Up at BlogaBull: The Bulls Strange Inability to Defensive Rebound

Through 4 games, Tom Thibodeau’s defense is outside of the league’s top 10 in points allowed per 100 possessions and that is, obviously, very surprising. Granted, Chicago is 11th in defensive efficiency, so they are barely outside of the top 10, but given that under Thibodeau the Bulls have finished 1st, 1st, 5th, and 2nd in overall defense in the last 4 seasons, it is a bit surprising to see them anywhere but the very tip top of the league’s defensive rankings, even at this early juncture.

More at: http://www.blogabull.com/2014/11/6/7162997/the-bulls-strange-inability-to-defensive-rebound

Up at BBALLBREAKDOWN: Pau Gasol Beat Up the Knicks, But Might Be Slowing Down

The subject of this article might not seem the most timely, given that Pau Gasol just beat the stuffing out of the New York Knicks in the Chicago Bulls’s season opener, recording an efficient 21 point, 11 rebound night in just 29 minutes of play. In fact, this might all end up being a bit of nagging worry that doesn’t amount to much of anything at all. But although the Knicks’ interior defense is incredibly bad, there were some not so great signs for Pau’s prognosis in that game, ones that were also evident during the pre-season as well.

More at: http://bballbreakdown.com/2014/10/31/pau-gasol-beat-up-the-knicks-but-might-be-slowing-down/

Up at Nylon Calculus: More 2014–2015 NBA Season Win Totals

Like Andrew Johnson, I did some season projectin’ for the season starting tonight. The basic method was to take ESPN’s Real Plus-Minus (RPM) numbers and run them through a simple aging curve and then to project out the season minutes.

Read more here: http://nyloncalculus.com/2014/10/28/more-2014-2015-nba-season-win-projections/

Never Hate a Player: On Doug McDermott

[Ed. note: This piece originally ran on July 23, 2014 on a site where all traces of my former presence, along with that of all of the other members of the team, has been wiped away. I didn’t want it to disappear from the internet, so I’m reposting it here, where I have control over what happens to it.]

I am dumber than I like to think that I am. From time to time, it’s important to remind myself of this. This article is me doing just that.

It is often easy when writing about sports to fancy yourself just as qualified as anyone else, including say general managers and talent evaluators, to say whether a player might be bad, average, good, or great. This is, for the most part, harmless, and in some cases, it is possible to outperform the average or bad general manager, if you’re pretty good at scouting talent (like all of the armchair GMs among us would like to believe we are). But this attitude can morph into the worst kind of narcissistic hubris, and well, make you (and when I say “you,” understand that I mean “me”) come off as a bit of an ass. Now, front office types do not have a corner on the market for knowledge about the game, and this should not read like an argument or claim that they do. What I’m wrestling with is that none of us has a perfect understanding of the game, and we’re all learning new things every day we’re lucky enough to watch the world’s other beautiful game. It’s easy to be blind to that. It’s so easy, especially in the super overreactionizer that is Twitter, to have a strong reaction to something and spew it out without really stepping back to question your own assumptions.

It’s quite hard to challenge those assumptions regularly; it’s so much easier to allow them to calcify and constrict my thinking. It requires no effort at all to fall back on my default setting: I know this and that and this too about basketball. But, then, every once in a while, I get to take a moment and breathe and really think, and what I always come back to is this quote from Socrates:

I know one thing: that I know nothing.

It is probably the most important sentence I’ve ever heard about the nature of knowledge. I should always be striving to learn and understand better. Whenever I decide I know something, I’m lost, because I’ve stopped learning.

This is all a bit abstract, so let me be more concrete. When the Chicago Bulls, my favorite basketball team, traded a bunch of assets to acquire Creighton star Doug McDermott, I basically had a Twitter meltdown. McDermott is probably one of the best shooters in the world on a team that was terrible at scoring last year, but all of my favorite statistic-based models cast lots of doubts about whether Doug could play at the NBA level. He didn’t pass very much, he basically never got blocks or steals, and his rebounding was merely decent. McDermott’s low block and steal rates and just okay rebounding made me worry about his athleticism, as those three stats have traditionally been pretty reliable at predicting which players will have the athleticism to hang in the league and those who won’t. The concerns about McDermott’s athleticism matched my own eye test concerns about him. So I decided I knew who Doug McDermott was as an NBA player before ever seeing him play in the NBA. I ignored people, like my friend Ricky O’Donnell of BlogaBull, who pointed out Creighton’s ultra-conservative defensive scheme as a reason for his low defensive counting stats. I scoffed off people who told me he was a good body position defender. I disregarded the common-sense idea that when you’re scoring as much as McDermott did and moving all over the court non-stop with literally five defensive players all aimed at stopping you, maybe defense takes a bit of a backseat. I stopped thinking and started ranting. I was lost.

The statistical models I love have had a pretty good success rate, especially when compared with the average general manager in the NBA, but they’re not infallible. There are plenty of misses, just as there are with any attempt to predict the future of very young men making the transition to a totally new atmosphere and level of competition. There are simply too many things we can’t know at the time of the draft which effect how well a player will do at the next level. How they played in college or internationally and their resulting counting stats is a big piece of that puzzle, but it is only part of it. I ignored that, too.

I watched Doug McDermott in summer league, and many of my concerns still linger. He probably won’t score as prolifically as he did in college and his lateral quickness is not great. But McDermott is so, so smart. He makes extremely quick decisions with the ball, which is a still undervalued skill and its value is multiplied exponentially by the threat he represents as a shooter and floor spacer, especially given his lightning quick shooting release. He’s going to bend defenses, just by virtue of these two skills. He’s also a much better passer and decision-maker than his low college assist totals would suggest. Additionally, McDermott is, as I was told, a solid body position defender, who will mostly funnel players towards his help defense- perhaps not coincidentally, the Bulls have two of the league’s very best help big men in Joakim Noah and Taj Gibson. Yes, McDermott will give up blow-bys to more athletic players and yes, that will be frustrating when it happens, but it won’t hurt as much as it might on another team because of those two big mobile guys behind him. Context matters very much in basketball, and well, maybe on draft night and after I didn’t think enough about the context in which McDermott will be operating. The defensive warts can be more easily hidden in Chicago than nearly anywhere else and his skill-set is a much needed one on any team, but especially for these Bulls.

Film Crit Hulk is one of my absolute favorite writers, and he has a tremendous piece which centers around a bit of advice given to him by the famed director, Quentin Tarantino. During a conversation in which a younger, perhaps less thoughtful Hulk ranted against a movie he “hated,” Tarantino told him, “Never, under any circumstances, hate a movie. It won’t help you and it’s a waste of time.” Tarantino went on to more fully explain that there is value and things to learn and enjoyment to be found even in the bombs or, for our basketball-watching purposes, busts. Tarantino finished his advice by saying of movies, “They’re gifts. Every f*cking one of ‘em.”

I am much less sure than I was about what sort of player Doug McDermott might be than I was on draft night. Part of that is a function of his summer league play, but a bigger part of it is me allowing myself to embrace that I don’t know nearly as much as I sometimes think and act like I do. What I do know is that regardless of whether he turns out to be a “bomb” or another “hit” for the Bulls front office, I’ll learn from watching him play. I’ll learn from seeing his struggles or successes and the how and why behind them. I’ll be entertained, as I always am, by the process. Doug McDermott is a gift, just like every player which I have the privilege to watch and root on.

Gulp. Stan Van Gundy to Detroit.

SVG
Pictured: A brilliant basketball man.

Stan Van Gundy was, to my mind, the best coach available on the market this offseason, and for the last few offseasons frankly. That stance, of course, comes with the caveat that there are many assistant coaches and coaches at different levels of basketball with whom I am unfamiliar and therefore, I exclude from any consideration in the “best coach available” discussion. There may be some diamonds in the rough who need a shot to show how good they can be. But I don’t know enough about the coaching development chain to say much about those guys. Still, SVG has proven himself as a great coach in the NBA.

Van Gundy ought never to have lost his job in the first place, but it’s a superstar league and the Magic wanted to appease Dwight Howard, before realizing how futile a game that really was. SVG has a winning percentage of 64.1% in the regular season over 579 games and he is 9 games over .500 in the post-season (48–39). Winning percentage alone, however, is not the only mark of a great coach. If you need any proof of that, check out Avery Johnson’s career after his Dallas turn, Vinny Del Negro’s run with the Clippers last season, or Scotty Brooks’ campaigns the last few seasons. Talent can overcome imperfect or even downright incompetent coaching, at times. Van Gundy isn’t that sort of coach. He’s exceedingly competent, curious, and supremely adaptable. So the following should hardly be surprising:

Stan Van took two separate and totally different style squads through deep playoff runs, ending in the Conference Finals and NBA Finals, respectively. In Miami, he paired Dwyane Wade’s pell-mell drives to the hoop with a still dominant Shaq’s low-post scoring to reach a top 5 offense and a 7 game conference finals loss to Detroit. It was a distinctly traditional offensive setup, and Van Gundy made it work incredibly well.

Then, from 2008–2011, he helped Dwight Howard reach his peak, which he has still not surpassed or even really approached, with a 4-out spread pick and roll attack which both incorporated the biggest lessons of the modern basketball analytics movement (the importance of spacing, along with the value of the three point shots and dunks) and best catered to the talents of his team. It’s also worth noting that no coach since SVG has been able to get Dwight to buy in to the pick and roll game as thoroughly as he did in Orlando. In 2009, the Magic overachieved in the playoffs, knocking off LeBron’s Cavaliers juggernaut — +8.68 SRS, well ahead of the second best team, the eventual champion Los Angeles Lakers — in the Eastern Conference Finals, only to get beat by that Lakers team in 5 games. It would be easy to be disappointed by the Magic’s performance in that Finals, but really, they shouldn’t have even been in that series at all. Orlando had already overachieved by reaching the Finals. Of course, that’s not the attitude a coach or his team should have in that series, but for fans and media analyzing Van Gundy’s resume, it’s important to note that important context.

So, in short: Detroit has nabbed a great coach. Van Gundy, after being burned by both his former Florida employers and unjustly let go, also managed to negotiate control over Basketball Operations for the Pistons. I don’t know how SVG will do on the management side of things, as there’s no history to go on. Given his understanding of how teams fit together and how to maximize the talents he’s provided, though, I suspect he’ll also do a heck of a job there, as well. He would be hard pressed to do worse than Joe Dumars the last few years.

I see most things in the NBA through the lens of how they will effect my beloved Chicago Bulls. Stan Van Gundy entering the Central Division scares the hell out of me. I’d imagine most fans of other Central Division teams feel likewise, which is as good an indication as any that Detroit has done very well for themselves here. Now it’s time to fire up the trade machine, because there’s almost no way SVG is keeping this trainwreck together next season.

UARPM100 for 13–14 Updated to Reflect Team Strength

Last post, I mentioned that Kevin Durant was the UARPM100 MVP, and I gave a top 10 list of players in Wins Above Replacement as well. After looking through the numbers, something that occurred to me was that the number of total wins under those numbers didn’t sum up to team level wins. That was primarily an effect of including raw per minute plus-minus numbers as part of the UARPM formulation. Basically, good teams had too many wins, and poor teams had too few wins. So I decided to correct that. I adjusted the UARPM100 numbers using a per minute adjustment for each player on the team so that total team plus-minus was equal to team SRS (basically point differential adjusted for strength of schedule) via Basketball-Reference. The final numbers are posted on the UARPM100 page.

The top 10 is basically the same, with Carmelo Anthony jumping into the 8th spot, and DeAndre Jordan sliding to 10th. The total number of wins are reduced across the board, and they are no longer set to above replacement, because I decided it’d be more interesting to just have total wins contributed. You can easily turn wins into Wins per 48 minutes by dividing by minutes played and multiplying by 48. Durant and LeBron were nearly exactly the same in per possession impact by UARPM100, with Durant’s heavier minutes load giving him the edge in wins. Chris Paul also remains the best per-possession player in the league, even after the team adjustment.

Here’s the updated top 20:

RankPlayersMINUARPM100Wins1Kevin Durant31226.720.02LeBron James29026.718.63Kevin Love27966.417.64Stephen Curry28466.117.35Blake Griffin28635.215.86Chris Paul21717.214.77Joakim Noah28204.614.38Carmelo Anthony29824.114.39James Harden27774.514.110DeAndre Jordan28704.213.911Paul George28984.113.912DeMarcus Cousins22985.913.713John Wall29803.513.114LaMarcus Aldridge24994.713.015Kyle Lowry28623.512.616Dwight Howard23964.612.217Goran Dragic26683.611.918Anthony Davis23584.411.719Al Jefferson25533.811.720Serge Ibaka26663.411.6
All in all, this seems like a pretty credible list. For what it’s worth, Rookie of the Year award winner, Michael Carter-Williams produced 6.1 wins under UARPM100, well ahead of runner-up Victor Oladipo who clocked in at 4.7 wins. The voters appear to be doing a pretty good job.

Kevin Durant is the UARPM100 MVP

I recently updated my UARPM100 numbers to reflect the end of regular season statistics. You can see the final numbers here. After all the games were played, Kevin Durant was the UARPM100 MVP in my version of Wins Above Replacement (WAR). Durantula provided roughly 21.8 wins over what we’d expect from a replacement level player taking over his minutes, while LeBron James came in as a close second providing 21.5 WAR. LeBron was a bit more productive per possession, by UARPM100, than Durant (+7.6 to +7.1), but Durant played more minutes, which ultimately made the difference.

Interestingly, Chris Paul was tops in per possession productivity clocking in at +8.1 points per 100 possessions better than average. CP3 missed a number of games with injury, which knocked him out of MVP consideration. Paul was still able to contribute 16.8 WAR despite only playing 61 games, which is pretty amazing.

The top 10 in WAR via UARPM100 were:

1. Kevin Durant, 21.8 WAR

2. LeBron, 21.5 WAR

3. Kevin Love, 19.4 WAR

4. Stephen Curry, 18.9 WAR

5. Blake Griffin, 18.2 WAR

6. Chris Paul, 16.8 WAR

7. Joakim Noah, 16.1 WAR

8. DeAndre Jordan, 16.0 WAR

9. James Harden, 15.7 WAR

10. Carmelo Anthony, 15.1 WAR

(Way to waste a really great season from Carmelo, Knicks.)

Also notable: Goran Dragic, who recently received the NBA’s Most Improved Player award, finished 20th overall in WAR. In 2012–13, Dragic put up a +1.3 UARPM100, while this season he put up a +3.8, along with the aforementioned 20th place finish in WAR. A pretty big leap, and one of the more difficult things a player can do- go from being the pretty good player he’s been his whole career- to jumping into the top echelon.

Prior Informed UARPM100 Through Roughly 30 Games

In my last post, I mentioned that I would, from time to time, produce UARPM100 numbers that were prior-informed by xRAPM numbers from Stats for the NBA. Today, after roughly 30 games played for each team, I’ve gone ahead and produced those numbers. Enjoy!

Another Update to UAR: Introducing UARPM100

Last time out I explained, in detail, how I calculate Usage Adjusted Rating (a usage adjusted version of Alternate Win Score). I liked the results, but I thought that they could be better. In order to try to better value defense, I decided to try to include a weight to factor in minutes per game played. I made this decision based on the idea that coaches, generally, won’t play someone a lot of minutes if he’s got shaky counting stats- which basic UAR covers- unless he’s providing other value. So I added a factor that gives a slight boost to players who play 20 minutes or more per game and gives a slight negative to players who play under 20 minutes a game.

In addition, I took the UAR with the minutes per game adjustment (70%) and blended it with non-adjusted +/- per pace adjusted 48 minutes (20)% and added a zero-weight to regress it to the mean (10%), as this was the blend that best correlated with xRAPM. Then I made the metric 100 possessions, instead of per 48 minutes pace adjusted. I call this new metric UARPM100, which is a bit of a mouthful, but it conveys the information contained within the metric, so I’m sticking with it.

I ran a correlation of UARPM100 over past years against xRAPM from Jeremias Englemann at stats-for-the-nba.appspot.com. The r-squared for UARPM100 against xRAPM was roughly .67. The r-squared becomes much, much stronger if prior year xRAPM is blended with UARPM100. The r-squared for blended prior year xRAPM and UARPM100 is roughly .82 with in-year xRAPM, which is obviously very strong. Given my belief that xRAPM is probably the best one-number metric in the public domain, I feel pretty good about UARPM100’s results. Here are the results for UARPM100 through the December 16, 2013 games (minimum 120 minutes played):

Going forward, I will be updating UARPM100 as close to daily as possible. Periodically, I will also post UARPM100 that’s prior informed by 2012–13 xRAPM. Hope you enjoy!

Usage Adjusted Rating, Further Explained

Usage Adjusted Rating, as I discussed previously, has Alternate Win Score (AWS) as its base. Alternate Win Score is a simple per minute measure of performance, which has proven to be the best linear weights metric for prediction across high continuity and low continuity contexts. High continuity contexts are situations where a team is the largely the same as it had been when the players compiled the statistics being used to make predictions. Low continuity contexts are the opposite. AWS, as Neil Paine has demonstrated, is the best linear weights metric for prediction when dealing with both of those situations. So how is Alternate Win Score defined?

AWS equals Points+0.7*(Offensive Rebounds)+0.3*(Defensive Rebounds)+Steals+0.5*(Blocks)+0.5*(Assists)-0.7*(FG missed)-(FG made)-0.35*(Free Throws Missed)-0.5*(Free Throws Made)-Turnovers -0.5*(Fouls Committed) all divided by Minutes Played.

I wanted to make some tweaks to this basic formula. Namely, I wanted to include a usage-efficiency tradeoff. As I mentioned in the previous post, APBRmetrics forum poster v-zero provided a way to do that. I incorporated his math into the formula for AWS and after some tweaking, I arrived at UAR.

About that tweaking. Some people have expressed interest in knowing exactly how I arrived at the numbers I came up with. So here goes. I broke AWS into two separate figures. The scoring (and offensive turnover) portion and the Non-Scoring aspect. The Non-Scoring portion of UAR simply is equal to .7*OREB+.3*DREB+Steals+.5*Blocks+.5*Assists-.5*Fouls Committed per pace adjusted 48 minutes.

Then I moved on to the Scoring portion of UAR, which includes turnovers because turnovers use a possession just the same as a shot attempt or free throw attempts, except turnovers obviously always result in 0 points. I calculated the league average for points per possession (PPP), using the simple formula for possessions (FGA+.44*FTA+TOV), and similarly calculated the league average for possessions per 48 minutes (USGper48), again using the simple possession definition. I then used the coefficients v-zero provided to create what I call average ScoreRating, which is simply 5*(PPP)+.076*(USGper48). For this season, thus far, the league average for that number has been roughly 6.2. Next I calculated the Score Rating for every player in the league and subtracted out the league average rate, so that if you’re an average scorer you break-even in Score Rating, if you’re above average you contribute a positive value through your combined scoring volume and efficiency whereas if you’re below average, you detract value from your team through your inability to score. I also had to multiply Score Rating by a coefficient in order to properly value scoring in UAR relative to the NonScoring parts of UAR. The Scoring Rating needed to be worth roughly 2.7 times the Non-Scoring Rating, based on some math resulting from the Four Factor weights discovered by Evan Zamir here. In order to get the scale right, the coefficient turned out to be roughly 2.4. This owed to the league average for Score Rating being 6.2 and the league average for Non-Score Rating being about 5.5. Then I set total league average UAR to 0.

These numbers change year over year but they are pretty consistently in this range. I then added the Scoring and Non-Scoring parts together to get UAR. The equation for this year basically looks like this:

UAR = (2.4*(5*(PPP)+.076(USGper48))+ (7*OREB+.3*DREB+Steals+.5*Blocks+.5*Assists-.5*Fouls Committed per pace adjusted 48 minutes)-((lg avg Score Rating)+(lg avg non-score Rating))

The numbers, as I said, vary year over year depending on what the average numbers league wide are.