A Vision for the Future

Since starting to write about basketball on the internet a few years ago, I’ve been very fortunate to have some great opportunities to write for a number of websites with a very wide reach. As a result of those opportunities, I was able to intern for the Dallas Mavericks last year in their analytics department and I’ve been able to launch the official podcast for the website on the bleeding edge of NBA analytics and information in the public domain, Nylon Calculus.

Over and over again, however, one thing has left me frustrated about the sports media landscape: It is very difficult for great writers to be compensated adequately for the amount of time they put into producing amazing sports content. Many of the websites that provide great platforms to learn and build a voice don’t provide much in the way of monetary incentive for the amount of time required to produce top-notch content. The reason for this is that there is simply a massive amount of writing labor supply. Simply put, if you decide to drop out of a content-farm because it’s not making sense for you in terms of ROI on your time, there will be someone glad to fill your place for no pay or close to no pay. Everyone is looking to make their big break. On the individual level, this makes total sense, but collectively it is a big problem. I don’t think this is sustainable in the long term.

There are really two models for making money on content. There are ad-supported models and there are consumer-supported models. I am biased in favor of consumer-supported models, because I find being tracked by advertisers wherever I go online (and in the real world, actually) to be kind of gross. In addition, ad-supported websites seem, to me, to be in danger of being eaten by  the Facebooks and Googles of the world. When an advertiser can go to Facebook and get every detail about a reader’s preferences, what would be  the point of general purpose branded advertising? It’s no coincidence that Facebook has begun partnering with news organizations to host articles directly on Facebook. I suspect Facebook is going to take a greater and greater share of the profits from the advertising dollars spent backing the production of content. As a result, the number of websites actually producing content and the amount of writers who can be paid through the ad-supported model will dwindle.

So, if that’s the case, where does it leave an independent writer on the internet who wants to make money writing about, well, anything? Increasingly, I think it’s going to mean that consumers are going to need to support the people who make the things they like directly, or if they are unwilling, the things they like to consume will go away.

A quick digression: Recently, I became a big fan of Ben Thompson of Stratechery and the Exponent podcast. My Twitter pal, Kostya Medvedovsky (follow him, by the way), recommended the podcast to me and I have listened to over 50 episodes now. Ben makes his living writing a daily paid newsletter on tech business strategy. This has become a big inspiration to me.

So, taking the inspiration from Ben and reading the tea leaves about what I think is the future of content creation, I have come up with a vision for a new business model for NBA content. This may not work. That’s fine. But I feel I have to try.

Here’s the pitch: I want to create a network of paid newsletters centered around individual NBA teams. The idea would be to produce 8-10 articles a month (potentially more if the subscriber numbers get to a point to allow writers to do the writing full-time), for a subscription fee of somewhere between $8-10 (so a buck an article, basically). Because readers will be paying for these articles, the goal would for them to be analytical and provide value above and beyond what you can find for free on the internet. I am going to be putting together the sign-ups for the first of these in  the coming days. Please let me know on Twitter or via e-mail (nbacouchside@gmail.com) if this is something you’d be interested in. The first team is going to be the Bulls, because that’s who I know how to write about best. I want to provide a proof of concept of this idea before I bring on anyone else to start another team newsletter up.

Thanks for reading.

Kyrie Irving is Overrated


Kyrie Irving is a supremely talented basketball player. Anyone with functioning eyeballs and any interest in the game can see  that. Irving was rated the second best prospect in his high school draft class. He was the consensus and obvious number one overall selection in the 2011 draft. He deserved those accolades. But with his Cavaliers down 2-0 to the historically world-beating Golden State Warriors and Irving about to wrap up his fifth NBA season, it’s worth wondering, just how good is Kyrie Irving?

It’s a question many are now asking.  It lead ESPN radio personality Bomani Jones to wonder on his excellent radio program, The Right Time (the discussion begins at the 30 minute mark), whether Reggie Jackson (Reggie Jackson!) might be a better option for the Cavs than Irving.

Deadspin’s Kevin Draper posed this exact question in a quick post yesterday, concluding:

I don’t want to overreact to a two-game sample, but Irving’s defense has been awful since he entered the league five seasons ago, and furthermore, no matter what advanced metric you look at, he grades out as good-but-not great. Despite this, he has a max contract, a signature shoe deal, three All-Star games, an appearance on the 2014 FIBA Basketball World Cup team, and all the other trappings associated with being one of the league’s superstars.

So, with two more days before Game 3, here are the big questions: Are the Cavs better without Kyrie Irving than with him, and is there anything they’re willing to do about it?

To all of this I say, welcome aboard, America to the “Kyrie is overrated” bandwagon. Irving’s standard issue box-score numbers have always seemed to inflate his impact and whenever you look at how Irving actually impacts the bottom line of points scored versus points surrendered, you can see just where he is lacking. Irving puts up great offensive stats. He’s hovered around a 20 PER (career: 20.9) every year for the last five years. His offensive Box Plus-Minus (OBPM) has hovered between +4 and +5.

Season    Age  Tm  Lg Pos   G    MP  PER  TS% 3PAr  FTr ORB% DRB% TRB% AST% STL% BLK% TOV% USG%   OWS DWS   WS WS/48  OBPM DBPM BPM VORP
2011-12    19 CLE NBA  PG  51  1558 21.4 .566 .245 .262  3.1 11.2  7.0 36.5  1.8  1.0 16.1 28.7   3.4 0.6  4.1  .125   4.9 -1.7 3.3  2.1
2012-13    20 CLE NBA  PG  59  2048 21.4 .553 .261 .271  1.8 10.8  6.1 32.7  2.3  0.8 13.8 30.2   4.2 1.1  5.3  .125   4.9 -1.6 3.3  2.8
2013-14    21 CLE NBA  PG  71  2496 20.1 .533 .278 .273  2.3  9.5  5.8 31.6  2.2  0.8 12.1 28.2   4.6 2.1  6.7  .128   4.1 -0.9 3.2  3.3
2014-15    22 CLE NBA  PG  75  2730 21.5 .583 .306 .296  2.3  7.5  5.0 25.0  2.2  0.6 11.8 26.2   8.4 2.0 10.4  .183   4.7 -1.4 3.3  3.7
2015-16    23 CLE NBA  PG  53  1667 19.9 .540 .298 .217  3.0  7.6  5.3 26.6  1.7  0.9 11.4 29.5   3.2 1.7  5.0  .143   3.0 -1.4 1.6  1.5
Career            NBA     309 10499 20.9 .555 .280 .267  2.4  9.2  5.8 30.0  2.1  0.8 12.8 28.3  23.8 7.6 31.4  .144   4.4 -1.3 3.0 13.3

Data courtesy: Basketball-Reference.com

It’s worth noting here that Irving had his worst season of his career this season and he’s never really shown much improvement by the box-score numbers. Still, if those offensive numbers were representative of his actual impact, then even without improvement, he’d be a really good and valuable player worthy of his maximum contract. Unfortunately, there is a lot of evidence to indicate that’s not the case.

Looking at Irving’s career Real Plus-Minus numbers1 we see a picture of a player who is, by and large, a slightly above average impact player.


It’s important to consider here that a significant portion of Real Plus-Minus comes from Irving’s box-score numbers. The box-score prior – which is fairly similar to BPM – conveys a fairly significant chunk of the information in Real Plus-Minus. So what if we remove the box-score influence? What does regularized adjusted-plus minus (RAPM)2 say about Kyrie for his career?


Kyrie again hovers right around the average point of 0. That’s hardly the impact we’d expect from a guy who is touted as a superstar level talent. For comparison’s sake, Chris Paul, in his age 30 season, just put up a RAPM of +4.5 per 100 possessions with splits of +2.9 per 100 on offense and +1.6 per 100 on defense.

Irving has incredible skills handling the ball, snaking through tight spaces, and finishing through contact. He’s a very good shooter. When you watch him, unless you’re paying close attention, it’s very easy to miss why he’s not a superstar level player when it comes to impact. There are a few reasons that Irving is not as great or impactful as his skills suggest he should be. He dribbles the air out of the ball, stagnating the Cavaliers’ offense, and taking more difficult shots than he could perhaps get for his teammates or even himself if he swung the ball and got open off-ball more often. The next time you’re watching Kyrie pull this act, watch how frustrated LeBron looks watching him do it. Then remember that LeBron is one of the smartest basketball minds the world has ever seen. Kyrie also doesn’t have particularly great court vision. For all his fancy dribbling and ability to slither into the lane, he doesn’t use the defensive panic he causes to create a great look for a teammate nearly often enough. Most damaging, Irving is just absolutely lost on defense, almost all of the time. He can occasionally hold up one on one in an isolation situation, but involve him in any sort of screening action both on and off the ball (which the Warriors have done over and over), and he is toast. Add all those things up and most of the positives he brings get neutralized.  

Real Plus-Minus and other adjusted plus-minus metrics aren’t everything, but they can tell us quite a bit, particularly about how a player is succeeding in a given role. Irving has been in two wildly different contexts in his career, first as the lead guard on a relatively talent-less team that went nowhere and now as the second option on a stacked team with one of the two or three greatest players of all time. In neither situation has he particularly thrived. In fact, looking over his career, he’s been roughly average for almost all of it, and he has shown few signs of improvement. At this point, 5 years into his career with Irving wilting on the league’s biggest stage, it’s fair to question whether he ever will.

  1.  2011-12 and 2012-13 season numbers come from Jerry Englemann’s old website where they were called xRAPM, Real Plus-Minus’s precursor
  2. RAPM numbers come from this thread, again thanks to Jerry Englemann

Up at BlogaBull: Let Jimmy Be Harden

Since Derrick Rose has gone out with his newest knee injury, the Bulls have returned to last year’s Joakim Noah centric offense. They should fight against that impulse to rely on the familiar, because this year Jimmy Butler has shown he can handle a much bigger role. In fact, even when Rose is back, it would serve the Bulls well to allow Jimmy to handle more and more of the offense.


Did We Expect Too Much from The Chicago Bulls?

[Ed. note: I pitched this piece a little while back and it didn’t get picked up where I pitched it, so I’m posting it here. Chicago has picked it up some of late, so it’s not exactly timely, but I wrote it and wanted it to have a place on the internet.]

The Chicago Bulls have been one of the NBA’s weirdest teams this year. They’ve been labeled a contender since they signed Pau Gasol in the offseason. While Pau’s put up big numbers, and Derrick Rose has played 41 out of 52 games, the Bulls don’t look anything like a contender on a night to night basis.

Chicago’s defense, typically the backbone of the team, has fallen off significantly from its usual spot amongst the top 3 to 5 teams in the league to something much more average. The Bulls’ offense, on the other hand, a putrid eyesore last season, is back to respectability. In total, the 2014–15 Bulls have been a pretty good, not great team. It’s worth wondering why anything more was expected.

In some ways, it’s easy to see the logic behind the anointing of the Bulls as sure bets to make the Eastern Conference Finals. No one expected the Atlanta Hawks to become what they are and the Cleveland Cavaliers’ defense was a huge question mark. Chicago won 48 games last season and upgraded from Carlos Boozer to Pau Gasol and from D.J. Augustin to Derrick Rose. Predicting a significant boost in the win totals from those two seeming upgrades didn’t seem like much of a leap.

The truth of this season, though, has been that Rose has hardly been an upgrade over the performance Augustin provided and, in some ways, has actually been a downgrade. Part of that is how exceptionally well Augustin played last season in Chicago and part of it is the reality of Rose returning from nearly two entire years having played almost zero basketball.

Returning from one major knee injury is a difficult enough task, mentally and physically. Rose is returning from two. As a result, Rose is finishing at the basket worse than ever. Rose still helps the Bulls immensely when he charges into the paint over launching a long jumper, but his forays into the paint are just coming with less frequency than they once did. This shouldn’t be that shocking, really. To have expected Rose to return after being away from the hardwood for so long and be anywhere close to his former self was always absurdly optimistic. That’s not to say he can’t get ever back to that level, it’s just that it’s going to take time for him to get there and may require a change in how he plays. We might not see the best of Derrick Rose this season, and really, that makes total sense.

Gasol has absolutely been an upgrade over Carlos Boozer, that much can not be disputed. But Gasol does present many of the same problems that Boozer did. He is incredibly slow-footed on defense and his effort on that end is often demoralizingly poor. The Bulls defensive drop-off is not all Pau’s fault, but he is a big part of the problem. The other part, which Pau’s presence has exacerbated, is that Joakim Noah has not been himself this season. This, too, could have been predicted, were it not for the Rose-colored glasses with which everyone seemingly viewed this year’s Bulls.

Noah, after being an MVP candidate last season, was absolutely demolished in the playoffs by the Washington Wizards’ frontcourt. It was clear something was wrong with him, and as it turned out, he had to have offseason knee surgery. As a result, Noah, up until very recently, has seen his lateral movement significantly limited and his usual frenetic energy sapped. With Gasol manning the back-line center spot in the Bulls’ defense instead of Noah, last year’s Defensive Player of the Year has been forced to chase around power forwards on a bum wheel. Not exactly an ideal fit or situation.

Finally, a hugely under-discussed portion of the Bulls’ relatively disappointing season has been their lack of wing depth. Even with the problems that Rose’s rustiness and Noah’s balky knee have caused, the Bulls looked pretty close to the contender everyone expected up until January 1st.

What happened on January 1st? Mike Dunleavy Jr. jammed his ankle in a tilt against the Denver Nuggets. He’s been out ever since, as the injury has been the nagging sort. Without Dunleavy’s shooting in the starting lineup to space the floor around the driving games of Rose and Jimmy Butler and the post-ups of Pau Gasol, the Bulls have seen their offense stagnate. It’s been even worse on the other end of the floor. Dunleavy is a classic glue guy on defense, as his 6’9″ length and smart use of angles makes him tough to score over for most other wings in the league. The Bulls have missed him immensely.

Behind Dunleavy on the depth chart is Kirk Hinrich, a formerly solid two-way player who has seen his once stout defense slip markedly this season and who has been an offensive non-entity for a few seasons now. Second year player Tony Snell provides a decent approximation of the length and shooting of Dunleavy, but without the veteran savvy or, more problematically, the trust of Coach Tom Thibodeau. The Bulls have also tried rookie Nikola Mirotic on the wing, but he is really a stretch power forward more than anything, and doesn’t have the foot speed to defend most wings in the league. Doug McDermott, last year’s college Player of the Year, is also around as an option, but he was out until very recently with a knee injury of his own and appears to be firmly entrenched in Thibodeau’s dog-house, a not unusual phenomenon for rookies under the Bulls’ demanding and, at times, unforgiving head coach.

The Bulls’ lack of wing depth was a problem that was there to see for those that, knowing Thibodeau, could guess how reluctant he might be to use the young players who make up the majority of the Bulls’ wing rotation. If Dunleavy didn’t get hurt, this weakness of the Bulls’ roster might never have been exposed, but that’s life in the 82-game grind of an NBA season. Injuries invariably happen.

So are the Bulls a disappointment or did we simply expect too much? It’s a little of both. The Bulls had real weaknesses coming into this season that were glossed over due to the simple math of Rose plus Pau Gasol plus 48 wins last year equals contender. It’s also a long season, so even if the Bulls don’t take the East’s number one seed in the regular season, Rose and Noah could get right, with Dunleavy’s return on the horizon, and the Bulls could still come out of the Eastern Conference playoffs. It’s just been a more frustrating and difficult road so far than they or anyone else really expected.

Up at CBS Local Sports: Derrick Rose’s Diminished Vertical Explosiveness

It’s tough to disentangle how much of Rose’s reducing his forays to the paint and the rim is mental and how much of it is a drop in physical ability. In watching him night in and night out, he seems to be just as fast as ever, but he doesn’t appear to have the same vertical explosiveness he once did. Even if Rose never gets that burst off the ground back, he can still become an extremely effective player based on his speed, but he has to fundamentally alter his game and improve on aspects of his skillset that he never really needed so badly before. Maybe that’s what his struggles this season have been about: evolving and building a new skillset, a new path to effectiveness. That has to be the hope for Rose fans, because it looks like the version of Derrick Rose who can explode up and over opponents (hi, Goran Dragic) might be lost to us.


Up at BlogaBull: Derrick Rose Struggling, But Don’t Panic Yet

This season, Derrick Rose has not been his old MVP self. This is not a revelation, I’m not telling anyone noteworthy information by making that observation. Rose has shown flashes of brilliance here and there, but mostly he’s been okay to pretty bad depending on the night. His overall numbers and efficiency have all trended down over the course of the year. I’m still not quite ready to panic regarding Rose’s struggles, though. Some of this is based on positive things Rose has been able to do and part of it might be wishful extrapolation from a select few (some would say cherry-picked) numbers. Let’s get into it.


Up at BlogaBull: Signing Pau Gasol Was A Mistake

Pau Gasol was selected as an All Star starter by the fans. This is, honestly, a joke. Pau Gasol is the Bulls fourth best big man. Yes, fourth best. Behind Joakim Noah, Taj Gibson, and even rookie Nikola Mirotic. The idea that Gasol was somehow one of the best big men in the entire Eastern Conference is ridiculous. At first glance, it’s easy to see why Pau has received this recognition. Pau is averaging 18 points, 12 rebounds, nearly 3 assists, and over 2 blocks in his roughly 35 minutes per game. Those certainly seem like All Star numbers. However, it’s pretty likely that no one is putting up emptier stats than Pau Gasol is this season.



Building A (Hopefully) Better Simple Metric

It’s been a while since I’ve written anything in this space, but I’ve not forgotten it, and I’ve not stopped tinkering with some of the ideas I’ve tried to tackle here before. The idea I’d like to return to today is my enjoyment of simple, easy to calculate, transparent boxscore metrics.

In the past, I built off of an easy to calculate and understand linear weights metric (Alternate Win Score) to create Usage Adjusted Rating, which essentially tried to adjust AWS to credit heavier usage players for the greater degree of difficulty they generally encounter in getting points and remaining efficient. The results were pretty good and passed the laugh test. But calculating UAR and the subsequent variant blend with plus-minus (UARPM) that I developed was best done on season long numbers and well, there are much better one number metrics out there for analysis of season long data. Daniel Myer’s Box Plus-Minus (BPM) and ESPN’s Real Plus-Minus (RPM) being the best examples. So from here on, I’ll be retiring UARPM from the website.

But for broad strokes analysis of single games, the current best linear weights metric is probably Alternate Win Score. Some people like to use John Hollinger’s Game Score, since it’s readily available on Basketball Reference for every game. I wanted to improve upon AWS and Game Score and build a transparent, easy to calculate and understand metric to quickly analyze game to game performance.

To build my game score metric, I looked to Jerry Engelmann’s 14 year RAPM data set, since it is, to my mind, the best estimate of long run  +/- impact that’s in the public domain. After that, I ran a regression of the most basic boxscore stats (per 100 possessions) to get the coefficients or weights for my simple linear weights metric.


New Win Score

I tried to include personal fouls, but they were not statistically significant predictors of RAPM (+/-) at all. All of the other boxscore stats I picked were highly statistically significant with strong p-values. Then, I translated the coefficients so that they were weighted relative to points score (i.e., so that the coefficient or weight for points was equal to 1). The resulting weights for my simple game score metric are as follows:

PTs + .2*TRBs + 1.7* STLs + .535*BLKs + .5*ASTs – .9*FGA – .35*FTA – 1.4*TOV

If you want to translate this linear weights metric directly to a simple statistical plus-minus, you can just subtract the average performance league-wide from the player’s total. Per pace adjusted 36 minutes the average performance in the league currently is roughly 4.9. Here’s the top 25 in the league as of the games played January 16 per game, pace adjusted, with the per-minute average subtracted out, so as to make it roughly +/- impact per game:

Anthony Davis7.1
Stephen Curry5.9
Chris Paul5.7
James Harden5.0
Kevin Durant4.3
Jeff Teague3.9
Jimmy Butler3.8
LeBron James3.7
Kyle Lowry3.5
Damian Lillard3.3
Russell Westbrook3.1
DeAndre Jordan2.9
Tyson Chandler2.8
Brandan Wright2.6
Kawhi Leonard2.6
Kevin Martin2.6
John Wall2.6
Mike Conley2.5
Klay Thompson2.4
Paul Millsap2.3
Marc Gasol2.3
Hassan Whiteside2.3
Derrick Favors2.2
Kyrie Irving2.1
Kyle Korver2.1

Those results definitely pass the laugh test. Anyway, I like this as another tool in the tool kit. I even won over noted one number metric skeptic Seth Partnow to use the metric for some broad strokes performance analysis.

Good enough for me!

If any of you have a good idea for a name for this new Win / Game Score linear weights metric, let me know in the comments or on Twitter: @NBACouchside.