sunsfn 3/25/2005 report Stockton/Nash

[sunsfn]

sunsfn 3/25/2005 report Stockton/Nash

Updated: March 25, 2005, 11:05 AM ET
Stockton's PPR makes Nash's year seem routine


By John Hollinger, ESPN Insider

Ever done something out of habit even though you know it doesn't make any sense?

You're not alone. NBA execs do it, too. For a good example, consider how they evaluate point guards.

Coaches and personnel people almost instinctively look to a player's assist/turnover ratio to check how he's doing. But ask them why they look at assist/turnover ratio, and you'll get lots of blank stares and convoluted answers. Probe further, asking if they think Reggie Miller would make a better point guard than Steve Nash, and you'll quickly get a series of guffaws. But guess who had the better assist/turnover ratio last year?

Yet, in spite of the logical disconnect of a name like Miller's showing up near the top of the list, some still insist on using assist/turnover ratio to gauge point guards. They're getting some confusing information. For instance, this year's leader, for a second straight season, isn't Nash. It isn't Jason Kidd or Baron Davis or Brevin Knight, either. It's (drumroll, please) … Antonio Daniels. Does that mean Daniels is really better than those guys at running a team? Of course not.

What it really means is almost nothing, because assist/turnover ratio is a flawed stat. The problem isn't with "assist" or "turnover," it's with the "ratio."

Using a ratio is faulty for two reasons. First, it assumes assists and turnovers are equal, when in fact a turnover is more costly than an assist is helpful.

Second, it equates very different amounts of productivity. If Player A just sits in the corner all season and finishes with three assists and one turnover, while Player B directs the offense all year and has 300 assists and 101 turnovers, then according to assist/turnover ratio, we should assume that Player A is "better" at running the offense.

Fortunately, there's an easy way to fix assist/turnover ratio. It's a stat I call Pure Point Rating, and it mends the two flaws I mentioned.

First, it adjusts for the fact that assists do less good than turnovers do harm by multiplying assists by two-thirds. There's a factual basis in this. As I noted in a recent column, of the three acts of creating the basket (getting open, making the pass and making the shot), the passer does one. So we give him one-third of the credit of a 2-point basket, or about two-thirds of a point. Since turnovers cost almost exactly one point (teams average about 1.02 points per possession), we needn't make any adjustments to that part of the equation.

The second adjustment is measuring productivity, to avoid the Player A vs. Player B situation above. The way to do this is to sum a player's accomplishments on a per-minute basis, then adjust them for his team's pace. Finally, multiply the end result by 100 to make the numbers more user-friendly. The final equation is:



Pure Point Rating = 100 x (League Pace / Team Pace) x ([(Assists x 2/3) - Turnovers] / Minutes)Using Pure Point Rating instead of assist/turnover ratio yields vastly more believable results. The Reggie Millers and Antonio Danielses disappear, replaced by some names we're used to seeing in discussions of the game's best "pure" point guards.



Pure Point Rating (2004-05 Leaders)PlayerTeamPPR
Steve Nash Phoenix Suns 12.18
Brevin Knight Charlotte Bobcats 12.05
Rick Brunson Los Angeles Clippers 8.03
Marko Jaric Los Angeles Clippers 7.39
Jason Kidd New Jersey Nets 7.23
Jason Hart Charlotte Bobcats 7.12
As we can see, Nash and Knight are miles ahead of the pack in Pure Point Rating, with Nash leading the way.

Considering the lead Nash and Knight have on the rest of the pack, it's reasonable to ask where their seasons rank in historical terms. This is where things get really interesting, because both are in exalted territory in terms of Pure Point Rating. Let's look at some other notable seasons from the past decade, adjusting the pace of each of the previous years to 2004-05 levels.



Top Pure Point Rating 1996-2005 (Min. 1,000 minutes)PlayerYearPPR
John Stockton 1995-96 12.63
Mark Jackson 1997-98 12.49
Steve Nash 2004-05 12.18
John Stockton 1999-2000 12.08
Brevin Knight 2004-05 12.05
Mark Jackson 1998-99 11.98
In terms of Pure Point Rating, only John Stockton and Mark Jackson match what Nash and Knight are doing. However, including Jackson and Knight in the comparison is a bit unfair to Nash. Though infinitely better than assist/turnover ratio, Pure Point Rating does have one fly in the ointment: Because it measures only assists and turnovers, it gives an advantage to non-scorers such as Jackson and Knight. Guards can make turnovers going for their own shots just as easily as they can passing it to somebody else, so those who aren't looking to score will benefit from this measure.

If we limit the discussion to guards who have taken on some kind of an offensive burden – those who average at least 15 points per 40 minutes – then we see how rare Nash's performance has been. It's the best Pure Point Rating by an offensive-minded point guard since Stockton in 1995-96.

Based on that, you might wonder whether it's fair to compare Nash to Stockton. After all, both are West Coast Conference products who grew up in the Pacific Northwest, albeit with slightly different grooming habits. But once we include Stockton's prime years, Nash doesn't measure up. I expanded the chart to go back another decade, showing how much work Nash has to do to make the comparison stick.


Pure Point Rating Offensive-Minded Players 1986-2005
(Min. 1,000 minutes)PlayerYearPPR
John Stockton 1989-90 15.56
John Stockton 1987-88 15.52
John Stockton 1990-91 14.59
John Stockton 1990-91 14.54
John Stockton 1994-95 13.86
John Stockton 1993-94 13.76
John Stockton 1992-93 12.83
John Stockton 1988-89 12.69
John Stockton 1995-96 12.63
Steve Nash 2004-05 12.18

When people say Stockton is the best pure point guard ever to play the game, this is what they're talking about. As you can see in the chart, Nash's season, as great as it has been, barely cracks Stockton's top 10. The Man from Spokane remains in a class by himself.

However, we shouldn't let the Stockton comparison blind us to the genius of Nash's play this year. While it's true he can't touch Stockton's best years, it's equally true that no other scoring point guard can touch what Nash is doing.

But the really incredible part is that all of that information is invisible if we rely on assist/turnover ratio, because it spits out misleading data that makes us think Daniels is better than Nash at running an offense. Thus, out of all the amazing numbers put up by Nash this year – or by Stockton in the two preceding decades – the most telling stat is this: Stockton never led the league in assist/turnover ratio.




---------------

 

[playstation]

well written, and nash's case against stockton is further hurt in that stockton played defense, and did it well. nash is great, stockton is a god...

 

[hafey2]

Riddle me this: an assist which means, by definiton, 2 or even 3 points were scored is worth less than a change of possesion where, on average 1.02 points are scored. It would seem to me that in terms of point production an assist is more valuable than a turnover.

 

[frdbtr]

I disagree that Stockton is a god. Frankly (and living in Utah you can understand that I have almost been stoned to death on a few occations for this opinion) that Stockton is the reason that the Stockton/Malone duo never got a championship. Before you flame me completely off of this board, listen me out. Stockton had an unbelievable shot, he could score any time that he wanted to, but most of the time he was content to just bring the ball up, pass it to malone and chalk up another assist. In the regular season this was fine but when it came to the playoffs, he never raised his play. I remember one year in the playoffs, the Jazz were playing the Sonics in an elimination game. The Sonics led the game by about 20 for most of the game and Stockton had that same expression that he always had until about 2 minutes left in the game (with the outcome already decided) he starts hitting 3's like crazy and brings the Jazz back within about 10 but it was too late. This is just one example but there are many others. I always told my "friends" here in Utah that if Stockton would average 20 pts and 10 assists throughout a playoffs series that the Jazz would win a championship and guess what, he almost did, until they got to the finals. Then he went back to his normal stone faced self and just handed the ball off to Malone for the dirty work. Guess what, now Malone is the goat here for the Jazz's losses in the finals and Stockton still remains above reproach. Stockton was probably the best pure passer that the league has seen but I would probably pick KJ, Magic, or Oscar Robertson if I had a choice of any point guard from history. If I could insure that KJ would stay as healthy as Stockton did (and had to choose between point guards from the same era) I would definately choose KJ over Stockton. Magic would be a no brainer just because of the versatility that he had. This love fest for Stockton doesn't make sense to me, my opinion of him is that he was selfishly unselfish. Sloan has always said that he told Stockton to shoot more and he refused to do it, in my opinion he is completly responsible for the Jazz's failures in the playoffs. Course since I predicted that the Jazz would never win one with Stockton and Malone, I am only too happy that he remained this way, if he had decided to take over some games against Chicago, I might have been proved wrong.

 

[stompg]

Riddle me this: an assist which means, by definiton, 2 or even 3 points were scored is worth less than a change of possesion where, on average 1.02 points are scored. It would seem to me that in terms of point production an assist is more valuable than a turnover.

You need to also add in the loss of points on your side. A turnover not only gives the other team points but results in no points for you. So, where you might have scored a 2 point basket, a turnover that results in the other team scoring 2 points is a four point turnaround. Two that you didn't score and 2 that the other team did.

 

[AZZenny]

an assist which means, by definiton, 2 or even 3 points were scored is worth less than a change of possesion where, on average 1.02 points are scored. It would seem to me that in terms of point production an assist is more valuable than a turnover.

While I like that people are starting to apply Sabremetric analyses to basketball, with that comes the 'seat-of-the-pants' aspect that a real statistician wouldn't accept - guesstimates or assumptions used as shorthand for numerical facts because really analyzing enough data would be a serious pain in the butt. Happens all the time in baseball stats - the famous DIPS formula was seriously flawed because it was based on too small and biased a sample and made some 'logical' but incorrect assumptions. The concept was valid, but the specifics and the initial conclusions were off.

This assists piece looks like one of those things. He decides - based on one logical but simplistic and arbitrary model - that an assist is only credited with 2/3 point, (Delete long erroneous rant-passage)
This also doesn't factor in and weight the percentage that goes to 3 pointers, which would vary by team/year.

You'd have to find a way to weight each guy's assists - perhaps according to how he compares to the league average, so someone like Nash might have assists that count for 1.0 pts this year, if we say the league average is .67.

When you look at that kind of complicating detail, you can understand why these guys take shortcuts (and why baseball has begun to turn from these handheld models, and increasingly to complex computer algorithms). But garbage in, garbage out, and if one of your basic assumptions is flawed, so are the results.

Obviously assists/turnover ratio is a primitive stat, and thus fairly useless. The PPR is an improvement, but it ignores some important data, and ought to come out and say so.

 

[elindholm]

Riddle me this: an assist which means, by definiton, 2 or even 3 points were scored is worth less than a change of possesion where, on average 1.02 points are scored.

The author's argument is that the assist man doesn't get sole credit for the basket. It's sort of like in the NFL when they credit "half a sack." Yes, the assist leads to a 2- or 3-point basket, but the guy who actually makes the shot has to get some of the credit too. (According to the author, so does the guy who gets the shooter open, although I think that's stretch. Also, since he's in to weighting everything, he should figure the average made basket to count more than two points, by a percentage reflecting the average ratio of made threes -- ideally on a team-by-team basis. But never mind.) A turnover, on the other hand, is pretty much the sole responsiblity of the guy who coughs it up.

So, where you might have scored a 2 point basket, a turnover that results in the other team scoring 2 points is a four point turnaround.

Not really. The other team was going to get the ball back anyway, so the only thing you lose through a turnover is your own posession. It would be helpful to break down turnovers into those that lead to quick easy baskets and those that don't, but the stats don't do that.

I disagree that Stockton is a god... (long interesting analysis deleted)

This gets back to the general fallacy of the "pure point guard" as being a player who is somehow absolved of the responsibility to score. As I've written before, such narrowly defined roles might work at the high-school level, but in the NBA, everyone has to be a scoring threat in order to be a great player. We've seen time and again how Kidd has cost his team victories through his inability to shoot, even though he's allegedly this awesome point guard who "unselfishly makes his teammates better." Before the Spurs won their 1999 title, it was often said that they'd never get there until Avery Johnson could start hitting open shots. Eventually he did, and they won.

I didn't watch the Jazz that carefully, but I know from the stats (and the games that I did see) that Stockton was a superb shooter, probably one of the best of his era. Maybe frdbtr is right and Stockton's reluctance to score, even though he could when he wanted to, ended up holding his team back.

 

[elindholm]

So let's say 7 out of 10 of Nash's assists this year result in 2 pointers, because one of his special talents is that he passes the ball to where guys can shoot effectively. If you compare him to someone else, maybe a Marbury, who lacks that quality, maybe you find only 5/10 of Steph's assists earn a 2 point shot.

An assist results in a made basket by definition, so this particular criticism isn't valid. Overall, however, I agree with you that the article substitutes hand-waving for legitimate analysis.

 

[AZZenny]

Ooops. Sorry - Got on my statistical rant and shot first, aimed later.

Sincerely,

Emily Litella.


But yeah, saying its only 2/3 point is not necessarily a valid assumption.

 

[sunsfn]

I would like to comment on the Stockton played Defense.

Stockton could never guard KJ when the suns played them, I mean never!
I have seen many Utah games over the year and Stockton got away with more fouls than Michael Jordon. He would grab jersies and hold on many plays during the game. For some reason the refs let him do this all the time. Oh they would call a few every game, but if he held 10 times they would call it a couple times.

Stockton was slow afoot and could not guard the fast players. Nash has the same problem.

Stockton was a great NBA player and deserves to be treated as such, but saying he played good defense is quite a reach.

 

[JPlay]

I don't remember Stockton being very clutch, I do remember that he was known for being one of the most dirty players in the NBA.

Also you have to look at assists for dunks. I would argue that Nash is probably having one of the best seasons ever for a PG in that regard. Many times stockton passed to Malone and he took a pull up J.

 

[George O'Brien]

I'm not really sure that assists and turnover are really useful stats because they lump together too many things and don't include others.

1. A pass to a wide open catch and shoot guy gets the same value as a pin point pass to an inside guy for a layup. The first is something almost anyone can do unless blind. The second is an art.

2. Turnovers that lead to breakaways are far worse than balls that get thrown out of bounds.

3. Turnovers that result from not protecting the ball while dribbling is clearly the player's fault, but when a pass gets fumbled by the other player it is not clear who is at fault. If Amare can catch balls that Hunter can't, does that make Nash a better point guard merely because he is playing with Amare?

4. Indirect offense does not produce assists, but may be more valuable. Nash passes to Amare who hits a cutting Marion. Amare gets the assist, but it was the entry pass by Nash that created the play initially.

In the end there is only one stat that coulds - W's. That's where Nash is a clear number one.

 

[seattle_sunsfan]

A turnover, in my mind, is more costly than this writer gives credit. The writer assumes that the turnover gives the other team a standard possession where they would score 1.02 points. I'll bet anything, it is far worse than this. Picture Nash losing the ball in the lane, and having three defenders race down the floor immediately in the other direction for a layup. Nash's team simply cannot get back in time to defend. This would probably hurt Nash's rating because he is so turnover-prone. But he is obviously a fantastic point guard. So there must be a more correct way to analyze this. I agree that assist to turnover ratio by itself is flawed.

 

[JPlay]

huh, an assist is a guaranteed score, where a turnover is not. I put much more weight on the assist.

 

[Gaddabout]

Assist stats are predicated on subjective opinion as they are collected, so it doesn't surprise me nobody has a great way to measure ball distribution on an individual basis. If Nash delivers a bounce pass from the halfcourt line to Stoudemire all alone under the basket, he gets the same credit as Luke Ridnour tossing the ball to Ray Allen who regularly hits contested three pointers. When is an assist an assist? And how many great passes have been blown by stone-handed big men who couldn't shoot?

Assists aren't as subjective as, say, QB hurries in the NFL (that's a BS stat), but you'll be hard pressed to get me into a heated debate over how to account for them. I've sat next to too many scorekeepers who give their point guards way too much benefit of the doubt to say it's a good science. I feel confident reasonably intelligent people can measure good point guards with their eyes. If you didn't know Steve Nash was a better passer than Antonio Daniels in five minutes of watching them, you've got no business putting your nose in stats and telling someone otherwise.

 

[jibikao]

Stats are very interesting to look at but Nash's offense is more than just assist/turn over.

When Nash is on the court, we have "fluent" offense. Everyone knows what he is supposed to do. Even if the pass doesn't end up an assist, Nash's leadership and floor vision will still make sure our offense runs smoothly. That's why Nash is the Top PG this year.

Nash is just like Stockton, who directs everything on the court and even sets picks for big guys.