Why the NBPA Is Playing Against a Stacked Deck

I thought now might be a good time to look into why a league that appears to be entering a renaissance is having such a difficult time reaching a labor deal.

THE PROBLEM

Consider the economic reality of the players. The NBA is often called a league of stars, and for marketing purposes it is. However the stars aren’t the only one voting, and when the players vote to ratify a CBA, the minimum-salary player’s vote counts just as much as Carmelo Anthony’s. The average NBA player salary is $5.15 million, but that number is skewed by a few players at the top making considerably more than the average hoopster. To compensate we can look at the median wage, which better represents the general NBA player. The median salary in the NBA is $2.3 million. $2.3 million might sound like a lot, but there are more factors to consider:

1) The Average NBA Career Lasts Five Years: Since most NBA players forego the years wherein they’d most likely be building a “normal” career, their earnings after their NBA careers are likely severely diminished. In other words a normal person would work into their 60s, but a basketball player is lucky to earn money in his 30s, and highly unlikely to still be getting a paycheck in their 40s. Yes, stars can go on to sell sneakers or underwear. And a few will be suited coach or work in broadcasting, but for most players, this is not likely an option. If we look at this median player’s salary spread out over the average worker’s forty years of income, that $2.3 million salary turns into a $287,500/year. That’s great money for most people, but it’s not filthy rich. In essence the average NBA player is more a six-figure salaried employee than millionaire.

2) Missing Games For a Better Deal Is Still a Net-Loss For Current Players: Let’s say you’re halfway through your five year NBA career. If you hold strong, the Player’s Association may finagle an extra 3% in salary, but that’s chump change compared to the 20% of your overall earnings you lose by not playing for a season. Almost without a doubt you are sacrificing income. Maybe it’s for a good cause, but how much is that good cause worth to journeymen NBAers?

3) One Season Constitutes 20% of the Average Player’s Career Earnings: That’s a huge sacrifice. Compare that to a person in an auto-worker union, who even if he strikes for a year only gives up 2.5% of his lifetime income (assuming a forty year career). To put that figure into context, losing a season drops the average yearly income of a player from $287,500 to $230,000.

4) Taxes: Because he makes most of his money in a short period of time, Mr. Median NBA Player will find himself in a higher tax bracket than the person with a longer career but the same lifetime earnings. He will send Uncle Sam up to a million dollars more than his dentist neighbor. Now his real-world salary is down to $200k/year. Again, that’s a good living, but not one on which you should be buying fourteen luxury cars

5) Poor Money-Management: As the article linked above explains, NBA players as a group are bad at managing money. This subject has been well-chronicled, but due to its importance to the current negotiations, it bears repeating. No matter how well the union has tried to prepare its players, many of them (Eddy Curry, take a bow) still live paycheck to paycheck and end up in financial trouble when the checks stop coming.

All these factors combine to put the NBPA at a huge disadvantage in negotiations. It is in a bargaining position that no amount of Dwyane Wade outbursts or Billy Hunter All-Star Game “soliloquies” can overcome. The owners know that the more checks the players miss, the more the union’s stance will soften. Issues like a hard cap and reducing bad contracts are mostly just smokescreen so that the owners can appear as though this lockout is “for the good of the game,” so that they can keep the pressure on the players until they’ve gotten the best deal possible.

THE SOLUTION

I am dubious about the claim that the NBA is losing money (1, 2, 3), but even if it is, the current offer from the players should more than make up for those losses. Moreover, with the renewed interest in the NBA due to the Heat’s big three; our Knicks’ acquisition of two stars; the rising careers of marketable players like Blake Griffin, Kevin Durant and Derrick Rose; and, as Mike Kurylo explained, the increased popularity of basketball among younger people, revenue has nowhere to go but up.

The problem is the owners benefit by making the lockout longer and even canceling games, up until a point. For instance if the players are clinging to an extra 3%, and the owners cancel 15% of the games this season, that means the common NBA player has gained absolutely nothing. (Common player plays 5 years and 3% * 5 = 15%). The biggest stars and players who are likely to play multiple years will gain by such a hold out, but not those on the short end of the spectrum. Hence among the players there isn’t consensus, which weakens their position, since some would take a lesser deal than others.

The owners know this & therefore are operating from a hard stance. (Steve Nash Tweets: “Why are people saying the players are asking for more money?” and “The players are negotiating to take less money & let’s be clear that’s not going to lower ticket prices, it just lines the owners pockets.”) And at this stage the owners are all together on this issue, because their interests are all being met here.

But the owners can only push so far, because the players have one tactic they can use to strengthen their hand: decertify the union. In the NFL labor dispute the players disbanded the union, and after some legal wrangling both sides negotiated a deal in time to save the season. By breaking up the union, the players can sue the NBA using class action lawsuits & challenge everything including the draft, the salary cap, restricted free agency, etc. Obviously this is something that could hurt the owners severely, if they had to openly bid for rookies and free agents without any restrictions. (see: The Dollar Auction for why owners would inevitably pay more than a player’s worth.)

Billy Hunter may not like what decertification suggests about his negotiating abilities. However, the hardline stance of the owners, even after the players have conceded to the owners a higher revenue percentage than that in any other professional sports league, is proof that the owners are trying to rig the game. The only way for the players to get their fair share is if they move the game to a casino where the league isn’t dealing the cards.

Carmaré: Brotherly Love or Sibling Rivalry?

Let’s say you have a player — we’ll call him Max — that, whenever the offense is run through him, the team’s offensive efficiency increases by 30%. Unfortunately, Max is not the greatest of defenders. In fact, he prefers to take short naps while his team is on defense, curling up at center court and allowing his teammates to play four against five. As a result of his poor defense, the opposing team’s offensive efficiency improves by 20% whenever he is on the floor. Still, as long as Max is the center of the offense, his team is looking pretty good. A 30% increase is, after all, better than a 20% one. 

But wait! Max has a twin brother named Hortense, and they are really dying to play together. They feel nostalgic for those games of two-on-two from their childhood. Hortense is not the same player as Max, but he’s similar. When the offense is run through him, the team’s scoring efficiency improves by 20%. On defense he’s slightly less comatose. He rebounds pretty well, and sometimes he tries to stay in front of his man. Still, due to his inconsistency of effort and focus, opposing offenses improve by 10% with him on the floor.

So what happens when Amare and… err I mean Max and Hortense are on the floor together? Well, their combined poor defense results in a 30% increase in opponents’ offensive efficiency, and if they’re well-disciplined and always run the offense through Max (despite Hortense’s continual complaints), they also score 30% more efficiently. They are a .500 team.

Obviously this is a simplification, but I don’t think it’s that far off. When you have guys who are only elite when they have the ball in their hands, do not play good defense, and aren’t great passers, the law of diminishing returns weighs heavily on them. Sure, there is the benefit of being able to always have a scorer on the floor, and certain matchups may favor one scorer, but overall, it’s the worst kind of skill replication there is. For evidence, you can look here and see that New York was at least as successful with Carmelo on the floor and Amaré off as with the two playing together. Compare those numbers to Boston’s lineups where any lineup without three of their big four has a negative +/-. Every other lineup is dominant. That’s healthy. The Knicks are not. However, as the minutes for New York’s alternate lineups post-trade are too small to make any grand conclusions, let’s compare this union to past unions or disunions of high usage players:

Zach Randolph —
Age 25-27 average WS/48: .086
Age 28-29 average WS/48: .169 

Notable changes: Left the Clippers/New York, where his interior scoring ability was replicated by Kaman and Curry, for Memphis, where the low-post belonged to him.

Ben Gordon —
Age 23-25 average WS/48: .123
Age 26-27 average WS/48: .048

Notable changes: Left Chicago via free agency for Detroit, where his skills were replicated by Richard Hamilton.

Vince Carter —
Age 34 WS/48 (with Orlando): .160
Age 34 WS/48 (with Phoenix): .060

Notable changes: Was traded to Phoenix, where he was expected to be more of a role player, catching and shooting or else moving the basketball.

Kevin Durant —
Age 21 WS/48: .238
Age 22 WS/48: .189

Notable changes: Westbrook’s usage rate jumped 5%. Durant’s usage only fell by 1.4% but the way he was using those possessions changed as Durant became more of a catch-and-shoot role, evidenced by his increased three point attempts and decreased free throw attempts.

Players that are at least as valuable as off-ball scorers and/or are good defensive players do not suffer from these sorts of fluctuations as often and are more likely to benefit from a diversity of offensive talent. Pau Gasol, for example, has been far more effective in LA, where he can focus more on defense (his rebound rate improved significantly) and doesn’t have to force his offense (thus the lower turnover rate and more efficient scoring). Ray Allen, who is just as comfortable (or perhaps more comfortable) catching and shooting as being the number one option, has arguably had four of his top five seasons since joining Boston.

What does all this mean? Well, I think Carmelo has made a case with his hot shooting that he can be effective off the ball. However, is he willing to play that role? Is his friend Chauncey willing and able to facilitate an offense where Carmelo is playing that role? Is our new general manager (or owner) willing to employ a coach who will spread the floor and run pick and roll?

If the answer to any of these questions is “No,” I see Stoudemire’s effectiveness declining in a big way. Sure, he’ll make his share of 18-footers, but without pick and roll action with a good PnR point guard, Stoudemire is a huge liability. He is bad off the dribble and so can’t punish defenders who are closing out. He is capable from 18 feet, but that’s the most inefficient shot in the game, and he is a bad defender and rebounder. If we don’t need him as an initiator of the offense, then he is no better than, say, Thaddeus Young, who while maybe not the shooter that Stoudemire is, is just as good a rebounder and a superior defender. You know what happens when 28 year old All-Stars with major knee surgeries in their past start looking like Thaddeus Young? Their trade value plummets. 

That means the Knicks have to go to their local Kinko’s, and they have to ask the pimple-faced dude behind the counter to print a poster at the largest size that says, “Get a point guard that has the backbone, the will, and the skill to run the offense through Amaré Stoudemire, or else trade him. Deadline: 2/23/12.” Heck, even if you have a secret paper with Chris Paul’s signature across the bottom that says he’s committed to coming to New York, I still think if an offer comes along this summer for a player of equal value to Stoudemire but who makes his living more off defense and/or playing off the ball on offense (say Andrew Bogut or Joakim Noah), that’s a trade that you jump on, especially if a certain coach with a striking resemblance to the Pringles man is no longer employed here.

What is Mike D’Antoni’s Offense? (Part III)

The chart below shows all of the players who have played 1,000 minutes for SSOL Suns (including the last two non-D’Antoni years, since they’ve largely held to the same system). I also included Carter and Barbosa, since they will both qualify by the end of the season.

LEGEND

Threshold age: The age the player was during the full season before or after that player left or joined Phoenix that is closest to that players prime. I defined a player’s prime as his 27 year old season. For example, Raja Bell played for Phoenix from ages 29-32. His stats with Phoenix below are those from his 29 year old season, and the stats for when he was not with Phoenix are from his 28 year old season (with Utah). I do this so as to prevent interference from players’ rise and decline due to age.

WS/48 w/ SSOL: The player’s win score for the chosen season (as described above) that he played for Phoenix.

WS/48 w/o SSOL: The player’s win score for the chosen season (as described above) that he played for another team.

WS differential: The amount a player’s win score increased/decreased when comparing his Phoenix season to his season without Phoenix (positive numbers mean improved WS/48 with Phoenix).

subjective player type: My subjective description of the player’s skillset.

Player Name threshold age position WS/48 w/ SSOL WS/48 w/o SSOL WS differential subjective player type
Shawn Marion 28 3/4 0.191 0.101 +0.09 2 way offensive double threat
Channing Frye 26 4/5 0.141 0.054 +0.087 offensive perimeter big
Quentin Richardson 25 2/3 0.098 0.023 +0.075 2 way 3pt threat
Raja Bell 29 2 0.110 0.060 +0.05 2 way 3pt threat
Shaquille O’Neal 36 5 0.166 0.119 +0.047 slow scoring big
Steve Nash 30 1 0.203 0.162 +0.041 MAESTRO!
Amare Stoudemire 27 4/5 0.181 0.145 +0.036 offensive mobile big
Brian Skinner 31 5 0.069 0.035 +0.034 defensive mobile big
Steven Hunter 23 5 0.121 0.090 +0.031 defensive mobile big
Joe Johnson 23 1/2/3 0.112 0.086 +0.026 distributor, volume scorer
Leandro Barbosa 27 1/2 0.038 0.014 +0.024 offensive double threat
Tim Thomas 28 3/4 0.069 0.054 +0.015 offensive double threat
Grant Hill 35 3 0.127 0.123 +0.004 jack of all trades
Jason Richardson 28 2/3 0.119 0.116 +0.003 offensive double threat
Matt Barnes 28 2/3 0.069 0.077 -0.008 wing defender
Jim Jackson 34 2/3 0.078 0.102 -0.024 3pt threat
Kurt Thomas 33 5 0.087 0.112 -0.025 slow defensive big
Boris Diaw 26 4/5 0.063 0.103 -0.04 jack of all trades
Eddie House 27 1/2 0.054 0.094 -0.04 3pt threat
James Jones 26 2/3 0.077 0.139 -0.062 3pt threat
Vince Carter 34 2/3 0.065 0.154 -0.089 3pt threat, volume scorer

I make five conclusions from this list:

  1. D’Antoni seems to get a lot of value out of defensive mobile bigs when he uses them.
  2. Guys who do nothing but shoot threes (i.e. don’t add anything defensively and can’t dribble drive) contribute less to SSOL than to more traditional offenses.
  3. If you are capable of defending the four or the five, and either faster than the average 4/5 or a good perimeter shooter, you can help SSOL a ton. If you can do more than one of those things, then you are a SSOL god.
  4. Volume scorers, if they don’t reform their ways, can be cancerous to SSOL (Vince Carter and Joe Johnson).
  5. Shaq likes to prove neigh-sayers wrong.

Let’s start with number three and who I dub the king of SSOL. Shawn Marion is the crowning example of what SSOL values. Despite the fact that during the 07/08 season his primary backups were Shaq and Boris Diaw, both of whom are bigger and stronger than him, the defense was a full 5 points per game better with him on the floor. The offense, meanwhile, improved by 1.9 points. I’m sure if we incorporated pace into the equation, that difference would be even larger. On offense, Marion can do things that pressure opposing bigs in ways they are not used to being pressured. He can run the floor, he can shoot, and he can get to the basket quickly.

Remember how the pick and roll forces a defender to hedge in? Well, Marion’s man is likely bigger, slower, and quite possibly uglier than the guy guarding the two or the three. These differences make it harder for the Marion’s man to recover to Marion after a pick and roll, giving Shawn more space.

The reason Marion’s numbers declined so much, even in the season immediately following his departure from Phoenix, is because he is not a good scorer. He has no post game and he is not good at beating a set defender off the dribble. These are the skills that most offenses demand. His strength on offense in SSOL mostly comes from his speed.

The last point I want to make about Marion’s success in the D’Antoni offense is that it he was never a good three point shooter. He was a 33% three point shooter on his career with D’Antoni, exactly the same number as his career average, and only averaged a little over one make per game. Having a 33% three point shooter attempt threes is not good value — the equivalent of a 49.5 TS%: less efficient than the Milwaukee Bucks’ offense this year, the least efficient team in the league. (I have this weird feeling that there’s a guy on our bench with exactly these skills. I can’t figure out who though…)

In regard to scorers, I think the plain and simple fact is that if they can reform their game to that of the offensive double-threat (shoot, drive, or pass, but don’t stop the ball) then you can be effective in SSOL (Joe Johnson), but if you are used to stopping the ball and taking over the offense, then you are not going to be as efficient in SSOL.

Finally, I want to briefly (D’Antoni style…) talk about defense. Much of Marion’s value comes from his defense — his rebounding, shot blocking, thievery, and general capacity to do things that bigger players can do without the cost to his speed. In general, I think that’s why seven of the ten guys whose value increased most are above average defenders. Two of the remaining three (Nash, Stoudemire) are the dynamos of the offense, and then there’s Channing Frye, whose value comes from his amazing efficiency from the perimeter. This to me is a result of the fact that because SSOL interior players are generally less imposing, it falls to the wings to compensate. Q-Rich helped on the boards, Marion helped with boards, blocks and steals, Bell helped with his man defense.

What is Mike D’Antoni’s Offense? (Part II)

As I said in part I, only the point guard has complicated decisions to make in D’Antoni’s offense. For the rest of the team it should be a series of simple reactions. Hence point guard is the most important position in the SSOL. This year the main thing plaguing New York’s offense is Felton’s inability to make the perimeter shot. The more he misses, the more the defense slacks off onto Amare. This is why Felton is shooting so much. Additionally, the more the defense slacks off onto Amare, the fewer options Felton will have to make plays for others. Here’s a dandy chart showing the relationship between Felton’s TS% (X-axis) and Stoudemire’s TS% (Y-axis):

According to the above chart (data through the Clippers game on 2/9), on average each percentage point increase in Felton’s TS% increases Stoudemire’s TS% by 0.24 points. In other words, the more efficiently Felton scores, the better Amaré is.

Here’s the graph for Amare’s TS% (Y axis) and Nash’s (X axis) last season:

There is virtually no relationship between Stoudemire and Nash’s TS%. This to me suggests that, while teams respect Nash from the get-go and commit to defend him regardless of how well he shoots, they likely take a “wait and see” approach with Felton. If Felton can’t prove to them that he is a dangerous shotmaker, they lay off of him and commit to keeping Stoudemire out of the paint. The statistics bear this out: Stoudemire is taking a lot more shots out of isolation plays. Last year in Phoenix, 61% of his baskets came off assists. This year, only 48% of them do.

Why the skilled passer is important in D’Antoni’s offense:

• As we saw in the video clip from my first post, the window to make a pass to a perimeter player and to make interior passes to a roller opens and closes very very quickly. If the passer mistimes the pass, it may result in a turnover (either on the pass, or as we’ve seen sometimes, forcing Stoudemire into a charge) or more likely simply in a shot with a higher degree of difficulty (a covered three or a shot in the paint with more traffic).

• Understanding spacing and angles is also critical. When Felton dribbles through the pick, he has to make a judgment regarding how the defense is responding. Depending on how the defenders position themselves, Felton will have to go to different spaces on the floor in order to create an angle to pass the ball into Stoudemire.

• Finally, if Felton gets into the paint, good defenses will cover the easiest passes to the perimeter. Sometimes, the player who is most open will be the guy at the top of the key. Passing the ball out to a covered perimeter player, then rotating the ball takes longer and allows the defense more time to recover.

What is Mike D’Antoni’s Offense? (Part I)

The element of surprise is as old as life itself. With species competing for life & death, every advantage is critical. Anyone that’s witnessed a sucker punch knows humans have been no exception, as catching an opponent unprepared can lead to a easy victory. The general philosophy of Mike D’Antoni’s offense, aka Seven Seconds Or Less, follows on the same principle. The goal is to get the ball up the floor before a defense is prepared for an attack. It’s the NBA’s version of the Trojan Horse.

D’Antoni features a lineup that is quicker than the other team. Ideally, this sleeker group races up the floor where the ball-handler gets into the paint before the slower opposition, especially the big men, can get back. Backpedaling wings may be able to collapse in order to compensate and prevent a score in the paint. However the SSOL team’s other players each have a spot around the three-point arc, and the ball-handler’s job is to pass the ball out to them when the defense turns inward. The fast break has been used for decades, however without a concentration of perimeter players other teams will not be able to take advantage of as many of their fast break opportunities. This is due to D’Antoni’s use of three point shooters to spread the floor wider and more efficiently.

In lieu of a fast break, D’Antoni’s offense relies on the pick and roll as a means of simulating the above situation. The point guard will run a pick and roll with a big (for the Knicks, usually Amare, sometimes Chandler, Turiaf, or Mozgov), forcing the defense into making difficult decisions. In the video below, you can see how different types of defensive approaches affect SSOL and what kind of skills it demands of the offense in order to maximize efficiency.

The defense can react to a Felton/Stoudemire pick & roll with:

BIG SHOWS: Stoudemire’s man can come out and “show” on Felton, preventing him from having a direct line into the paint or an open shot, then either try to recover back to the diving Stoudemire or stay with Felton. Because Stoudemire is quicker than nearly all the defenders who cover him, most bigs can’t recover to him, especially when Stoudemire is playing at the center position. Additionally, if Stoudemire senses that the big is going to show, he will often slip the screen, diving to the basket before Felton dribbles through the screen and making it even harder for his man to recover. In this situation, teams often send a third defender to cut off Stoudemire’s path to the hoop.

BIG AND SMALL SWITCH: In this option, the big takes Felton and the small takes Stoudemire. When this happens, Felton will always have space behind the screen to take a jump shot. If he can nail that shot, he can force the big to show, which opens up the options above. If he can’t, then he can try to beat the big with his quickness. Additionally, if the small doesn’t commit early enough to the switch, Stoudemire can sometimes still slip the screen, keeping Felton’s man pinned to his hip. However, this will only happen against poor defense. The other option is for Felton to try to beat the big using his superior quickness. So far, he has not been particularly adept at this. It’s another thing that Nash is superb at.

BIG STAYS WITH STAT: Stoudemire’s man can stick with Stoudemire, preventing him from charging to the basket. If this happens, then Felton can take the ball into the paint himself (his man will be behind him), forcing a third defender to come in and help and leaving a man outside open. The angle on the pass to the perimeter player is often a difficult one, so this is the time when Felton’s passing becomes most important.

BIG SPLITS THE DIFFERENCE: The defending team can do something in the middle, leaving the big around the free throw line, close enough to Stoudemire to cut off the pass inside and close enough to Felton to cut off a drive. The best option here is for Felton to take the open 20-24 footer, a shot that he has struggled mightily with of late and one which whatever point guard we have will have to make consistently if we want the offense to thrive.

A couple of points I want to emphasize:

  • The dribbler on the pick and roll has to make a lot of tough decisions, but outside of him, the thing that sets SSOL apart from other pick and roll offenses is, in my opinion, that it demands less decision making to get a good shot. It increases the probability (by spreading the floor and moving quickly) that the first pass will lead to a high percentage shot.
  • SSOL does not thrive off of guys who are good at scoring on their own. Players who score for themselves are unnecessary, as SSOL is designed so that when players catch the ball, they already have the advantage.