I’m Andrea Bargnani and I endorse this stat.

Since the start of the 2010-11 season, there have been 17,301 instances of an NBA player taking 10 field goal attempts or more. According to a “new” “stat” “I” just “invented,” as I’ll outline below, Andrea Bargnani’s 2-for-19 performance from last Sunday’s game against the San Antonio Spurs was the 33rd least efficient scoring performance of all of them.

The purpose of this article is not to be overly statistical – I’ll use math terms, but only to explain the logic – the point of this article is actually to simplify. The point is also not to rag on Bargnani. There has been some misunderstanding and disagreement among the Republic of late over what exactly makes a good game for a scorer.

For example, while some see “22 points” in the box score and deem it a strong performance, others see “10/21 FG, 0/2 FT, 2 TO” and deem it a pretty poor performance. What I wanted to do was take interpretation of the box score out of the equation and provide a simple “scoring efficiency” rating we can look at on a game-by-game basis.

Of course, box score stats are flawed, and even something like field goal attempts doesn’t tell you much without watching the game. But, for the purposes of improving our discussion, I introduce to you my “new” “stat” which isn’t really either of those things…

PRIMO – Points Rated In My Own (Way)
Simply put, PRIMO is just points divided by “chances,” defined as FGA + Turnovers + (Free Throws / 2). Chances, here, is an approximation of possessions that we can calculate easily to avoid going math-heavy.

What PRIMO will tell you is simply how many points a player scored based on how many possessions they used. (Again, this ignores other important aspects of the game like assists, rebounding, defense, and more – this is JUST a scoring efficiency measure.)

So What is a Good PRIMO Score?
To keep it easy in your head, you can consider a PRIMO score of 1.00 to be an “okay” score – a player scored one point for every possession they used. Since teams have averaged about 92 possessions per game over the past few years in the NBA, if a player with a PRIMO score of 1.00 used every possession for a team, the team would score 92 points (under the poor assumption of no offensive rebounds – forgive me for the assumption, but this IS a Bargnani-based analysis). Not great, but not terrible. If you wanted your team to score 100 points (free pizza!), you would want players to hit a PRIMO score of 1.09, for an average-paced team (the Raptors play at a slightly slower pace, 91 possessions per game).

For reference, the league average PRIMO score so far this year is 0.9 at the team level. Over the past three seasons, for any player who took 10 field goal attempts or more, the distribution of PRIMO scores is as follows (I should note that PRIMO will look silly for players with less than five field goal attempts, and the point of it is to compare higher-volume shooters).

PRIMO Distribution, 2010-11 to Present
For all player games with greater than 10 FGA.

Basically we can see that these scores are close to normally distributed – that is, the bulk of the games are around average, with fewer and fewer games further away from the average. I’ve drawn a few lines in there for reference.

PRIMO – Raptor Example

As you can see, the stat does a decent job backing up what our eyes have told us so far: Bargnani has been inefficient in scoring his heavy points total, DeRozan has been a bit more efficient, Lowry’s value is negated some by his turnovers and trigger-happy nature, etc. At the same time, as I mentioned at the outset, this stat tells you nothing of Jose’s passing, Jonas’ rebounding, etc. This is only a stat to measure scoring efficiency.

So What’s the Point?
There isn’t one, really. When trying to figure out just how bad Bargnani’s recent game was, I wanted to roll out a VERY simple and basic stat we can use when quickly scanning the box scores.

But next time you’re looking at a box score (which, of course, is absolutely no replacement for watching the game and analyzing what you SEE), keep efficiency in mind. And specifically when analyzing the Bargnani box scores, think PRIMO.

Follow @BlakeMurphyODC

How does Lowry rank with other free agent point guards?

So we pooched our Fields/Nash analysis. It did end up in a situation where New York could, in theory, still obtain Nash, but it required Phoenix to take on a lot of parts (and NY having to give up Shumpert). Almost media outlets (as well as myself), didn’t fully consider (although our team did) the Lakers TPE – largely as the Lakers would have already over $70 million in salaries committed for next year to five players, two being World Peace and Blake. It completely makes sense for Nash given his “criteria”, but the Lakers will have to be shrewd to round out their roster.

More importantly, it meant the Raptors rapidly moved towards “Plan B”. And, in some respects, its a much better one.

The following charts compare six key free agent point guards. The first section is based on “per 36 minutes” numbers while the second section is “advanced” metrics (click on image for full view).

Source: basketball-reference.com

My original post was going to be why I believe Lowry would be a *much* better addition then Dragic. Lowry matches Dragic in most metrics and is a much better rebounder and gets to the line much more. Lowry is a steal compared to Dragic’s $8.5 million average salary over 4 years. Houston was wise to pass on Dragic in favour of Lin’s $7.2 million average salary over 4 years.

In terms on adjusted +/-, Lowry also screens very well:

Source: basketballvalue.com

Goran Dragic’s adjusted +/- is more mixed and it appears he is as good of a defender.

And what about his numbers versus a 25 year old Nash?

Source: basketball-reference.com

Kyle Lowry may be the steal of the free agent class offseason.

Questions? There is a dedicated to “Statophile Q&A” forum thread here . If you prefer to send questions privately, you’re welcome to email me at tomliston [at] gmail [dot] com or find me on Twitter (@Liston).

Addendum (7/6/2012 8am): Here’s a link comparing Lowry and Jerryd Bayless. Many similar numbers.

Five Points on Fields

1. You could listen to Bill Simmons
   or you could rub a few facts together. I know its hard. But if you bothered to look, Landry Fields played the second most minutes of any Knick last year. No, he wasn’t the 11th man as much as you wanted to believe it (thinking is hard). And the offer is for $19 million. And its back end loaded. And Steve Nash turned 38 this past February. But making crap up to illicit a passionate response is the name of the game and Bill knows clicks are clicks (yes, we advertised it as well and most everyone simply accepted it as the gospel).
2. You could also look at the the structure of the deal. Some of the very same Raptors’ fans endorsed Daryl Morey’s move to offer offensively challenged Omer Asik $25 million over THREE years as brilliant. Why? It was a “poison pill” type structure – in this case severely back end loaded, which essentially guarantees Chicago cannot match (note: this is not a direct comment on the merits of the Asik deal – I realize he’s an excellent defender). Landry Fields’ reported deal starts (see TSN’s Tim Chisholm’s article) at $5 million this year and ends at $8.9 million (see more on this later). Houston needed a big defender while Toronto needed a wing scorer. Why isn’t this deal structure just as brilliant?

   Maybe its even more so. Thankfully, reporters like Chisholm caught it right away: “Without a sign-and-trade, the Knicks are stuck at offering Nash only $3 million per year, or roughly one-quarter of what the Raptors are prepared to pay to get a deal done [...] if the race is really between Toronto and New York, Colangelo’s pre-emptive strike against New York’s sign-and-trade options illustrates the seriousness with which he covets Nash as a future Raptor. Of course, Fields is more than just a pawn in the Nash chase; he’s a player that the Raptors are going to integrate into their attack next season.”

3. Landry Fields can spend minutes at SF. Some have speculated DeRozan would have to move full time to SF. For New York last season, he spent more than half his time at the 3 spot. He’s 6’7″ and can match up fine in many situations there.
4. Some have also noted Fields’ shot only 25.6% from beyond the arc last year versus a hot 39.3% in his rookie year. This was obviously cause for concern. However, if you think about the “new look” Knicks offense (predicated on little ball movement, except for Lin) in his second year, it is understandable. Last year, NYK had the 4th highest turnover rate and was ranked 22nd in % of field goals assisted on. As an example, Fields’ 2pt FG% “improved from 50.2% to 55.7% playing with Jeremy Lin last season” (as per Kevin Pelton). Its likely he’s not a 40% three point shooter, nor a 25% shooter. Coupled with a pass first PG that can also shoot (hint: Nash or Calderon), he’s likely somewhere in between.
5. His adjusted +/- is quite good. Yes, this can be “noisy”, but (click image to see it completely)

   
   Fields appears to make positive contributions to both offense and defense when on the floor. Fields has the best 1 year and 2 year adjusted +/- of all Knicks who played over 1,000 min last season. And Landry just turned 24 and most players continue to improve significantly at this age.

Net, net: its a very smart gamble.

You miss 100% of the shots you don’t take. – Wayne Gretzky

This is a shot worth taking. This is not 30 year old, beat up Jermaine O’Neal. This is not a four year deal for a defensively challenged Jason Kapono.

Here is an example of the decision tree (well, its not exactly a “tree”, but you get it) I would make in this situation:

I’m sure I could’ve displayed this better, but I hope you get the idea. The most important piece of the analysis is the removal of New York from the Nash equation (although *maybe* they could try a sign and trade with Lin + parts, but I don’t see it) and thus the increased odds of Nash joining Toronto. It also introduces option 2b, which not only lands Nash in Toronto (again making it near impossible for NYK to be a factor), but also has them accepted a large deal in the face of mega deals for Carmelo and Amar’e. Thus, 2a becomes a likely scenario and is not without risk. Toronto could increase its odds to land Nash, but still has a chance of only landing Fields – I peg this scenario at 20%. In this worst case scenario, we’d be paying a player that is much more productive than most believe, especially when part of a pass first PG and better ball movement. At $5M or so the first two years, then an expiring deal, this is a bold but calculated move. If you are concerned by the combined 3rd year salary for scenario 2a, then consider the dramatic increase in revenue MLSE will get by this move. Would the Raptors average attendance go from 16,800 last season to something like 19,500 like it was in 2008? Using TMR’s Fan Cost Index of approximately $75 a ticket, that’s $25 million in additional ticket revenue alone. Add increased money for TV deals (viewership would be up significantly with Nash), a few playoff games (~$2 million per game in ticket revenue alone), merchandise, Real Sports etc and you could probably double it. So, by year 3, MLSE will have significantly more resources handle these salaries and deal with any potential cap tax issues (i.e. will be able to easily handle deals which put them over the cap – given the large amounts Nash and Fields will earn in year 3).

There are very few “no risk” deals. Teams always have to balance risk and reward. I believe the Raptors have been savvy with this deal. It’s always a gamble, but this is a good one.

Questions? There is a dedicated to “Statophile Q&A” forum thread here . If you prefer to send questions privately, you’re welcome to email me at tomliston [at] gmail [dot] com or find me on Twitter (@Liston).

Older and Damn Productive

A quick one since its timely.

Everyone seems concerned with Nash’s ability to contribute at 38. While its very rare for a player to be productive at this age, its not without precedent. Here’s how “old and unproductive” Nash was last year – in a compressed season where most of the older guys suffered through the schedule.

Source: basketball-reference.com

He’s obviously too old to play any more. How did he stack up last season against other PGs – of any age?

Source: hoopdata.com

So Nash has been very productive despite his age. But can he hold up and be productive at reasonably the same levels? NBA aging curves suggest Nash should be in steep decline. However, PGs can often hold up longer, especially “pass first” PGs. And Nash’s diet and exercise regimen is well documented.

A PG playing at a high level at age 36 or older isn’t without precedent. The great John Stockton played at a very high level for several years beyond 36. I don’t see why Nash couldn’t replicate this feat.

Source: basketball-reference.com

I’m not sure how one can argue bringing a winning culture – with a NBA champion as a coach (Casey) and a two time MVP as a leader – to an organization is a bad thing.

Could he get hurt and it turn to be a bad deal? Sure. Could a #1 pick get hurt and play less than 100 career games? Yeah. There are no sure things – but this is as smart of a move as it gets.

Any time you have an opportunity to get a top ranked producer and a first class leader, you do it.

Questions? There is a dedicated to “Statophile Q&A” forum thread here . If you prefer to send questions privately, you’re welcome to email me at tomliston [at] gmail [dot] com or find me on Twitter (@Liston).

Draft Q&A with Alex Rucker

We are planning to do a more comphresive feature on the Raptors’ investment in Analytics, including a focus on consultants Alex Rucker (who’s moved to Toronto and will be full time on July 1st) and LA based Keith Boyarsky. However, I thought it was best to “jump the gun” and focus on their role in this year’s draft as I was especially interested in getting their insights about the Raptors selections.

For background, Mr. Rucker (@Alex_Rucker) has a BComm from UBC and a Law degree from Notre Dame. He’s worked in various analytics roles with the Vancouver Grizzlies, Notre Dame Basketball and came to the Raptors in 2009. He met Jay Triano back when he was in high school and worked with him at Simon Fraser and the Grizzlies. He recently retired from the U.S. Navy after a 10 year commitment and will be working full time with the Raptors on July 1st with his analytics partner Keith Boyarsky (who was flying back to LA during this interview).

Tom Liston: When the Raptors selected Terrence Ross with the 8th pick, my immediate reaction was “jack of all trades, master of none”, meaning he does a lot of things reasonably well (TS%, 3pt%, rebounding for his position, etc) but doesn’t appear to excel in any one area.

1. Where did the analytics team have him ranked?
2. The big hole in my analysis is his potential impact on defense – were you able to distill some data on his impact?
3. How did the process work in selecting this pick and how was the analytics team part of it?

Alex Rucker: Analytically, Ross was a very easy selection at 8th overall.

1. We had him ranked as the 7th best prospect – and one of the guys ahead of him slid down the board due to medical concerns. When we got to our actual pick, we took the best player available. I agree with the general sentiment that he does a lot of things reasonably well – and that’s precisely one of the reasons he projected well in our in-house analytic tools. Unless you’re looking at a guy who can provide elite production in a key area, you’re often better served taking a player who can contribute positive value in multiple key phases of the game. I’d much rather have a guy who’s a slight positive in 6 areas at the pro level than a positive in 1-2 areas and a negative in the others. Bottom line:athletic wings who can hit the 3, play defense and get after it on the boards have positive value on nearly any team, regardless of how it’s constructed.We’re a team that’s building and improving, and based on the data we’d accumulated on him, Ross was a guy we assess will very quickly become a player who can contribute on both ends of the court.
2. With respect to his defense – I agree that this is the hardest thing to discern from college data. College data is inconsistent and often difficult to both collect and work with, but there are absolutely some important inferences which can be drawn from it, even on the defensive end. Ross projected as a guy who will be a “plus” wing defender, both as an on-ball defender and as a help defender.
3. The analytics team has been integrated into the draft process for the past 2-3 seasons. We are involved in the initial screening of candidates, essentially using our analytic suite to identify players who’s college data does not support taking them at our draft position – or at least make clear that if they are drafted, it will be purely on the basis of potential, not prior performance.
   Additionally, our analysis has gotten pretty well-refined in terms of translating how a college prospect projects at the pro level based on identified roles and positions. Many players played a different position in college than they will at the pro level, and in a different role, and historical analysis allows us to do a better job of projecting what they’ll look like at our level.
   At the end of the day, the final determination of a prospect is a mix of player evaluation, analytics, background checks, medical checks (both physical and psychological), etc. Our final draft board, I think, reflected our team’s invesment in each of those areas – but yes, it is clear that the analytic piece helped shaped both that list and the ultimate selection.

Liston: ESPN’s John Hollinger said “I had Ross 27th on my board. Not a fan of the pick and thought they could have traded down if that was their guy.” What is your rebuttal?

Rucker: It’s entirely possible we could have traded back and still gotten Ross. I know that Bryan Colangelo was extremely active throughout the entire draft, and he investigated a wide spectrum of options for our team. While we did have opportunities to move back, there was a very real chance that doing so might cost us the player we wanted. I think that Bryan weighed the benefits and risks associated with the various scenarios, and made the optimal decision for our franchise.

I’d also offer that the media’s perspective of player values is interesting, but it doesn’t always align with pro team’s perspectives – and one of those two enjoys a significant information advantage by virtue of the resources invested into the draft process.

If you look back historically, I think you’ll find that the real decisions (actual drafts) fare better in terms of assessing player value than the media projections (mock drafts). Neither is perfect, but one has less error than the other.

Liston: Who do you think is the biggest potential “steal” of the first round?

Rucker: Sullinger to Boston at 21. He has a chance to be a really good player at the pro level. Henson at 14 was another excellent pick. Both of those guys went later than I’d have selected them, and they ended up in great situations.

Liston: For the second pick, I was big on Jae Crowder, who went to Dallas – known for their big analytics team – right before your pick at 37.

1. Was he on the radar?
2. Why Quincy Acy, who doesn’t appear to fill a need a much as perhaps a Scott Machado, Tyshawn Taylor or Kim English (three others I was big on).
3. Hollinger also didn’t like this pick – he noted: “Also, Quincy Acy over Quincy Miller continues a WTF night for the Raptors.” Your rebuttal?

Rucker:

1. Crowder was definitely a player who was in the mix – but as you noted, he ceased to be an option 3 picks before ours.
2. We liked Machado and Taylor, and they were given strong consideration – but the strong play by Ben Uzoh at the tail end of last season as a backup PG was a factor. Once you get into the 2nd round, analysis suggests that your best options are niche guys, players who can fit into an identifiable role and contribute positive value in that role. Acy is appears to a great fit with the team we’re building. Our assesments of him supported the position that he’s a guy who will do just that – give us a rugged, low-post battler who doesn’t take anything for granted and will be looking to compete anytime he steps onto the court.
   We’re continuing to build a winning culture, one with a strong work ethic, a commitment to defense, a toughness – and Acy is going to be a positive contributor to that.
3. Quincy Miller is a talented player, and I’m looking forward to seeing how he does in Denver – which I think is a great spot for him. However, for our needs and what we were looking for, Acy was the guy. “Best player available” doesn’t exist in a vacuum. It exists in the context of your team, the players on your roster, and how that player will fit in with those things. I think sometimes people confuse “most potential” or “most skilled player” with “best”. Analytically, “best” relates to the sum of a player’s contributions to his team – and that’s rarely going to align with either of those two alternate definitions.

Liston: Who do you think is the biggest potential “steal” of the second round?

Rucker: While the odds are against any 2nd rounder being a useful NBA player, both Doron Lamb and Kim English are players who can clearly fit a role, and contribute positively in that role. I think both landed in great spots and am looking forward to seeing what they do going forward.

Liston: What can you tell us about Tomislav Zubcic?

Rucker: Zubcic is a 7′ combo forward from Croatia. He’s athletic, runs the floor, and looks to push the ball in transition. While he’s not a guy who can necessarily create shots for himself, he’s a strong spot-up shooter. The key for Zubcic will be his continued development. If he can get stronger and continue to refine his 3PT shooting, he could be a useful piece down the road.

Liston: Several Raptors’ fans were screaming (well, tweeting in ALL CAPS) for the team to acquire a late 1st rounder to pick up a dropping Perry Jones III. A) Was this considered? B) Where did you have him ranked?

Rucker: It should be no surprise that his analytic ranking would suffer as a result of his underwhelming play at the college level. He was not a player who projected well in our models, and his ability to contribute at the pro level will depend entirely on significant improvement going forward.
As with any trade scenario though, it’s easy to get excited about what you might be getting – the question is whether it’s a net positive once you factor in what you’re giving up, and what competing scenarios come off the table.
I think that we made positive-value decisions last night, and I can’t wait to see how the next few weeks unfold – there’s no question that we’re at a very high-leverage moment for our organization.

Liston: Thank you for your time. It was quite insightful.

Questions? There is a dedicated to “Statophile Q&A” forum thread here . If you prefer to send questions privately, you’re welcome to email me at tomliston [at] gmail [dot] com or find me on Twitter (@Liston).

The purpose of this post is twofold:

1. To highlight potential “value” picks later in the draft.
2. Try to highlight “risky” players at the 8th spot.

Whether its valuing stocks, houses, or college players, the “secret” is the same – observe factors that are consistently over/undervalued to find opportunities others may be missing.

Often one can find value later in the draft. This draft does not appear to be unique in that aspect – if the mocks are anywhere close to reality.

Several biases have been observed when it comes to drafting players:
David Berri (of Wins Produced fame) and Martin Schmidt from “Stumbling On Wins“:

All else being equal, players will be drafted higher if the player is younger, recently appeared in the Final Four, and is relatively taller.

And it can be signficant:

A player who appears in the Final Four can improve his draft position by about 12 spots.

Most of this is understandable. Scouts and GMs will, in general, give too much credit: A) to one player if his team makes the final four; B) the impact on measurements C) “Upside” and “ceiling”

Sometimes a good rebounder is just a damn good rebounder. I don’t particularly care if he’s 6’8″ in heals or 7’1″ without shoes. If he puts up the numbers, he puts up the numbers. You know where to shove your wingspan numbers.

“… you guys just sit around talking the same old ‘good body’ nonsense like we’re selling jeans.”

I’ll be upfront that “but you NEED TO WATCH THE GAMES!” comments could be particularly valid in this analysis. No, I didn’t have time to watch Kim English’s 35 games played (nether did you, BTW). It’s not the purpose of the article. Most scouts and GMs don’t get to watch all 35 games of 75+ players. Nor do their memories always serve them well.

Our focus is: who has put up the numbers and may be undervalued? As we hinted at above players that: may be a bit older, play for “non-elite” program, and perhaps have a less than ideal wingspan/handwidth/eyebrowlength often emerge as great candidates.

Late 1st round, second round values

The Pint Sized SF/PF Jack of All Trades

Source: draftexpress.com

These, and other impressive metrics, led to the third highest Win Score per 40 min in our draft database, 7th in PER, and 6th highest (in all of the NCAA) Win Shares score (including 17th highest in the NCAA for defensive Win Shares). While I won’t like on combine testing for much, I don’t mind the fact he bench pressed 185 lbs more than anyone else there. And he was Big East Player of the Year. Not bad.

Draft Express’ Johnathan Givony had this to say: “One of the most efficient players in the NCAA, on both ends of the floor, Crowder is the heart and soul of a Marquette squad that has exceeded expectations.” Sounds like a Casey-type player.

Is he a top 15 pick? Probably not. Is he potentially very good value in the second round – even late first round? Absolutely.

The Sharp Shooter, part 1

The Sharp Shooter, part 2

The Glass Cleaner

Options for our 8th pick?

Mr. Average at 8th?

Source: draftexpress.com

His WS/40 is even lower at 36th in our database as he doesn’t rebound much either (only 5.2 per 40 min). His TS% (59%) and eFG% (56%) are both a couple percentage points above the group averages, but not exactly in the top tier. Defensive performance is tougher to measure, but this interview is also a bit cause of concern (~ 2:15 mark). One shouldn’t have to “find ways to keep (their) energy up” at an elite level.

Rivers Doesn’t Flow

Source: draftexpress.com

Could be Worth Waiting on Waiters

Source: draftexpress.com

Likely the best choices?

37th Pick Jae Crowder, Marquette
57th Pick Kim English, Missouri (other options Jenkins [above] or Will Barton [not mentioned], Memphis if they drop)

You may see an obvious problem with these picks. Our biggest need is at the wing. While we’ve picked up good value and potentially quite productive wings, they are not likely starting caliber wings right away. We’ve added another PG with a top pick when we have two capable players already on the roster. Its why, if the rumours are true, we’re looking to make a trade for a stud wing. While I wouldn’t like to part with Calderon, it may make sense to package him with our glut of PFs to acquire a wing. If you move Calderon, draft Lillard. If you happen to find a good deal to move Bayless, you draft Waiters. I rather try to work a trade than draft a mediocre wing at the 8th spot.

(Near) Final Numbers

James Johnson continues to rank highly on most defensive metrics. He could make a strong case if he accepts a role of a big rebounding wing who guards the opponents’ best wing (or perhaps PF on occasion). He has his moments of Jordanitis (symptom: thinking your Michael Jordan; thankfully its curable) on offence too often, but I suspect Dwane Casey will have this solved by next season.

Passa on Sasha

and

Source: basketballvalue.com

and (PER rank of all guard that played over 500 min in 2010-11)

Source: http://basketball-reference.com/
and

(Sasha is 28 years old)
Source: APBRmetrics Forum

The only interesting metric relating to Sasha Vujacic is his fiancé’s leg length (35″ inseam).

It also appears the Raptors could get slightly better productivity at a better price with the likes of a Courtney Lee or Landry Fields.

Update (4/25 10:30pm): Marc Stein of ESPN:

Bullet dodged.

Note: The green Statophile logo for the 33rd edition is a subtile tribute to the great Larry Bird (#33), who I saw in my first live NBA game. In 1984-85, he become only the second player (with the “Big O”) to record >28 points/game, >10 rebounds/game and >6.5 assists/game. And he did it on 58.5% true shooting percentage.

Questions? There is a dedicated to “Statophile Q&A” forum thread here . If you prefer to send questions privately, you’re welcome to email me at tomliston [at] gmail [dot] com or find me on Twitter (@Liston).

Measuring a Coach’s Success

His first screen was based on year over year improvement in winning percentage.

Coach Casey ranks a solid 5th. However, as Mr. Paine points out, “there has to be a better way of accounting for the raw talent on a team beyond simply using the franchise’s previous W-L”. Certainly, player changes (such as LAC’s addition of Chris Paul) should be considered.

Here’s where Daniel Myers’ Advanced Statistical Plus/Minus (ASPM) is valuable.

In essence, ASPM is a boxscore-based, more stable version of Regularized Plus/Minus that is probably the most predictive all-in-one advanced box-score stat out there right now. Denominated in efficiency differential, ASPM is ideal for projecting a team’s performance via a minute-weighted average of the individual ratings of its players. – Neil Paine

Coach Casey still ranks well, but drops to 8th using this methodology:

If he had a healthier Andrea Bargnani, Jose Calderon, and Jerryd Bayless he would likely have screened much better. Starting two 10-day contract players isn’t ideal if you’re trying to win games, yet he’s pulled out a couple.

Advanced Statistical Plus/Minus

I like the ASPM methodology and the presentation of both offensive and defensive scores. It also appears to meet the “eye test”. Most Raptors’ observers would agree Bargnani, Calderon and Bayless are solid offensive weapons, yet are not as effective on defence. Meanwhile, James Johnson, Amir Johnson, Ed Davis and Aaron Gray are solid defenders, but struggle to varying degrees on the offensive end. What’s glaring is the absence of any player in the quadrant for both positive offensive and defensive production.

Questions? There is a dedicated to “Statophile Q&A” forum thread here . If you prefer to send questions privately, you’re welcome to email me at tomliston [at] gmail [dot] com or find me on Twitter (@Liston).

Deuces

DeMar DeRozan

The big hole in his game remains his three point shooting. While he was off to a hot start, he has cooled significantly – shooting only 13% since mid February. The good news is he’s not taking many, but he MUST be able to consistently shoot over, say 33%, to remain a solid starter at the 2. The league average SG shoots 35.2% from deep.

Gary Forbes

Full Season Comparison

Questions? There is a dedicated to “Statophile Q&A” forum thread here . If you prefer to send questions privately, you’re welcome to email me at tomliston [at] gmail [dot] com or find me on Twitter (@Liston).

Tanking – What it Really Means And How to Fix It

“… tanking means intentionally trying to lose games.”

Our TrueHoop network is on a mission to find solutions to “address” the tanking (potential) problem and the draft.

He’s a run down of ideas:

Malcolm Gladwell / Jeff Van Gundy

Concept: Gladwell notes “…the idea of ranking draft picks in reverse order of finish — as much as it sounds ‘fair’ — does untold damage to the game. You simply cannot have a system that rewards anyone, ever, for losing. Economists worry about this all the time, when they talk about ‘moral hazard.’” Van Gundy’s view:”I would either have an inverse lottery, like the best record gets the most chances — so trying becomes of paramount importance. Or at the very least, everybody has an equal chance, so there is absolutely no benefit to trying to be bad.”

Advantages:

• Teams try to win the maximum amount of games.
• As per Henry Abbott (who makes a great point here): It would put a real premium on great long-term team management, which could be the best news ever for fans of bad teams.

Disadvantages:

• The rich get richer.
• The ‘moral hazard’ argument doesn’t completely fly.
• This is not the a “survival of the fittest” competition problem (more later).

Grade: F
Note: I’m ignoring Van Gundy’s “Or at the very least, everybody has an equal chance, so there is absolutely no benefit to trying to be bad” comment, as “F-” isn’t a real grade.

Adam Gold from the MIT Sloan Sports Conference

Concept: “Give the first pick in the draft to the team that wins the most games after being officially eliminated from playoff contention. Then the team with the second highest number of wins gets the second pick. And so on.”

Advantages:

• Teams try to win the maximum amount of games down the stretch.

Disadvantages:

• Just promotes more tanking (different time) and manipulation. E.g. have big cap space entering season, play youth and lose as many games early in the season, bring in a star or near star at the deadline (expiring year star he tells his team he’s not signing an extension, you better trade me – you tell him we’ll have a near lock on #1 pick who’s the “next” Durant.)
• The best manipulators of the system get a guaranteed top pick.

Grade: D

Kevin Arnovitz

Concept: Rookies are simply free agents.

Advantages:

• Removes any concept of tanking.
• “… self-determination for rookies would be a likely [lead to an] uptick in retention.”

Disadvantages:

• The rich get richer – despite what Kevin argues “I have a hard time buying it.”
• Incentives for rich teams to add several scouts to follow and pitch kids at a very young age. Could easily promote bribes, “loans” and other unwanted behaviour to the NBA. Wouldn’t be more than two years before you had an investigation and penalties on a team.
• Some have supported Kevin’s argument “Are we certain a supernova would accept a role as the fourth or fifth option with the Lakers at the rookie minimum” Can easily happen (let’s tweak it to a 3rd option as generally a “big 3″ gets you to the dance). You call Nike and tell the player how much his shoe deal is worth as a “big 3″ player that’s on national TV every 3 nights versus a top banana that’s on national TV twice a year. You “encourage” other local sponsors your big market to also give Mr. 3rd Option a nice contract. Add it all up, he’s making his money while winning multiple championships and being in his favourite city.
• Many smaller market teams that cannot afford Mark Cuban’s world class facilities and don’t have local sponsors “subsidizing” the rookies means you “eliminate” the weaker competition.
• If I lived in LA, Boston, Miami, Dallas, etc I would support it as well. Otherwise you’re potentially facing contraction.

Grade: D

David Lee (via Abbott)

Concept: “Instead of helping teams out who are really bad for a one-year period, the league should distribute lottery chances based on how many times a team has missed the playoffs, or failed to advance past the first round, during the last five years. For example, lets give team two lottery balls for every year over the past five in which they failed to make the playoffs and one lottery ball for making the playoffs but failing to get past the first round.”

Advantages:

• “it removes the incentive to be really really terrible because barely missing the playoffs for a year is going to give team is the same increase in lotto odds as winning only 13 out of 82 games.”
• A team will have to sign up for more years of pain to maximize its chances for a particular super prospect.
• It “flattens” the system.

Disadvantages:

• Puts “luck” at a premium over smart maneuvering.
• An advantage may also be a disadvantage: “A team will have to sign up for more years of pain to maximize its chances for a particular super prospect” – or have a bad GM. Penalizes fans.

Grade: C+

Evan at “The City”

Concept: “Eliminate the lottery. All non-playoff teams are placed in a queue. The first year will be based on the current lottery system (you need some initial condition like this) Thereafter, teams either a) move up in the queue or b) make the playoffs. In the case of a team not making the playoffs they will move up the number of slots equal to the number of teams ahead of them exiting the queue. When a team reaches the #1 spot, they go to the back of the queue the following season (or hopefully, make the darn playoffs finally)”

Advantages:

• It reduces the incentive to lose games.

Disadvantages:

• Encourages more aggressive tanking in a few scenarios. E.g. If you were top 5 in the queue and near a playoff spot – it is the difference between a *guaranteed* top 4 pick or a 16th pick. My guess is many teams would take the guaranteed top 4 pick over a first round exit and 16th pick.
• The team that originally gets the 3rd or 4th pick gets a windfall. Don’t surround those picks with anyone good. This guarantees you 4 top 4 picks – and all will be still young and can grow together. I.e. a team would be incentivized to tank for 3 or 4 straight years in this case.

Grade: B-

Sandy Weil

Concept: Its a bit complex, but it essentially weights the ping pong balls based on winning percentage (i.e. the lower it is, the more ping pong balls) before the trade deadline and provides for some “bonus” ping pong balls by *winning* after the trade deadline (please review the article for a more complete description)

Advantages:

• Discounts the concept of tanking (throwing games definition).
• Still a lottery and thus partially avoids the pitfalls.
• Appears to strike a reasonable balance.

Disadvantages:

• Still can be a bit prone to “working the system”. By tanking early, trading for a good player at the deadline and then playing optimal lineups down the stretch, teams would have an even greater shot at #1.

Weil’s counterpoint:

• Argues that this is a good thing (obviously not on the “tanking early” point): “Imagine a lottery-bound team being a buyer at the deadline because they want to get the right chemistry to try to pull together some wins. If they think that this will help their team and they aren’t giving up too much, doesn’t that sound like fun?” Solid argument.

Grade: A-

“The Team Rebound” (also highlighted in Arnovitz piece)

Concept: “The basic rules: The salary cap space bid would have to exceed the rookie salary slot for the first pick in the draft, which was $4.4 million for 2011-12. The rookie drafted with the “purchased” pick would have a cap number equal to the total amount bid, although he would only receive the $4.4 million rookie salary for the first pick. The amount of the bid which exceeded the rookie’s salary slot would be distributed among the other non-rookie players on the team for the duration of the draftee’s rookie contract, although those players’ cap numbers would not change. Teams willing and able to use additional salary cap space on rookies would jump in line ahead of teams in the traditional draft, forfeiting their own draft pick in the process. Those teams would be required to bid above the maximum rookie salary slot for that pick. So if the first pick in the draft would normally have a salary of $4.4 million, a team would need at least that amount of salary cap space plus whatever additional amount it wanted to bid. Each team could only purchase one draft pick per year using salary cap space. If the team had multiple draft picks, it would forfeit the highest of those picks when signing a player with cap space.”

Advantages:

• Discounts the concept of tanking (throwing games definition)
• Rewards teams that are smart with cap space – which some define as tanking (trading salary away when season looks lost)
• Discourages teams from signing bad long term deals (or, to put another way, penalizes bad GMs)
• Thus, addresses Gladwell’s “moral hazard” concern of “rewarding” bad teams, while still allowing smart GMs to maneuver to get better.

Disadvantages:

• Perhaps is a big harder to “stockpile” picks and take advantage of other GMs
• “The amount of the bid which exceeded the rookie’s salary slot would be distributed among the other non-rookie players on the team” should probably be tweaked – could get complicated and provides unequal “bonuses”.

Grade: A

Some overall comments:

• You can build a great team via: free agency, trades and the draft.
• Some teams have a distinct disadvantage in free agency. Trades are not easy given cap limitations and other CBA rules.
• Thus, the draft is key in a building process for any team that’s not the Lakers, Celtics, Bulls, Heat or doesn’t have billionaire owners willing to spend big.
• A few great small market teams (e.g. San Antonio and OKC) needed a very high draft pick (not exclusively, but it was key) to be great.
• Thus, you need to secure a system where a smart GM can trade off near term results for future success.
• Frankly, we’re partially asking the wrong question: the problem isn’t so much the draft system per se, the problem is how not to reward really bad GMs – while enabling good ones to make smart moves.
• Gladwell does not appreciate the difference in the last two points. E.g. if you have a .500 team early in the season and your best player is lost due to injury (for the year), it probably makes sense to move a good player on an expiring deal. This is just a smart move. A good team is able to acquire a player to potentially “get them to the dance” while you make the smart trade off of near term versus long term investment as an unfortunate injury essentially forced it.
• As an economist by training (I try not to admit this often), I know a sports league like the NBA is much different than say, the auto industry. I’m puzzled why some want to make direct analogies to the “real world” (see “why would we not reward success”). As an example: in the auto industry, companies generally has access to the same raw materials and thus ingenuity, hard work, and marketing should be, and is, rewarded. And part of that reward is market share gains and competition going out of business – leading to further market share gains. In the NBA, not everyone has access to the same “raw materials”. One player can materially skew results. And while you do not want to “reward” bad GMs, you also don’t want a system makes it near impossible to field a good team. A great GM should have a fighting shot to build a winner and that is at least partially done via a high draft pick or two (unless you’re in a premier market).
• This post started at 1/3 this size, but four (at least) “how to fix tanking suggestions” were posted in the last 24 hours and I felt they should be included. Thus, it made for a late night and I’m sure I’ve missed some key points – so comment away below!

A Quick Comment on Pizzagate

This promotion has to be changed. Stan Van Gundy says it best:

How to make it work: Dwane Casey’s coaching style is defense first. Why wouldn’t you leverage that style and alter the promotion to when the Raptors win and hold their opponents under 100 points. (This idea came from the Lakers via @DrewUnga)

Why? It makes much more sense to have that passion displayed for the last 4 min in a 88-84 game than for the last 12 seconds, down 18.

Culture is important. If I’m a free agent I’m looking at money first. But, while it won’t likely “move the needle” materially, I’m not sure the message of “hey we will boo you when the opponent goes on a 10-0 run / won’t try to pick you up / yawn if you take a charge / but… if you’re down 18 and we a chance at a free $2 pizza: Boo yeah!!” resonates well.

Oklahoma City fans know how to get behind their team. The whole game. Let’s copy that.

Really? We got a standing ovation for a slice? They told me they got a free pizza. A slice? A slice you have to sit in your seat and clap; you can’t stand up on a slice, that’s bad etiquette. – Stan Van Gundy

Questions? There is a dedicated to “Statophile Q&A” forum thread here . If you prefer to send questions privately, you’re welcome to email me at tomliston [at] gmail [dot] com or find me on Twitter (@Liston).