

  23. TDM

    Off-topic alert: someone threw out Bassy’s name yesterday (I think) as a potential target for the Knicks. Now, Isola is reporting that the Knicks are interested in him.

    http://www.nydailynews.com/sports/basketball/knicks/2010/12/07/2010-12-07_stoudemire_honored.html

    Never been a big fan. He always seemed like more hype than substance. Is he really the best the Knicks can get?

    http://www.basketball-reference.com/play-index/pcm_finder.cgi?request=1&sum=1&p1=telfase01&y1=2011&p2=walkebi01&y2=2011

    Still say Antonio Daniels would be a better fit with this team.

  24. Abasi

    Nick C.: As I understand it consistency would be by the game to game fluctuations (standanr deviation?) but you compare him to others to get an idea of whether or not this is normal.  (Quote)

    That would tell you if his performance is consistent with OTHER players. It may be that they are all INCONSISTENT in their production, but CONSISTENT with the production of each other. Two VERY DIFFERENT things.

  25. Abasi

    njasdjdh: That would tell you if his performance is consistent with OTHER players. It may be that they are all INCONSISTENT in their production, but CONSISTENT with the production of each other. Two VERY DIFFERENT things. Abasi

    That would tell you if his performance is consistent with OTHER players. It may be that they are all INCONSISTENT in their production, but CONSISTENT with the production of each other. Two VERY DIFFERENT things. Abasi

  26. Abasi

    njasdjdh: [20] Yes, you do, because without knowing whether every player of a similar profile plays that way we have no idea what is and is not consistent.  (Quote)

    Sorry for the mess – the quotes are killing me :-)

    That would tell you if his performance is consistent with OTHER players’s nightly production. It may be that they are all INCONSISTENT in their production, but CONSISTENT with the production of each other. Two VERY DIFFERENT things.

  27. Mike Kurylo Post author

    Abasi: to make my point much easier to understand: if Gallo produces 31 points with 10 rebounds today and goes for 6 points and 0 rebounds tomorrow, folloed by 24 points and 14 rebounds the next day and 2 points with 18 rebounds the following night, he is NOT consistent. You don’t really need stats or other players to come to that conclusion, do you?  

    That is an extreme example. Gallo hasn’t had a game with 14 or 18 rebounds. Actually he’s only had 1 game with 10 or more rebounds.

    OK take these 3 consecutive games, and answer whether this player is consistent:

    Game 1: 20 pts, 4 reb
    Game 2: 24 pts, 13 reb
    Game 3: 37 pts, 15 reb

    That’s Amar’e Stoudemire. He’s also had 3 games with 4 rebounds and 4 games with 14 or more rebounds. He’s had 2 games with under 14 points, and 3 games with 35 or more points.

    There’s a reason we use statistics and logic to deduce such outcomes, and not just slap a label on someone because that’s we deem to be true by our own made-up standards.

  28. njasdjdh

    Abasi:
    That would tell you if his performance is consistent with OTHER players. It may be that they are all INCONSISTENT in their production, but CONSISTENT with the production of each other. Two VERY DIFFERENT things.Abasi  

    But if everyone is inconsistent…doesn’t that make them all consistent?

  29. cgreene

    @22 I think what you are missing in the comparison to other players is that if they too showed similar deviations as Gallo that the type of inconsistency would be the norm and, therefore actually, not inconsistent. By comparing him to other players with similar stats/styles it shows that his deviations are wider and he then is more inconsistent than them and so the characterization of him as inconsistent is correct because his deviations are not the norm.

  30. cgreene

    guess i wasn’t the first w that thought. got sidetracked while writing and everyone said the same thing in between (-;

  31. Abasi

    Mike Kurylo: That is an extreme example. Gallo hasn’t had a game with 14 or 18 rebounds. Actually he’s only had 1 game with 10 or more rebounds. OK take these 3 consecutive games, and answer whether this player is consistent:Game 1: 20 pts, 4 rebGame 2: 24 pts, 13 rebGame 3: 37 pts, 15 rebThat’s Amar’e Stoudemire. He’s also had 3 games with 4 rebounds and 4 games with 14 or more rebounds. He’s had 2 games with under 14 points, and 3 games with 35 or more points. There’s a reason we use statistics and logic to deduce such outcomes, and not just slap a label on someone because that’s we deem to be true by our own made-up standards.  (Quote)

    you’re very conveniently presenting a very reduced sample – we can’t tell much from that. Extend your example to the entire season and then we will have a better idea of how consistent the player is. Three games doesn’t tell you much.

    That said, perhaps your mistake is in not clearly defining what type of “consistency” you’re alluding to: consistency as in “producing more or less the same every night” or consistency as in “showing more or less the same PRODUCTION PATTERNS of other players playing the same position”.

    Those are two very different things.

  32. Mike Kurylo Post author

    Abasi:
    And now you’re making my case: you don’t need to know about other cars on that road: you excluded them, the same way I’m asking you to exclude the other ‘cars’ (players) in the Gallo analysis. you only need to know the “weather, cargo, etc”, not what other cars are doing. I’m glad you posted this example, because it better explains what I have been saying.  

    You’re missing the point. Even a singular event can’t be compared on it’s own merits because it’s missing context. Here’s a set of numbers, and you tell me if this is consistent:

    3-1-8-1-2-1-3-5

    Another set of numbers:

    12-7-7-17-11-10-6-6-4-18

    Are these consistent, yes? or no? Can’t tell? How about if I tell you the first is Dwight Howard’s blocks by game, the second is Duncan’s rebounds by game. Still can’t tell? Neither can I. Do blocks and rebounds vary much per game? I don’t know. How would I know?

    Take a standard deviation and compare it to other similar players. That’s the only way.

  33. Frank

    njasdjdh:
    But if everyone is inconsistent…doesn’t that make them all consistent?  

    This is sounding more and more like the semantic argument that Ted and I had about Anthony Randolph’s shot selection — you two are pretty much disagreeing over little details. In summary, Gallo is obviously inconsistent from day to day. Whether it has to do with the different opponents’ defenses, real inconsistency on Gallo’s part, etc. is not totally clear to me, but IMHO it is logical that a guy that shoots so many 3 pointers will always be more inconsistent than a guy that shoots all dunks. While Abasi is right in saying that you don’t need player-player comparisons to see that his point production is all over the board, it IS useful to see whether he has the normal amount of “inconsistency” in his #s you would expect in a guy that plays the offensive role that he does (ie. shoot 3’s). I think is what Mike is trying to say.

    This kind of analysis should be very useful to Gallo himself if he reads it — he should see that it’s not just “my shot was a little off”, which happens to all players and may generally even out over time — it’s that his shot is either ALL ON or ALL OFF much more than it should be, or at least more than happens to other players. This might point to a mechanical issue that crops up from game to game that gets corrected in practice or on film study etc. as opposed to expected game-to-game variance on a relatively low-probability shot.

  34. villainx

    njasdjdh, didn’t notice that you post here too until today.

    Abasi, maybe I’m not reading you right, and heck, I could also be misreading Mike Kurylo, but if you just look at the consistency or inconsistency of Gallo, without referring to other players, you wouldn’t be able to say if Gallo’s relative (in)consistency is typical or extreme. No?

  35. Mike Kurylo Post author

    Abasi: That said, perhaps your mistake is in not clearly defining what type of “consistency” you’re alluding to: consistency as in “producing more or less the same every night” or consistency as in “showing more or less the same PRODUCTION PATTERNS of other players playing the same position”.

    Those are two very different things.

    This makes no sense either: “consistency as in “producing more or less the same every night” ” which you’ve said in your first post “You don’t need to compare Gallow to anyoone ir order to determine whether he is consistent or not. you only need to compare his production game to game. ”

    What NBA player is consistent more or less on every night? By this definition I can argue that all players are inconsistent based on what ever I deem to be “more or less”. Why would I measure such a thing?

    BTW no need to capitalize random words. I’m perfectly capable of understanding the meaning of words when they are in lower case.

  36. Mike Kurylo Post author

    Frank: While Abasi is right in saying that you don’t need player-player comparisons to see that his point production is all over the board,

    No he’s not right. You can look at a lot of players and see point production over the board. Go ahead spend a day finding guys you think are solid as a rock, and look at their game logs. You’ll be surprised. Even the most stable of the players above, Wilson Chandler, has a standard deviation of 5.65 points. That’s a lot.

    Unless your human mind has the amazing ability to calculate standard deviation in your head of every game of every player in the league and innately understand exactly what is expected of a player of every type (bangers, rebounders, slashers, shot blockers, etc. etc.) then you need statistics to make that claim. Otherwise it’s just some guy’s opinion. And everyone has an opinion.

  37. JK47

    Okay, capologists. Is there any feasible way to make this roster a reality for next season:

    PG Ray Felton
    SG Landry Fields
    SF Carmelo Anthony
    PF Amar’e Stoudemire
    C Marcus Camby

    Bench:
    Danilo Gallinari
    Timofey Mozgov
    2011 1st round pick
    Shawne Williams or whoever else

    That would be a pretty fierce SSOL team. This would assume we’re subtracting Eddy Curry’s carcass, Wilson Chandler, Anthony Randolph and Toney Douglas.

    But I’m not gonna lie– I don’t fully understand the salary cap and all the contracts involved. I think Camby is owed $10M next season and Melo of course would want a max contract… or would he? Could we perhaps convince Melo to take a salary “haircut” like his d-bag friends in Miami?

    The beauty of it would be that Camby’s contract is coming off the books after 2012– it’s not like it’s a brutally long contract. In terms of pure basketball, Camby is such a perfect fit for this squad it’s ridiculous. His scoring efficiency has been terrible but his usage is so low that it almost doesn’t matter. But he still doesn’t kill you on the offensive end, because he’s a really good offensive rebounder– 7th best in the NBA this year. On defense, he’s a beast. Fearsome shot blocker, ranks #2 in defensive rebounding percentage behind Kevin Love, has a defensive rating of 101.

  38. Abasi

    @37

    “You’re missing the point. Even a singular event can’t be compared on it’s own merits because it’s missing context. Here’s a set of numbers, and you tell me if this is consistent:

    3-1-8-1-2-1-3-5″

    No, they are not consistent, if you define “consistent” as “close to the “average”. The average for that sequence is 24/8 = 3. So the problems becomes: how CLOSE does it have to be to be considered “consistent”? That’s a very subjective value. For example: you could define it as, say, no more than 1 over or under the average: 4-2-4-2-3-3-2-4. This is far more “consistent” than 3-1-8-1-2-1-3-5. So if the 4-2-4… sequence is howard’s production for April and the sequence 4-2-4… is his production for May, we can esily conclude his night-to-night production was more consistent in May. And he was “very inconsistent” in June when when he gave us the same 24 rebounds for the month, but in this pattern: 0-8-1-0-2-9-0-4. No other players need to be involved to reach these sensible conclusions.

    As you can see, the problem is determining *exactly* what type of consistency we are trying to measure. After all, in the above example, Howard was VERY CONSISTENT in giving us 24 rebounds per month :-)

    In the case of Gallinary, he is inconsistent on a game-to-game basis and that’s what I thought this thread was about.

  39. mred696

    It seems to me that the debate is centered around the definition of “inconsistent” and perhaps the wording of the question, “Is Gallo inconsistent?”

    Inconsistency, as Abasi describes it, is the game-to-game fluctuation in stats. To the extent that any player’s production will vary each game, that player is “inconsistent”. This would be analogous to saying that my car’s MPG changes each day. A better analysis of absolute inconsistency in this regard might be called “adjusted inconsistency” in which you adjusted for factors such as opponent’s defense. The analogy would then be to say my car’s MPG changes each day even after accounting for rain, miles driven in city vs highway, etc.

    What Kurylo describes in his answer to “Is Gallo inconsistent?” is not absolute inconsistency (Yes, to the extent that Gallo’s stats are not identical every day, he is in fact inconsistent) but rather whether this absolute inconsistency is an acceptable reflection of Gallo’s situation. It might be that all players who fill Gallo’s role are absolutely inconsistent, as Abasi points out. Abasi is also right in saying that Kurylo’s analysis is measuring players’ consistency with each other, but “consistency” in this sense is not game-to-game consistency as used above, but consistency in terms of similarity.

    Thus, there are two applications of “consistency” being used. First, game-to-game absolute consistency within a player. Second, consistency between players. The first is what we’re ultimately interested in. The second is the measuring stick we’re using to figure out if the first is at an acceptable level.

  40. Abasi

    @44 mred696

    May god bless you for expressing my point far better than i could in my 275 clumsy posts!

    (with apologies to everyone – i don’t want to be argumentative or hijack the thread – i’m just trying to explain my position. sorry!)

  41. Thomas B.

    Great discuss here at calculusblogger.net. I am very curious as to where Gallinari is inconsistent. As you all know, I am far to dumb to run a what do you call it, a standard deviation? But looking at Gallinari’s boxscores in which he has played at least 30 minutes and scored at least 20 points (8) games his eFG% is .584 8.75 FTM made. 106 FGA in these games. Knicks are 6-2 in those games (.750).
    In games that he played 30 but didn’t score 20 (11 games) his eFG% is .481 with 4.72 FTM. 106 FGA in these games. Knicks are 7-4 in those games (.636).

    Maybe it just as simple as this: When Gallinari shoots well from the floor and gets to the line, he scores more points than when he shoots poorly from the floor and does not get to the line. The team is above .500 whether Gallinari shoots well or poorly, but the team fares much better when he shoots well.

    I have no idea what any of this means. One of you smart people can try to figure it out.

  42. Frank

    Mike Kurylo:
    No he’s not right. You can look at a lot of players and see point production over the board. Go ahead spend a day finding guys you think are solid as a rock, and look at their game logs. You’ll be surprised. Even the most stable of the players above, Wilson Chandler, has a standard deviation of 5.65 points. That’s a lot.
    Unless your human mind has the amazing ability to calculate standard deviation in your head of every game of every player in the league and innately understand exactly what is expected of a player of every type (bangers, rebounders, slashers, shot blockers, etc. etc.) then you need statistics to make that claim. Otherwise it’s just some guy’s opinion. And everyone has an opinion.  

    What #44 said. Nicely done.
    Meanwhile Mike, I was actually making your point. Badly apparently. Obviously Gallo has 30 point games and 8 point games and everything in between. As far as I can tell, your question is whether he is more inconsistent than expected.

    I think this exercise is really useful — not just in points, FTA, etc. but in rebounding, deflections, etc. I assume teams probably keep track of this stuff because it uncovers deviations outside the “expected” that may reveal deficiencies in consistent effort, mechanics, etc.

    I’d love to see “hustle” stats (ie. ORebs, steals, deflections, etc.) and their standard deviations within games (ie. 1st quarter vs. 2nd quarter vs. 4th etc.) also. My guess just from watching games is that many players loaf through most parts of most games, then turn on maximal effort in crunch time. Would be interesting to see whether guys like Fields, Noah, etc. who are known as “effort/hustle players” have far smaller standard deviations than others.

  43. JK47

    Does Gallo seem WAY more inconsistent than Wilson Chandler? Not really. Maybe slightly, but not by giant leaps.

    If you’re good enough to play in the NBA, you’re pretty much “consistent” to some degree. If you stink it up in too many games, you’re not going to keep getting minutes, unless Isiah Thomas is running your team.

  44. Mike Kurylo Post author

    mred696: Inconsistency, as Abasi describes it, is the game-to-game fluctuation in stats. To the extent that any player’s production will vary each game, that player is “inconsistent”. This would be analogous to saying that my car’s MPG changes each day. A better analysis of absolute inconsistency in this regard might be called “adjusted inconsistency” in which you adjusted for factors such as opponent’s defense. The analogy would then be to say my car’s MPG changes each day even after accounting for rain, miles driven in city vs highway, etc.

    OK so let’s say I find that Gallo’s “adjusted inconsistency” is X. That still doesn’t answer whether or not he’s inconsistent. Abasi is saying, repeatedly, that you don’t need another player to see if something is consistent. That’s not true, and there’s an easy way to show that. What value of X is needed for Gallo to be consistent or inconsistent?

    You can’t answer without looking at other players.

    Let’s take my post above. Let’s keep Gallo’s value at 8.04 points. What if I found that Chandler, Wright, Allen, Gilbert, etc. all had higher stdev? (9,10,12, etc.) Would Gallo still be thought of as inconsistent?

  45. drm1977

    Nice analysis.

    Gallo hasn’t played in an offense with this many options before, so it feels like he’s learning how to use his skills effectively. One thing about how the Knick’s offense flows right now is that they don’t look for the weak side 3pt spot-up that often as a primary option. Stoudamire tends to pass back out to the strong side wide, or the far side guard, and when Felton keeps the ball on the PnR, he doesn’t go to the weak side that frequently. When the ball’s moving around the key, Gallo most frequently shifts it down into the corner to catch the defense out of rotation.

    There was an interesting play last night when Chandler took the ball on the wing, Gallo set up on the weak side 3pt spot, and Chandler saw a seam develop in the defense. He drove across the key, Raptors collapsed, and he kicked it over to Gallo for an open three. Play was waved off because of defensive three seconds, but it was one of the rare times when they worked a play to take advantage of the kick-out opportunity.

    Gallo’s pretty efficient with low usage, keeps his turnovers to a minimum and tries to find places on the court where he can take advantage of imbalance on the defense. (His defense is another story…) It’s the kind of solid player that good teams take advantage of, and this year the Knicks have been doing that.

  46. Mike Kurylo Post author

    Thomas B.: Great discuss here at calculusblogger.net.

    Funny thing is this isn’t calculus. It’s basic statistics.

    Abasi: 3-1-8-1-2-1-3-5?

    No, they are not consistent, if you define “consistent” as “close to the “average”. The average for that sequence is 24/8 = 3. So the problems becomes: how CLOSE does it have to be to be considered “consistent”? That’s a very subjective value.

    Exactly. Your method of looking at a player’s games on it’s own requires a subjective judgment. The purpose of using logic/math/statistics is to eliminate the subjective element. So if I take an average and look at how close something is to that average (*cough* standard deviation *cough*) I need some way of knowing if that is an acceptable deviation from it, not just squinting my eyes and saying “yup that’s right.”

    I mean could you imagine going to the doctor, having him drawing a blood sample then just holding it up to the light and saying “yup you have X disease?” If you go half the way through a scientific method, you might as well follow it as far as possible before you allow a subjective take on it. Dr. Abasi is saying Gallo is inconsistent because he believes it and the numbers kinda look right. I’d rather have the lab technician take care of that blood sample.

  47. The Honorable Cock Jowles

    Mred696 with the serious technical writing skills. Well done.

    Posts like these are why KB is among the best sports blogs on the market. I’d like to see this study again, but controlled for other variables, such as defensive rating of opponent, game score of typical SFs against those opponents, lineups, usage rates, and per 36 numbers.

  48. ess-dog

    I think it is agreed that by sight and by statistic that Gallo is inconsistent. The question really should be “is Gallo’s inconsistency acceptable?” and I think the answer is “yes, of course.”
    Firstly, the nature of the small forward position is inconsistent: we expect some scoring, some slashing, some rebounding, some shooting, etc. but it is less of a specialist position than say point guard or power forward, although people often seem to generally expect “points” from their small forward in some fashion.
    Gallo’s fg% is low – .397 – even for someone who is primarily a 3 pt shooter. But his 3pt% is fairly high despite a “slump” at .374 and he has shown the “inconsistent” ability to get to the line on top of that…

    Oh and btw, he is 22 years old.

    Here are Gerald Wallace’s game logs from his 22 yr old season if anyone is interested in looking:
    http://sports.espn.go.com/nba/players/gamelog?playerId=1026&year=2005
    Wallace doesn’t excel in shooting the way Gallo does, but I just picked him b/c he’s at about the level that we would like from Gallo eventually (low tier all star) and I’m too lazy to look for anyone else better right now. He also strikes me as a 3 that can play the 4 like Gallo.
    He averaged 5 and half boards and 11 points that year. If you look at the log his stats are similar to Gallo’s: 18, 16, 6, 22, 6, 12 points… 10, 7, 3, 2, 5, 7 rebounds… fairly inconsistent. And not a much better fg% of .447 considering how many less 3pt shots he takes.

    Yes, if you compare Gallo’s 22 yr old season to Lebron’s 22 yr old season, he comes up short. But I think he really is more or less on track, he just needs to define his role and figure out what he needs to focus on giving the Knicks on offense while improving his defense.

  49. Abasi

    @52

    Pure nonsense, mike – sorry to be so blunt.

    If you bring home 200, 2000, 1500, 160, 800 bucks every week, your earnings are ‘inconsistent”. substitute “earnings” for “points scored”, “rebounds” or anything else and the same principle appiles. you don’t need to know what the neighbor is making to know that your week-to-week earnings are inconsistent. As simple as that.

    End of thread for me (don’t want to be banned!).

  50. Thomas B.

    I have to say that the Mike vs. Abasi tete-a-tete (which I see as more a philosophical difference) is made possible only by the Knicks being 5 games above .500. Think about it, a year ago we’d been 10 games under and we wouldnt have time for this. We’d be calling for jobs, proposing trades, and posting endless jokes about how fat Eddy Curry is. Look how far we’ve come as fans. Isn’t this a great sign? I love it.

  51. ess-dog

    Abasi: @52Pure nonsense, mike – sorry to be so blunt.If you bring home 200, 2000, 1500, 160, 800 bucks every week, your earnings are ‘inconsistent”.

    dude, you are driving me crazy. you are saying these paychecks are inconsistent right? but let’s say, to use an extreme example, that you live in a country with 100 people and 99 of those people get random paychecks like $10, $2000, $5000, $50, $3000…. then YOUR example is actually MORE consistent than anyone receiving paychecks in your little country, get it?

    You need comparables in order to have statistics.

    You’re taking the “rest of society” for granted in your example.

  52. villainx

    “If you bring home 200, 2000, 1500, 160, 800 bucks every week, your earnings are ‘inconsistent”.”

    I’m going to post this even after reading ess-dog’s 58.

    But if your neighbors all earn $0, $0, $0, $0, $4660. To average the same $932. 200, 2000, 1500, 160, 800 bucks every week would be viewed as relatively and probably preferably more consistent. No?

  53. DS

    @55 – Good logic. Accordingly, I think D’Antoni would be wise to play Azubuike at center when he’s healthy again, because he provides some serious height. 6’5″ is tall! Please don’t question me by comparing him to other NBA players though. Anyone w/ brains knows that 6’5″ is really tall.

  54. Mike Kurylo Post author

    Abasi: @52Pure nonsense, mike – sorry to be so blunt.If you bring home 200, 2000, 1500, 160, 800 bucks every week, your earnings are ‘inconsistent”.

    Abasi – What if you are a salesman? That salesman, when compared to people who get the same paycheck every week is inconsistent. However when compared to other salesman in the same field with the same product it might not. What if the other salesmen were more like 0, 3500, 0, 0, 1160. Your salesman would be consistent!

    I get what you’re saying. You can look at something at say it’s inconsistent. But what your brain is really doing is comparing it to what you think is the norm. Most people expect the same (or thereabouts) paycheck each week. They’ve already calculated an acceptable standard deviation. You’re not looking at that guy’s paycheck (or Gallo’s stats) in a vacuum. You’re comparing to a value you deem as acceptable. Hence you’re already comparing it to another value you have, you’re not looking at those numbers individually.

    For instance what if I told you a friend arrived every Monday at 1:05, 2:15, 4:50, 3:48, and 12:36. You’d think your friend was a flake if they were coming by subway because you know what the deviation is for a subway trip. But if you’re friend were flying in from Asia? That would be acceptable. That’s because you already have compared these numbers to what is expected or acceptable, which you’ve attempted to approximate. The numbers have to be put in context by comparison. Individually they don’t mean much.

    BTW I would never ban someone for making a point. It’s when these kinds of debates get personal that the banning mouse comes out of my drawer.

  55. TDM

    Thomas B.: Think about it, a year ago we’d been 10 games under and we wouldnt have time for this. We’d be calling for jobs, proposing trades, and posting endless jokes about how fat Eddy Curry is. Look how far we’ve come as fans. Isn’t this a great sign?nbsp; (Quote)

    I could use a good Eddy Curry joke after reading this thread.

    Eddy Curry is so fat he got arrested at the airport for ten pounds of crack. Rimshot. I’m here every Thursday, try the veal.

  56. ess-dog

    JK47: Okay, capologists.Is there any feasible way to make this roster a reality for next season:PG Ray Felton
    SG Landry Fields
    SF Carmelo Anthony
    PF Amar’e Stoudemire
    C Marcus CambyBench:
    Danilo Gallinari
    Timofey Mozgov
    2011 1st round pick
    Shawne Williams or whoever elseThat would be a pretty fierce SSOL team.This would assume we’re subtracting Eddy Curry’s carcass, Wilson Chandler, Anthony Randolph and Toney Douglas.But I’m not gonna lie– I don’t fully understand the salary cap and all the contracts involved.I think Camby is owed $10M next season and Melo of course would want a max contract… or would he?Could we perhaps convince Melo to take a salary “haircut” like his d-bag friends in Miami?The beauty of it would be that Camby’s contract is coming off the books after 2012– it’s not like it’s a brutally long contract.In terms of pure basketball, Camby is such a perfect fit for this squad it’s ridiculous.His scoring efficiency has been terrible but his usage is so low that it almost doesn’t matter.But he still doesn’t kill you on the offensive end, because he’s a really good offensive rebounder– 7th best in the NBA this year.On defense, he’s a beast.Fearsome shot blocker, ranks #2 in defensive rebounding percentage behind Kevin Love, has a defensive rating of 101.  

    It’s hard to say what Portland wants though. Their swingmen are under-impressing so far: Batum, Rudy, and Aldridge are all about average. Matthews is playing ok. They would want a cheap player under contract for a while and cap relief for next year, so either Landry, Gallo or AR plus Curry’s contract.
    Camby’s not playing that great by his standards either, plus he’s old as dirt… maybe they’d even take Shawne + Curry?
    We’d have to lose a lot of bodies anyway if we wanted to fit in Carmelo next year though. I don’t see DW taking that risk…

  57. NYK Ewing

    Mike, is what you’re saying akin to saying a player averages 35 points a game means nothing (not good or bad) unless you knew that other players average much less? As in, inconsistency by itself is a completely meaningless term (that st. dev you got) unless you know how it looks in comparison?

  58. SeeWhyDee77

    I am beginning to believe that people (myself included) only call Rooster inconsistent because of the expectations placed on him. Would I love to see him score more and shoot more consistently? Sure. But at the same time even I have to weigh him fairly. I remember him bein compared to Kukoc and being hailed as the best shooter Mike D has ever seen. Then I remember him bein compared to Dirk and Stojakovich. Wow..can u imagine…this kid would literally have to make EVERY shot to live up to expectations. He hasn’t developed Kukoc’s passing yet nor has he developed the ability to get buckets like Dirk and Stoy. But what he does is positively affect the game in every other way when his shots are not falling. He works hard on defense and rebounding, and he has been making good decisions when passing. He’s been vocal off and on the court and he’s NEVER wilted in the face of a challenge. And when his shots ARE falling, he has been a thing of beauty to watch. So..maybe not inconsistent, just not who everybody says he’s gonna be. He’s gonna be the 1st Gallinari, not the next Dirk or Stoy or Kukoc. His ceiling may be somewhere in between. It took me a while to get over the fact that he wasn’t living up to expectations becuz I drank the kool aid. But watchin him play, I have no problem with him at this point. Plus he’s still very young. Maybe the mustache should have kept his excitement about Rooster to himself.
    On another note, considering Mayo’s precarious position in Memphis…if it were available would anybody do a Chandler, AR, Walker, Mosgov swap for a package headlined by Mayo and Thabeet? Here’s my reasoning-they would get potential in AR and Moz and a solid rotation player in Chandler. Walker’s useful as well. Xavier Henry’s the future at the 2 out there it seems, and Thabeet would replace AR and Chandler’s shotblocking..and we would get Mayo-who I think needs a change of scenery.

  59. iserp

    @Abasi

    Take Gallo’s numbers and only Gallo’s numbers.

    Put him in the NBA, where every other player is more consistent than him. Me, as a spectator would call Gallo inconsistent.

    Now, put him in the Italian League where (let’s imagine) every other player is more inconsistent than him. Then, i would call Gallo consistent.

    So, not only i need the numbers, but i need some other players to compare him with. Why? Because being “consistent” is not an absolute value, but a relation of order. Like “big”, is 1000m2 big? Yes if you are talking about houses, but not if you are talking about countries. With only one caveat, consistency has an absolute maximum, deviation 0. If gallo hits 15, 15, 15, 15, 15, … he can’t be less consistent than anyone else, but that’s the only absolute you have, for anything else, you need to compare him with someone else (or better, with the average of the rest of the league)

  60. SeeWhyDee77

    That would give Mike D the freedom to bring Azu (when healthy) and Fields off the bench to play on the wing and give Shawne Williams more burn as a 3/4 hybrid. I guess the only problem with that is Thabeet sux worse than Turiaf on offense and it would force Douglas to primarily play the point which he hasn’t been doin 2 well lately. I don’t know..just tryin 2 get a cheap 2 guard who can score and handle the ball. I don’t know what happened to Mayo, but I know he’s way better than he’s shown this season.

  61. Thomas B.

    @64
    I think so. How do you establish success until you establish a norm? Is 35 points per game good? Well it all depends on what the other players in the same situation are doing. Let’s say Gallinari averages 35 in a league where the average player scores 60 and the best player scores 85. Gallinari might night look so good then. You need context to figure of how far an event is from the norm. Good and bad are just meaningless terms absent context. The tete-a-tete here is about how a person gets to define consistent. I think I get what Abasi is trying to say but the problem is that it does not apply to the set of rules Mike picked for the discussion. I think Mike’s model is a good one. It seems that Abasi advocates some other model and I’m not sure that is the best way to determine whether Gallinari is consistent. It’s all about what you are looking for in the pattern.

    If Gallinari scores: 10, 20, 30, 11, 21, 31, 12, 23, 32. You might say Gallinari was not consistent, due to the 20 point swing, but over every three game period, he is pretty consistent. It’s hard to argue with Mike as he identified consistent in terms of standard deviations. Then applied the numbers to the definition. Abasi seems to want to dismiss that model, in favor of some other means to define consistent.

    They seem to argue over the rules by which consistency will be defined, which is why I see the argument as philosophical in nature.

    Damn it now I’m sucked into this.

  62. iserp

    SeeWhyDee77: On another note, considering Mayo’s precarious position in Memphis…if it were available would anybody do a Chandler, AR, Walker, Mosgov swap for a package headlined by Mayo and Thabeet? Here’s my reasoning-they would get potential in AR and Moz and a solid rotation player in Chandler. Walker’s useful as well. Xavier Henry’s the future at the 2 out there it seems, and Thabeet would replace AR and Chandler’s shotblocking..and we would get Mayo-who I think needs a change of scenery.

    I don’t think Thabeet would earn more minutes than Randolph is currently. And i am not a fan of OJ Mayo; but he is not a bad player, and his rookie contract still lasts till 2011. So, in my eyes it is Chandler (who is a RFA in 2010, and we might not want to pay him) for Mayo (who is a RFA in 2011, so we can keep our options intact till then). Besides, if Melo comes, it is a SF less and a SG more, so our team would be more balanced. All in all, i’d do it.

    (Reading again, i realize i’ve given up on AR very easily; but the truth is that we’re better trading him before plunging totally his trade value)

  63. DS

    Forget Mayo and Thabeet…
    Behold, ESPN Trade Machine greatness:

    Knicks get: Carmelo, Balkman, and Maynor
    Thunder get: Landry Fields, J.R. Smith
    Nuggets get: Jeff Green, Eddy Curry, Azubuike, Toney Douglas

    The Knicks secure ‘Melo this year, get a legit backup point (2.3 A/TO), and replace some of Toney’s perimeter D (RENALDO!) and keep their best 4 players.

    The Thunder get a younger, more balanced player Fields to replace Green and a vet shooter. They don’t have to worry about losing Green to free agency nor leaving enough money/cap room from Westbrook.

    The Nuggets save a ton of money, get younger, and as those of us in the know, know they improve statistically. Green, Nene and Billups keep them in a playoff hunt. Green becomes a potential stud as a first or second (instead of third) option scorer.

    DONE AND DONE!

  64. ess-dog

    BTW has anyone looked at Carmelo’s TS% lately? Pee-ew. .524. He’s rebounding better than ever and his assists and blocks are improved, but Wilson’s TS% is at .561… and he plays more defense. I know Melo really wants to be a Knick, but I just don’t know… he needs to show us something first.

  65. Spree8nyk8

    I have not researched this, I am actually just asking….

    How consistent is Gallo’s usage? Because on night’s when the PnR is going strong or if Amare/Felton/Chandler are all playing well they just don’t really seem to go to him, which I don’t mind at all. But when he is actually needed he always seems to be there for the most part (first week of the season disregarded).

    My point is that I’m wondering if his inconsistency is linked to his usage. And I don’t even know how well that can be tracked because he is still on the floor at times when he’s not getting much opportunity. Idk honestly how to look at it. I just know that there are a lot of games where his stat line is incredibly efficient (17 pts on 7 shots, etc) but I mean that’s not a ton of usage. How often are you going to expect a player to get that much done with that few opportunities?

    Is it possible there is a variable missing?

  66. DS

    Knicks: Felton, ‘Melo, Gallo, Chandler, Stat Bench: Turiaf, Maynor, Williams, Balkman, Walker, Randolph?
    Nuggets: Billups, Affalo, Green, Forbes, Nene Bench: Lawson, Douglas, Azu, Birdman, Al Buckets
    Thunder: Westbrook, Sefolosha, Krstic, Durant, Fields Bench: J.R. Smith, Collison, Ibaka, Harden, throw in Mosgov!!

  67. Brian Cronin

    Not even saying that that is a bad trade or anything, Dan (as honestly, it seems fair enough) but there’s no way Denver sells to their fans trading Melo for Jeff Green and Toney Douglas.

  68. Abasi

    Mike Kurylo: Funny thing is this isn’t calculus. It’s basic statistics. Exactly. Your method of looking at a player’s games on it’s own requires a subjective judgment. The purpose of using logic/math/statistics is to eliminate the subjective element. So if I take an average and look at how close something is to that average (*cough* standard deviation *cough*) I need some way of knowing if that is an acceptable deviation from it, not just squinting my eyes and saying “yup that’s right.” I mean could you imagine going to the doctor, having him drawing a blood sample then just holding it up to the light and saying “yup you have X disease?” If you go half the way through a scientific method, you might as well follow it as far as possible before you allow a subjective take on it. Dr. Abasi is saying Gallo is inconsistent because he believes it and the numbers kinda look right. I’d rather have the lab technician take care of that blood sample.  (Quote)

    But what if “Dr. Kyrulo” is wrong and clearly more confused than “Dr. Abasi”?

    Most of the ‘lab technicians’ would be reluctant to call the doctor naked, I’m afraid…

    ;-)

  69. DS

    @74 – Brian, I agree it seems like better PR for the Nuggets to suggest to the average fan, “‘Melo abandoned us” than to look like they gave away ‘Melo for a bunch of guys they prob. never heard of… but I think that is short sighted. (1) ‘Melo is going to leave (2) you can throw ‘Melo under the bus and say you didn’t feel like playing the LeBron game w/ him (3) fans understand when it’s time for a change (4) that trade would make the Nuggets better this season anyway (5) Green might put up decent numbers in that scenario. You could make him a celeb.

    Sorry to be so argumentative over a complete hypothetical! :)

  70. Brian Cronin

    Oh, no, don’t get me wrong, I agree that it would be short-sighted of Denver. Just noting that NBA teams are often short-sighted.

  71. latke

    As I’ve posted previously, the reason Gallo’s scoring is inconsistent is because he is largely effective as a secondary option. He doesn’t have the strength to post up or the quickness to beat a defense that is set. When he is most effective is after another quicker or stronger player — largely Felton or Amare, but also Chandler and Fields — breaks down the defense and the ball rotates out to Gallo.

    In these situations Gallo is deadly. He is a strong perimeter shooter, and when the defense does close out, he has an excellent head fake that he uses to draw fouls on defenders that run out on him, and he has a good sensibility about putting the ball on the floor and drawing contact from that. It’s not just that he has these skills, but he has excellent instincts in regard to how to use them. He knows when to put the ball on the floor, when to throw a head fake, and when to just put up a shot. This is why his turnovers are so low and why, despite the relatively low FG%, his TS% is so high.

    There are times, however, when a teams defense does not shut down that first move from felton, amare, fields, or chandler. The ball never gets out to Gallo after a defensive breakdown because the defense doesn’t stop the initial foray in to the paint. In these games, gallo’s scoring drops; however, he is still a weapon in that the reason defenses aren’t stopping the initial drives is partly because they choose to stay close on Gallo.

    Lastly, I want to say that this is an ideal situation for Gallinari to grow. The team can successful with him simply scoring in the ways he is already comfortable. This I believe will help him to become more and more comfortable on the court and to experiment in expanding his offensive repertoire without feeling responsible (as he seemed to feel early on) to do those things. Barring injuries or major moves, I think the Gallo you’ll see in game 82 of this season will be a vast improvement from the gallo you’re seeing now.

  72. Spree8nyk8

    Reg (London)

    Given Felton’s play, who should the Knicks target in trades / free agency?
    John Hollinger
    (2:02 PM)

    Well, given that Carmelo Anthony is pretty much hell-bent on going to New York next year, it would be awfully silly for New York to allocate any future cap money that could be used to sign him. Knicks could trade with Nuggets and get him sooner if they can interest Denver in some kind of Gallo/Fields/Douglas platter, but they can afford to be patient too.

  73. Mike Kurylo Post author

    NYK Ewing: Mike, is what you’re saying akin to saying a player averages 35 points a game means nothing (not good or bad) unless you knew that other players average much less? As in, inconsistency by itself is a completely meaningless term (that st. dev you got) unless you know how it looks in comparison?  

    A. It’s not what a person averages, it’s what he deviates from that average.

    B. The deviation from the average has to compare to something. How much is your guy deviating to the average when compared to others?

    If you don’t get this thread, read @55 @58, @59, @61.

    Maybe I can sum it up simpler. Take Abasi’s example, but put it in the abstract. Is the series of numbers 200, 2000, 1500, 160, 800 consistent? Depends on what these numbers are.

    Minutes earned in a season for an NBA player = Inconsistent.
    Distance from cork in a dart game for a teenager in MicroMillimeters = Consistent
    Number of minutes you’ll call that girl after your date ends = Inconsistent
    Number of ballots in the US presidential election where someone wrote in a cartoon character = Consistent

    Same numbers can be seen as consistent or inconsistent, based on what you are comparing them to. You need that *similar thing* for comparison.

  74. Bruno Almeida

    DS: @74 – Brian, I agree it seems like better PR for the Nuggets to suggest to the average fan, “‘Melo abandoned us” than to look like they gave away ‘Melo for a bunch of guys they prob. never heard of… but I think that is short sighted. (1) ‘Melo is going to leave (2) you can throw ‘Melo under the bus and say you didn’t feel like playing the LeBron game w/ him (3) fans understand when it’s time for a change (4) that trade would make the Nuggets better this season anyway (5) Green might put up decent numbers in that scenario. You could make him a celeb.
    Sorry to be so argumentative over a complete hypothetical! :)  

    Yeah, but Jeff Green is on an expiring contract… if they want him so bad to replace Melo, they can just wait and get him as a free agent.

    I too feel that it makes sense for Denver to trade him, but they seem to be waiting for a killer offer, specially if the rumored Nets offer is really on the table (which I highly doubt).

    What I mean is, if they trade Melo, it’ll be a desperate move and it will happen closer to the deadline imo.

    Also, there’s the Chauncey Billups issue, they’ll probably try to find a trade for him too, which really opens up the possibilities, specially for 3-team trades, if they truly want to gut the roster and start from scratch.

  75. mred696

    Again, I would say that you’re still arguing over the definition of “consistent”. I think the problem arises in that Kurylo is using the word “consistent” to mean two similar, but different concepts.

    Consistency, as it is strictly applied to a set of numbers, is the amplitude of deviation from the mean of that set. Standard deviation is one method in statistics to quantify this concept. Gallo’s play is inconsistent in this way because there is variation in his stats. If he put up the exact same stats every day, he would be consistent. Indeed, if he never played and put up zero stats all the time, he would be *statistically* consistent, and his standard deviation would be zero.

    Consistency is also used to denote the degree of approval of a player’s statistical consistency. When we colloquially say a player is “consistent”, we generally mean that he plays at a level we appreciate, and performs at this level often. We might say, for instance, that Pau Gasol is consistent because lately, he has been playing well and playing well often enough to earn praise.

    This duality of meaning is similar to the word “good”. Something that is “good”, like honesty, is morally praiseworthy. We might also say that it is “good” when someone is honest, to express our approval of his morally praiseworthy action.

    When we say that Gallo is inconsistent, we are really saying two things about Gallo, both of which are statistically supported. First, he is statistically inconsistent from game to game because his stats vary from game to game, resulting in a standard deviation greater than zero. Second, when we say he is inconsistent, we express our disapproval that he is performing at his expected level only intermittently.

    To further illustrate the point: Dirk Nowitzki is statistically inconsistent in that his stats vary daily. But we would say he is consistent (meaning we approve of his performance) because he usually performs at the level expected of him.

  76. nicos

    As others have noted, it’s not that surprising that Gallo’s numbers aren’t incredibly consistent- he’s the third option right now behind Stat & Felton (and sometimes Chandler as well). What annoys me is how inconsistent his effort level seems on defense and on the boards- there are some games where he’s very active and others where the only time you notice him is when he’s late on a rotation or failing to get back in transition. His rebound numbers do get impacted by who he’s guarding- some games he’s farther out on the perimeter than others- but he still just stands around far too often once shots go up rather than going after the ball or looking to put a body on someone. As people point out he’s only 22 but he has been playing professional ball for a while now and spent his whole life around the game via his father so sometimes I feel a little less forgiving of him than I do of someone like AR who’s played a lot less organized ball.

  77. villainx

    With the talk about Carmelo, sounds like he’d be a nice addition, but that would still not address the main deficiency of the team, which I assume would be center. Or are there other areas that require drastic upgrade? I’m assuming Felton has shown enough to be serviceable as point guard, or is the consensus is he’s not enough?

    Or would an upgrade somewhere else in the front court make up for what’s lacking in the center position, like say some defensive minded forward.

  78. Spree8nyk8

    Bruno Almeida:
    I too feel that it makes sense for Denver to trade him, but they seem to be waiting for a killer offer, specially if the rumored Nets offer is really on the table (which I highly doubt).  

    If that offer were on the table he would have been traded already. It’s posturing.

  79. Brian Cronin

    It doesn’t have to be posturing. It could be that that offer is on the table, but only on the condition that Melo signs an extension with the Nets, and the Nets are waiting for Melo to change his mind about signing with them.

  80. fanshawe

    mred696: Consistency, as it is strictly applied to a set of numbers, is the amplitude of deviation from the mean of that set. Standard deviation is one method in statistics to quantify this concept.

    But even this simple definition requires some contextual assumptions. Is a player who point totals are 10-10-10-10-11 “inconsistent” because the variance is more than 0? If so, then “consistency” is a useless descriptor for a basketball player because everyone would be inconsistent. If not, then you are essentially arguing that the variance is low enough for the player to be called consistent. But when does the variance stop being “low enough”? 10-10-10-10-15? 3-3-3-3-38? What about 20-0-20-0-10? If you want your answer to be anything more than an aestheic judgment of what you want boxscores to look like, then you need to know what a normal amount of variance is, and the way to figure that out is to look at other players.

  81. TDM

    Thomas B.: first anagrams, now this . . .

    By the way, for Ronny Turiaf do you prefer “Our Infantry” or “Furry Nation?”

  82. latke

    Bruno Almeida: I too feel that it makes sense for Denver to trade him, but they seem to be waiting for a killer offer,

    I think the most plausible explanation is that they don’t see the Nets’ offer going away, so they’re better off waiting and letting NJ lose more games so that the pick they get in the trade is better. I know a lot of people think that the Nets’ offer is not for really, that it’s just being put out there to try to get other teams to raise their offers, but even if that’s the case, I think teams in general will have a clearer picture of where they stand come january/february. Look at the knicks. If we are only a few games over .500 come the trade deadline, we may up our offer, and if we’re not, perhaps a team like OKC or Dallas realizes that carmelo really could put them over the top. Dallas would probably have to get a third team involved, but Beaubois is a very promising young player who would make a nice starting point in a trade for Anthony.

  83. Spree8nyk8

    Oh, ok let me rephrase…..THAT is completely possible. I’m sure that offer could be on the table if Melo would sign. My point is that it does’nt seem like he’s willing to sign outside of NY. There has been no indication that he would be willing to go to NJ. And that deal is NEVER happening if he doesn’t sign. I just really don’t see them getting much of an offer from any teams without him signing the extension. So they have no leverage and this time the Knicks need to be ruthless. Gouge them. People ALWAYS do it to us.

  84. Mike Kurylo Post author

    TDM: Thomas B.: first anagrams, now this . . .
    By the way, for Ronny Turiaf do you prefer “Our Infantry” or “Furry Nation?”  

    “Furry Nation” – by far!

  85. Spree8nyk8

    TDM: Thomas B.: first anagrams, now this . . .
    By the way, for Ronny Turiaf do you prefer “Our Infantry” or “Furry Nation?”  

    A Fury in Front!

  86. DS

    @82 – I bet you the scenario I came up with is better than what the Nuggets end up getting. Also, Green is a restricted free agent and the Nuggets would a big leg up in retaining him.

  87. Spree8nyk8

    And with regards to Melo, if it actually gets to the deadline, at that point I wouldn’t even give up gallo in the deal. I lowball them and say see ya in 4 months. If it gets to the deadline then they have really struck out finding a suitor.

  88. BigBlueAL

    Rumors of a possible trade for Andre Iguodola?? If it wasnt for the length of his contract I wouldnt mind but doubt it will happen for that reason anyway.

  89. Spree8nyk8

    BigBlueAL: Rumors of a possible trade for Andre Iguodola??If it wasnt for the length of his contract I wouldnt mind but doubt it will happen for that reason anyway.  

    link?

  90. taggart4800

    I think the trade for Melo is going to happen prior to the deadline. I just think that DW would be run out of town if he didn’t get him and thats not to say Melo is the be all and end all. The media has however put a big onus on the aquisition and both DW and D’Antoni hinted a lot about the aquisition of two max guys. Interesting note in Hahns blog that if Azu doesn’t play by Jan 21st then 80% of his salary is paid by insurance. That is a pretty big trade chip and I bet the team has mixed feelings about the Azu’s return to the floor.
    How about a deal centered around these components:-

    NY Denver Hawks
    Melo J Smith Nene
    Azu
    Curry
    Turiaf
    Obviously pieces would be included, maybe a draft pick from the Hawks. Works out pretty good for all the teams involved and Denver aren’t going to get much better returns than that and Atlanta should really look to acquire a quality big.

  91. Spree8nyk8

    If we did that trade we would definitely need an immediate trade for a big. Don’t really see how they can part with turiaf for a 3 right now.

  92. DS

    If ‘Melo isn’t traded at the deadline wouldn’t the Knicks WANT him to become a free agent rather than play ball on a 3 team trade?

  93. Bruno Almeida

    DS: @82 – I bet you the scenario I came up with is better than what the Nuggets end up getting. Also, Green is a restricted free agent and the Nuggets would a big leg up in retaining him.  

    yeah, I agree, it’s just very nebulous right now, we can’t know for sure what Denver is willing to do to move Melo before he leaves as a free agent.

  94. taggart4800

    Yes but then you can keep chandler as well. we are esentially playing without Turiaf now though and seem to be doing fine. If we sidn Melo as a FA we are going to have to renounce those traded peices anyway and this way we get access to the MLE.

  95. Frank O.

    About halfway thru this thread I nearly guaged my eyes out.
    You see, I’ve been tracking your U.S. Senate’s dance of futility today. Stopping for a moment to read Mike and abasi debate over the nuances of inconsistency, I found myself growing inexplicably angry.
    I liked when we were shown the variance in Gallo’s eFG%. The disparity was eye-opening.
    But I do believe any comparison requires a control, some context in which to view the variances. These are people, not machines. The people with the least amount of variance in their performance are probably the most machine-like.
    I suspect it would be useful also to determine the average variance overall, and then you have a good sense of how people perform.
    I think looking at just a certain type of player is too narrow.

    Either way, I had to stop reading. :) I blame the Senate.

  96. iserp

    @mred696

    If your definition of “statistically consistent” is to have 0 deviation… then it is little useful in this conversation and i think nobody used it in that meaning.

  97. Z-man

    I vote NO in Iguodala. Too expensive and Landry Fields might turn out to do the same things more efficiently in short order.

    Re: Gallo, Mike K, it is so obvious that he has been inconsistent that I don’t know why you bothered. I might have chosen a different methodology, though. The question on my mind is: is Gallo acceptably inconsistent given his age, the amount of NBA games he has played, and his current role? Comparing him to Johnson, Allen and Arenas, all seasoned pros yet all playing very different roles, seems weird. Even Chandler and Gallo play different roles on the Knicks, and Chandler has played nearly double the minutes that Gallo has.

  98. hoolahoop

    I forget who I heard say it, maybe Larry Brown, Phil Jackson, or possibly D’antoni. They said that any player in the league can have a big scoring game on any given night. The great players do it every night. Guys get into a zone like they never had before, and they look like an all star. Why don’t they do that every night is the first question that comes to mind. (I thnk the answer is obvious – mental). And some fans equate that performance with how good the player is, though it’s forty games before he has another night like that. Gallo is starting to look that way.

    Bottom line, if Gallo can’t shoot lights out, he’s just a great white hope with potential that’s really a mediocre player at best.

  99. hoolahoop

    Okay, you’re starting a team and you have to choose your first player between Gallo and WC. Who do you pick?

  100. KnickfaninNJ

    I am very happy we are looking at standard deviations to judge consistency. This is the right thing to do. But I think I can contribute a little statistical background that might help the discussion. The theoretical distributions of something like a basketball shot, where there is a specific probability of success and you either succeed for fail is called a binomial distribution. For this distribution if you look at the expected variation in scoring, the standard deviation in how many points are scored will depend on how many attempts per game are made. The actual variation should be proportional to the square root of how many shots are attempted. (It also depends on the shooters shooting percentage, but this is a much smaller effect for the shooting percentages we are talking about). This means that a high volume shooter will have a higher standard deviation. But, and this is a big but, that shooters standard deviation will be a smaller percentage of his total points. For example a shooter who shoots 40% on 3 pointers and who shoots ten a game will theoretically make 4 of them, score 12 points and have a standard deviation of 4.6 or about 39% of his total scoring. A shooter who shoots the same percentage but take 20 shots will score 24 points and have a standard deviation of 6.6, which is 27% of his scoring total. I think this would make the high volume shooter appear more consistent from game to game to the public, even though they are basically the same player.

    So if you want to accurately compare different player’s shooting variability, you should calculate the squares of the standard deviations of their points per game (this is called the variance) and then divide that number by the number of points each would have scored if all the shots went in. I would be interested in seeing this calculation done for different players but it’s not so easy to do, because you have to know how many shot attempts of each type the player took.

  101. KnickfaninNJ

    To finish the post.:

    The measurement I suggested would be purer measurement of the players shooting variability than just a standard deviation would.

    Looking at some theoretical numbers for standard deviations of binomial distribution I think you will find that all the actual game to game standard deviations are higher than theoretical. This suggests that there are more variables in how a player succeeds that just his intrinsic shooting ability (Defense matters (among other things). But it also makes the numbers more interesting.

  102. Brian Cronin

    Big no to AI.

    I like him, but he’d basically be taking minutes from Fields and Chandler, and the difference between those two and AI sure isn’t worth whatever AI is getting paid (not nearly as much).

    So no thanks.

  103. Brian Cronin

    Okay, you’re starting a team and you have to choose your first player between Gallo and WC. Who do you pick?

    Gallo.

  104. Bruno Almeida

    Brian Cronin: Big no to AI.
    I like him, but he’d basically be taking minutes from Fields and Chandler, and the difference between those two and AI sure isn’t worth whatever AI is getting paid (not nearly as much).
    So no thanks.  

    I don’t know, I think both Chandler and Landry need to develop A LOT on defense to be on Iguodala’s level.

    but I agree his contract isn’t worth it, we’d go back to the “get stuck on huge multi-year deals for non-superstars” era once again.

    I still think waiting Melo’s situation out is the best idea, if we can get him on a favourable trade or as a free agent, do it, but if we can’t, stick with this promising roster.

  105. Z-man

    I can completely tolerate Gallo’s inconsistency, just as I had no problem tolerating WC’s spotty shooting and lack of conviction going to the basket. Gallo is 22 and already is an above average, intangible-type player. He is way ahead of where Bargnani was at age 22 and not far from where Rashard Lewis was at age 21 with more NBA experience. He has shown the propensity to take over a quarter, or a game, at any time with an explosion of points, and does little things to help when the shot isn’t falling; he rarely turns the ball over. When he disappears, we have reinforcements at that position (WC, Williams, Walker.) I see no reason why he won’t get more consistent with age and experience, like many very good players do.

  106. Brian Cronin

    I don’t know, I think both Chandler and Landry need to develop A LOT on defense to be on Iguodala’s level.

    Totally agreed. But they’re both better offensive players, so when you put it all together, I don’t think he’s nearly worth paying soooo much more than both of them.

  107. Bruno Almeida

    Brian Cronin:
    Totally agreed. But they’re both better offensive players, so when you put it all together, I don’t think he’s nearly worth paying soooo much more than both of them.  

    yeah, the problem is that Iguodala’s contract is REALLY unreasonable, he was never deserving of that much money.

    I think that the fact that he’s a great defensive player kinda offsets the offensive loss we’d have by losing Chandler or Gallo simply because this team needs an elite defensive player more than an above-average offensive player.

    but he’s still not worth it.

    what about Marc Gasol? he’ll be a free agent next season, and every time I watch him I get the feeling he could become an elite player on the right situation.

  108. Brian Cronin

    Oh yeah, I’d totally be cool with them pursuing Marc Gasol.

    They can’t let him go, though, right?

  109. drm1977

    To paraphrase knickfaninnj, low usage high efficiency player will appear more inconsistent than high usage player of similar efficiency. Comes to sample set size. The variable data affecting consistency is too closely tied to usage to be statistically reliable.

    BUT, most of us agree that Gallo is inconsistent. That’s a judgment on the totality of his game and it isn’t wholly measurable with the current tracking data.

    That’s the beauty of basketball.

    Go look at Dennis Rodman. He had a high degree of deviation in short term game results.

    At the same time, he was one of the most consistent players I ever saw. He set unreal picks. He defensed his part of the court. He chase the ball.

    Great site, great dialogue. I’m three comments in and am going to go back to lurking.

    Go Knicks.

  110. ess-dog

    hoolahoop: Okay, you’re starting a team and you have to choose your first player between Gallo and WC. Who do you pick?  

    Gallo.

    But I might take WC over Iggy…

  111. Bruno Almeida

    Brian Cronin: Oh yeah, I’d totally be cool with them pursuing Marc Gasol.They can’t let him go, though, right?  

    well, with Gay’s huge contract and Conley’s unreal deal they’re pretty commited to this core, and Randolph’s contract expires this season… my bet is that the Grizzlies won’t let Randolph go no matter what, and he was reported as looking for the same type of money Pau Gasol got, so they’ll have A LOT of money commited to those 3 players + the need to keep something out there for Mayo when he’s eligible for an extension.

    Marc Gasol is actually a RFA, but if we make a big offer they probably won’t be able to match it, something like the Wesley Matthews deal or the Paul Millsap one.

  112. latke

    The Knicks do gain a certain advantage from trading expirings — if they can get over the cap, then they can resign chandler via bird rights and gain access to the MLE. I’m wondering what ya’ll think the best way to do this would be. Below is a list of players who could conceivably be available some time this year that aren’t expiring. Who would you choose to pursue?

    The rules are that the asterisks are indicative of what you’d have to give up in a trade. No asterisks mean all the player would cost is expiring contracts. The dollar amount is their salary this year. You have to bring in at least 16 million, more if you trade our prospects. Remember, Deron Williams and Chris Paul become free agents in 2yrs.

    Gallinari, Fields = **
    Mozgov, Randolph = *
    Douglas AND Walker = *

    I realize this is not really realistic, but I’m curious to see who people value most. Assume that you get change, so if you trade gallo and fields for Paul, you get one * as change, which you can use in another trade.

    Stephen Jackson 3 yrs, $8.5*
    Diaw 2 yrs, $9
    Wallace 3yrs, $9.8***
    Deng 4 yrs, $11.3*
    Brewer 3yrs, $4.8*
    Varejao 4yrs, $7**
    Mo Williams 3yrs, $9.3*
    Jamison 2yrs, $13.4
    Nene 2yrs, $11.3**
    Harrington 5 yrs, $5.8
    Chris Anderson 4yrs, $4*
    Balkman 3 yrs, $4.5
    Gordon 4yrs, $10.8
    Villanueva 4yrs, $7
    Biedrins 4yrs $9
    Scola 5yrs $7.8 ***
    Martin 3yrs, $11.1 ***
    Brad Miller 3yrs $4.4
    Baron Davis 3yrs, $13.1
    Kaman 2yrs, $11.3*
    Cook 2yrs $1.1
    Mayo 2yrs $4.5*
    Harris 3yrs, $8.9
    Paul, 2-3yrs, $15***
    Okafor 4yrs, $11.5*
    Ariza 4yrs, $6.3
    Jack 3yrs, $4.9
    Redick 3yrs, $7.5*
    Gortat 4yrs, $6.3*
    Brand 3yrs $16
    Iguodala 4yrs $12.3*
    Louis Williams 3yrs $5.4*
    Nash 2yrs $10.3**
    Childress 5yrs $6.5*
    Roy 4yrs $13.5
    Camby 2yrs $8.5*
    Andre Miller 2yrs (team option) $7.5*
    Fernandez 2yrs $1.2*
    Udrih 3yrs $6.7
    Garcia 3yrs $5.5
    Jason Thompson $2yrs $2.2*
    Calderon 2yrs $9*
    Barbosa 2yrs $7.1*
    Arenas 4yrs $17.7
    Hinrich 2yrs $9

  113. hoolahoop

    I don’t know how much Camby got in his tank, but that could be an interesting fit. We’d have a west coast style offense.

  114. Brian Cronin

    Celtics win again. It’s hilarious how the Knicks could win seven games in a row and not gain a single game in the standings.

  115. Robert Silverman

    Joel Pryzbilla might be gettable on the cheap. A Bill Walker/Mozgov/Mason combo might do it.

  116. Spree8nyk8

    Brian Cronin: Celtics win again. It’s hilarious how the Knicks could win seven games in a row and not gain a single game in the standings.  

    Yeah even more sickening considering they haven’t gained a single game going 11/12 :(

  117. KnickfaninNJ

    Mike Kurylo: @128 – how would you account for free throws?  

    Mike,

    I’m sorry this reply is so late. I was to get to the computer for a while.

    Free throws have a different probability of success and there is no defender so I think you have to take them out of the scoring totals. Two pointers and three pointers also have different distributions, because (they have different probabilities of success). In theory, only the distribution of one kind of shot is binomial. The total scoring of a player is the sum of this two point, three point and free throw shooting and this may not be binomially distributed. I think two pointers and three pointers can probably be combined, but you will have to figure out how many total points the player could have scored from taking 2pt shots * 2 plus 3pt shots *3.

    Separating out free throws I think is more realistic anyway. They aren’t defended and the shooter has plenty of time.

  118. protectthepaint

    two Words: Wrist Injury.
    Gallinari had a mad bout with his injured wrist.
    I love where this standard deviation measurement tool of consistency is heading, and hope to see it replace plus/minus, the worthless stat that got Brian Cardinal paid, but it must also scientifically account for the time a player is returning form injury, or toss those games out of the equation, as we all know that a return from any injury will skew all data that is being measured in this article.
    in other words, maybe it is the fact that Gallinari is dating a Gossip Girl actress?

Comments are closed.