LMA = Least improved?!?

by Philthyanimal on May 14, 2008 2:34 AM PDT Comment 25 comments

I found this from the Wages of Wins Journal site

http://dberri.wordpress.com/2008/04/29/monta-ellis-this-time-and-not-lemarcus-aldridge/

The media’s vote indicated that LaMarcus Aldridge was the third most improved player. But if we look at the 25 most improved, Aldridge’s name is nowhere to be found.

We do find Aldridge’s name, though, in Table Two.

Table Two: The Unchanged in 2007-08

Table Two reports the players who changed the least from 2006-07 to 2007-08. The 20th player on this list is Aldridge. Yes, a player who members of the media think is the most improved really hasn’t changed much at all.

At least, not in the aggregate. There were some changes in Aldridge’s performance if we look at the individual stats.

Table Three: Evaluating LaMarcus Aldridge

As Table Three indicates, Aldridge did manage to take more shots this season. And despite a lower level of shooting efficiency, he was able to score more points. Of course, if you score less efficiently, you are not helping as much as you could.

In addition to scoring, Aldridge also posted an increase in assists and personal fouls. But he rebounded slightly less and committed a few more turnovers. The net effect was a very small change in Win Score per minute, and as noted, WP48.

Of course the media doesn’t look at Win Score or Wins Produced. In fact, they don’t appear to look at shooting efficiency or per-minute stats. What they did see is that Aldridge increased his per-game scoring from 9.0 to 17.8. And apparently, that was enough to get Aldridge some votes.

The Least Improved

At least, Aldridge did improve a bit. A few players didn’t quite fare this well. Table Four reports the least improved players - or the players who posted the biggest declines - in 2007-08.

0 recs | Comment 25 comments

Story-email Email Printer Print

Comments

Better trade him then

Before other GMs catch wind of how bad LMA is.

I like PER, most new agey stats are fine, but I just don’t trust this Wages of Wins stuff. It always seems kooky to me. I ain’t no four-eyed expert on the matter though, but my heart tells me it’s buptkis.

I guess it isn’t a surprise he wouldn’t improve much on this sort of stat; his per-40 numbers stayed almost the same, didn’t they? Of course, that’s a great feat by itself, as most people don’t keep the same production when getting their minutes massively expanded. Plus it was his rookie year 2.0, so I’m happy with him “just” playing as good as his per minute stats from his shortened real rookie season.

Statheads out there in Blazer land: What do you think of this Wages of Wins stuff? Is it full of gross weiners, or beautiful lady parts? Is it ultra-delicious chicken thighs, or gag inducing liver n’ poop? Is it smelly like an uncleaned litter box, or like a deep whiff of Oden’s legendary Championship destiny in smell form?

Whenever True Hoop discusses it, it tends to be off, and lucky when it’s right. I think it said Dan Dickau was the best player in the NBA too, and that Antoine Walker is the most efficient scorer, so maybe Berri didn’t carry the 4 or misplaced a decimal point. That’s math stuff, right?

Mortimer

by Mortimer on May 14, 2008 2:59 AM PDT reply actions 0 recs

I think it's Do-Do

Extrapolating an indvidual’s performance from within the team system based on raw statistics is highly problematic. More often than not the stats are used to confirm perception rather than form it. In this particular case they certainly don’t take into account that the more player plays the more the opponent learns to counter his play. In fact, the dynamics of opponents adjustment over time, the quality of opposition, the timing of the opposition, etc., I feel, cannot be included in the calculations.

by LaughingJon on May 14, 2008 8:42 AM PDT up reply actions 0 recs

I don't like Wages of Wins

The authors seem to be really arrogant and think that their winscore is some sort of perfect stat that tells them everything that they need to know about a basketball player. The reality is that there are some thing (such as defense) that are difficult to measure with any system that uses weighted box score statistics, and even compared to other linear weights systems winscore isn’t very good.

Winscore undervalues scoring. It treats shot attempts the same way that is treats turnovers, ignoring that it is possible to get offensive rebounds on missed shots. It also ignores that volume scorers tend to increase the efficiency of their teammates by drawing the defense’s attention and making it so the other players don’t need to try to force up bad shots.

Winscore greatly overvalues rebounding. It thinks that getting a rebound is as valuable as getting a steal, ignoring that a steal is taking a ball away from the other team, while a rebound is just picking up a loose ball that might have been picked up by the players teammates anyway.

The usefulness of a system of linear weights like winscore can be judged by how closely it correlates with adjusted +/- data. I have read a study that did this type of analysis and found that winscore did pretty poorly. PER isn’t perfect either, but it did correlate significantly better with adjusted +/- data then winscore did thus should be considered a more accurate measure of a player’s performance.

Aldridge PER 2006-2007: 17.2
Aldridge PER 2007-2008: 18.5

So judging by the stats there was significant improvement, though not enough to warrant being a serious candidate for the most improved player award. Most of the improvement in per game stats comes from playing more minutes.

While it may seem logical that a player would need to improve significantly to get similar per 40 stats while playing more minutes, every study I have seen that tried to measure this has found the opposite: most players play at least as well when they get a lot of minutes as when they don’t, and it is actually common to see improvement in per minute stats when players get more minutes.

by trk on May 14, 2008 11:21 AM PDT up reply actions 0 recs

Adjusted +/- vs. Wins Produced

The usefulness of a system of linear weights like winscore can be judged by how closely it correlates with adjusted +/- data.

Since adjusted +/- data is “noisy” and sometimes ends up with players like Chris Paul having an adjusted +/- that is close to zero (plus the error), why should it be the metric that all other stats should be correlated to? Wouldn’t a stat that correlates to wins be a better choice?

by tingeyga on May 14, 2008 3:57 PM PDT up reply actions 0 recs

Large sample size

The noise can give you significant errors if you are just trying to figure out how good one player is, but if you are looking data from hundreds of players taken over several years the errors tend to cancel out.

http://www.uncg.edu/eco/rosenbaum/nessis.pdf

This is a good (somewhat long) study of different metrics and how well they predict player performance.

by trk on May 14, 2008 4:29 PM PDT up reply actions 0 recs

Basketball Prospectus

In the basketball prospectus by Hollinger he states that when players get more minutes they tend to increase their PER rather than having it stay constant or decrease.

One thing that might account for the seeming regression of LMA is that his role in 06-07 v 07-08 was worlds apart as he took over Zach’s role as main low post threat.

Life is exhausting when you are this stupid.

by jonestr on May 14, 2008 8:10 PM PDT up reply actions 0 recs

True

The Win Production metric values rebounding and shooting efficiency more than any other player metric that I am familar with. Thus if you become more of a volume scorer and rebound less your WP is going drop.

by tingeyga on May 14, 2008 9:58 PM PDT up reply actions 0 recs

Good News?

I kind of think it is… Who wouldn’t want a player who ‘Didn’t Change’ but contributed heavily to a +9 increase in wins from the previous season and had the biggest point per game increase on the entire team(+8), while seemingly going unheralded outside of Portland?

I DO!! I DO!!

I say keep him under the radar, but give him his dues at the same time.

Juan Dixon Owes Me $5 Dollars.

by Outlaw is Rejector on May 14, 2008 3:02 AM PDT reply actions 0 recs

Read my comments on the article regarding degree of difficulty

I post under my first name ~Nathan

Blazer's fan since '84, Currently exiled in Tennessee and North Carolina

by HurraKane212 on May 14, 2008 4:18 AM PDT reply actions 0 recs

I agree 100% with this analysis

Aldridge’s improved stats were mostly due to increased playing time. For much of the year, he was the same player he was in the 2nd half of his rookie year. Toward the end of year 2, he started to show some marked improvement, but not enough to show up in the sabermetrics.

Wages of Wins is fascinating and most certainly NOT buptkis.

Boomshakalaka

by jksnake99 on May 14, 2008 8:39 AM PDT reply actions 0 recs

IMO

this understates LMA’s improvement. Yeah, he got more minutes and enjoyed a corresponding spike in his statistics. He also played the entire season as our only semi-legitimate post threat, without a second competent true big on offense, and went from afterthought rookie to our second-leading scorer and having other teams game-plan against him on a nightly basis. And as Mortimer mentions above, he experienced his successes this season after a highly truncated rookie year (and playing in a very different offensive system to boot).

Given all that, I think it’s awesome that his per-40 stats increased at all. That’s my primary objection to these new statistical evaluative measures like WoW—they crunch numbers in a vacuum without taking into account non-numerical factors that are also relevant, then present their findings as if those factors had no influence whatsoever.

by BlazersOrBust on May 14, 2008 8:59 AM PDT up reply actions 0 recs

great take

I especially agree with this: “they crunch numbers in a vacuum without taking into account non-numerical factors that are also relevant.”

I agree WoW underestimates LMA’s improvement, but I also think it indicates that the voters probobly overestimated his improvement to some degree.

Boomshakalaka

by jksnake99 on May 14, 2008 12:27 PM PDT up reply actions 0 recs

Haha

I watched 50-something games this season. LMA improved immeasurably. Of course part of it is due to increased playing time, that’s basic logic, not a revelation.

by leeroyjenkins on May 14, 2008 8:41 AM PDT reply actions 0 recs

Perfect Example of OVER-ANALYSIS....

Look BRO…

1.) Young player stepping into starting line-up for the first time, getting a huge bump in minutes per game and total games played over his rookie year.

2.) Young player being thrust into a primary scoring role for a team that really had no other place to turn. Coach would telling him to shoot more, push it more, get to the line, and all the other encouragements that would come along with a coach trying to develop the right mind-set for a young key player.

3.) Statistics don’t tell the truth, they are manipulated and isolate occurrences rather than looking at the overall picture. Overall picture is 17ppg and 8rpg over 9ppg and 5rpg. Also in the overall picture is 35 minutes per game over 22 minutes per game.

If you want to breakdown efficiencies… do it with a Player who has started more than one full season.

Least Improved is like the worst call I have ever heard…. well other than Phithlyanimal being a decent poster… that is also a Super-weak call!

by Portland Dynasty on May 14, 2008 8:59 AM PDT reply actions 0 recs

yo

I just posted what I saw so that it can be up for debate…in no way do I agree with it.

by Philthyanimal on May 14, 2008 9:13 AM PDT up reply actions 0 recs

In that case...

Consider this a retraction of my comment about you not being a good poster on B-edge….

Uh, I guess my homer-tendancies get defensive quick….

But… whoever was claiming Aldridge didn’t improve is a complete idiot. Right? Philthy’s with me!

by Portland Dynasty on May 14, 2008 12:00 PM PDT up reply actions 0 recs

Don't mess with Phil

He is on the Blazer fan advisory board. He has the power to call Greg Oden, and make him dunk on you.

by Sabonis4Ever on May 14, 2008 2:52 PM PDT up reply actions 0 recs

isolate occurences?

That is the last thing that statistics do. There are all kinds of analysis that tries to take into account the effects of small sample sizes so that small sample sizes are not weighted excessively.

Life is exhausting when you are this stupid.

by jonestr on May 14, 2008 8:17 PM PDT up reply actions 0 recs

They also say that LMA is a below average player

My eyes tell me they’re wrong.

BINGO, BANGO, BONGO

by blzrfan on May 14, 2008 9:58 AM PDT reply actions 0 recs

He's a below average rebounder though which is disappointing

Hope he can improve his rebound rate next year.

BINGO, BANGO, BONGO

by blzrfan on May 14, 2008 10:19 AM PDT up reply actions 0 recs

Attributed to Mark Twain

“There are three kinds of lies: lies, damned lies, and statistics”

“Statistics are wrong 50% of the time” sung in the style of Yogi Berra

"Be wary of strong drink. It can make you shoot at tax collectors... and miss" Robert A. Heinlein

by 92wastheyear on May 14, 2008 11:02 AM PDT reply actions 0 recs

How many double-teams did LMA face during his rookie season?

The answer: practically none. This past season, when LMA was single-covered, he was practically unguardable. When he was doubled, that created opportunites for his teammates. Typically, the stats don’t tell the real story. LMA was greatly improved in his sophmore season. Not that he was chopped liver by the end of his rookie year; the media just didn’t notice.

"We don't back down to nobody." --Joel Przybilla

by hurryup09 on May 14, 2008 11:31 AM PDT reply actions 0 recs

ya

his stats may not have improved, but you cant deny that his skills haven’t. he went from a rookie to a first offensive option that commands double teams with no drop off in production. hes shown improvement in toughness and post scoring making us a little bit less 1 dimensional.

by Philthyanimal on May 14, 2008 1:33 PM PDT up reply actions 0 recs

A silver lining to Oden's injury

...was that LMA got a season’s worth of practice in dealing with double-teams. (Roy also benefitted in that way.) Re/ LMA’s toughness, I think he’ll continue to show growth. He was such a beanpole when he entered the league-nearly in Durant’s category. Last season, he was noticably bigger and stronger, and that trend should continue as LMA continues to fill out and hit the weight room. This season I expect him to be more effective posting up, rebounding, and defending. If he works on his ball-handling this off-season-as he said he intended to do-he’ll truly be a load in ‘08‘09. Pair an upgraded LMA with Greg Oden and look out!

"We don't back down to nobody." --Joel Przybilla

by hurryup09 on May 14, 2008 1:46 PM PDT up reply actions 0 recs