Nerdnumbers avatar

Is LaMarcus Aldridge a Star?

On this week's podcast, Patrick and Arturo talked a bit about Aldridge's game in the playoffs. We're a bit leery to call him a star just yet. I was asked about him on Quora. Some shot charts and small sample size points easily explain the playoffs so far.

Read Quote of Andres Alvarez's answer to As a casual NBA fan who loves playoff basketball, what do I need to know about Lamarcus Aldridge? on Quora

How would you go about recognizing if a player like Aldridge is being defended well (and I think different Rockets have been defending him at different levels of quality) but was shooting so hot that he still looked like LeBron?
@Capologist, watch the games! Or watch Durant play at any point in time :)
I do agree he is a terribly inefficient player I think that is made pretty clearly, and I do not believe hes either a plus player on defense or on the boards. But to measure a player's value based off of his production isn't necessarily accurate. Sure he's horrendously inefficient, but his isolation game is one that comes in handy when the defense gets used to your offensive schemes and setups. Also the fact he shoots a terrible percentage does NOT make him a bad shooter, nor does it make him a subpar player. Also, a player's contributions to a team in terms of winning goes much further than the box score and is extremely difficult to measure with a stat like win shares in such a complex game. Lastly win shares has just as much to do with the coaches use of a player rather than the value of a player.
James,

You strung a lot of words together, but they don't seem to add up to a rational position. You call LA 'terribly' and 'horrendously inefficient,' but not a 'subpar player.' You want to divorce production from value--which suggests you're a Kobe or Melo fan, so LA would certainly be your kind of guy--and you broke out that hoary old 'game's too complicated to measure' chestnut. Your scouting report on Aldridge would read ' good shooter who makes a terrible percentage of his shots; massively inefficient, subpar on defense and a below-average rebounder.' His upside? 'When your team gives up on running the offense, he's a not-entirely-terrible isolation option.' To top it off, you referenced win shares, which is not the metric used here. Nice to see you're interested; here's to hoping you will bring more effort to the discussion next time.
LMA is one of those "on the cusp" players. As a 2nd or 3rd option, where teams couldn't double and would occasionally leave him open for jumpers, he'd be fantastic. But as a 1st option that sees doubles constantly, his game is subpar. I think what's happening is that Houston is playing him like he's not a 1st option (not that they play much defense anyway). They have 6'8" Terrence Jones or the 7' but slower Omer Asik on him, and he's eating them alive. LMA's one real benefit to the team is that you have to treat him like a 1st option even if he isn't one, much like Al Jefferson, or he'll go off.

@JamesJamaalWu, this is not win shares.
You really arnt doing much better then someone who evaluates a player based on points scored per game when the entirety of your analysis is "Lol, TS%" or "lol, wins produced". Box score stats dont exist in a vacuum. Zach lowe made the point best a few days ago while advocating the advanced stats movement in Portland so i'll use his words not mine.

" In a macro vacuum, minimizing midrange jumpers is smart. Basketball in real life doesn’t always work that way. Some midrange shooters, including Aldridge, are so good when unguarded that it’s a fine shot — especially against analytically oriented defenses designed to encourage that exact shot.

But more that that, Aldridge’s midranger is the door opener for the juicier analytics shots. Portland’s other bigs get shots at the rim and offensive rebounds when their men rotate toward Aldridge. The Blazers get open 3s when teams send an extra defender at Aldridge’s pick-and-pops and post-ups."

Im not saying LA is a star. Him over dirk as an all-star is a travesty. But in the team concept, for Portland. Aldridge is probably the most important or second most important factor in Portland coming away with the 5th seed in a packed conference with a legit shot at the second round.
The Blazers have an awesome offense despite Lillard being the worst shooting high-volume guard in the league at the rim and Aldridge being an elite high-volume mid-range shooter. These actions open up a ton of space on the weak side for 3-point shooters. The Blazers finally employed a real center, allowing LMA to play the PF full-time- Matthews and Batum had significantly more 3-pointers available and more open than ever in their career, and they took off- the Blazers have a terrible offense without LMA on the floor.

Also, by employing Robin Lopez, they went from one of the worst offensive rebounding teams in the NBA over the last few years to one of the best. That extra space and big man 15 feet away from the basket allows Lopez to fight one-on-one for offensive rebounds (he led the league in Dean Oliver's ORtg by a wiiiiiide margin), and given that they miss a ton of shots, there are a lot of opportunities for the right big man. Obviously I'd rather have Kevin Love, but you can build an elite offense around LMA's midrange game. For most of the season, the Blazers hung around +10 when he was on the court and -10 when he was off the court, because offensive scheme and floor geometry are unsustainable without his particular skill set.
This post isn't going to be much more than me thinking out loud, but something I'm curious about, and something I'm not sure is given enough attention is the variance of a player's performance. Aldridge is someone I would consider a "high-variance" player. On average, LaMarcus Aldridge might be accurately described as overrated. However, what Aldridge does in a given game is what is actually important to a team, not what numbers he averages for the year, and I don't think it can be disputed that this year and indeed throughout Aldridge's career he has had some masterful performances. Should these games be dismissed as outliers when considering the value of a player like Aldridge, or is there an overemphasis on a player's season/career averages, especially when considering to sign or play a player who might have a relatively high probability of "getting hot"



Spoofy - the difference between "yay points" and "yay ts%" is scoring efficiently leads to more wins than scoring more points without regard to the attempts does. I agree with Lowe to a point; it's nice to have a player that can make the midrange jumper, but LMA isn't one of those guys. 189 players took at least 100 midrange jumpers this season. LMA shot 899 on the season (or 13/game which is just stupid) and only made 42.2% of those shots ranking 55th out of the 189 players.

I don't have any concrete proof that LMA isn't the #1 or #2 reason that Portland grabbed the 5 seed, but I would point more towards them starting the season very well due to a favorable schedule, the 3-point barrage of Matthews, Batum, and Lillard, and Lopez being a fantastic offensive rebounder/defender.
Shawn woods - I understand the value of efficiency but the basis of this entire article is basically LA shoots too many mid range J's (He does but there is obviously value in that shot) which is an inefficient shot therefore LA isnt a very good player. It's to narrow minded. it's looking at one stat in a vacuum with no regard to its effect on Portlands offence as a whole which would, in my opinion, be a more effective way to Evaluate LA.

Portland are 9 points per 100 possessions better with LA on the floor, a larger impact then any other Portland player.
The title of the post is "Is LA a Star" not "LA isn't a very good player". I don't mean to put words into Dre's mouth, but I'm guessing if you asked him, he would say LA is a fine player, not a star.

LMA has also played almost 70% of his minutes with ROLO, a terrific offensive rebounder and the guy that takes the tougher of the 2 post assignments. According to those same stats, the Warriors are 2 points better with Lee on the floor than Bogut.
SMALL SAMPLE SIZE ALERT #1
Dre points out that LaMarcus had TS% this season of 50.7%, compared to a league average for PF of 54.5%. He neglects to point out that for his career he has a 53.5% TS %, with a usage rate that far exceeds the league average PF. So cherry pick the data to support your narrative.
If you want a narrative, how about digging into the numbers, and understand why Portland went from 15th in the league in Offensive efficiency (105.8) to 2nd this season (111.5), while having LaMarcus Aldridge increase his usage rate from a career average 24.56% coming into this season, to 29.8% this season, and having his TS% fall from 53.5% to 50.7%. He shot way more shots, and did so at much less efficiency in 2013-14, yet Portland's ORtg improved 5.4% in the process.

SMALL SAMPLE SIZE ALERT #2
Portland and Houston finished the season with identical 54-28 records. Portland had an offensive rating 111.5 and Houston had an offensive rating of 111.0. Portland had a defensive rating of 107.4, Houston, 106.3. In aggregate Houston's ORtg - DRtg was 0.5 points per 100 possessions better than Portland.
We are told to ignore the large sample size of the entire season (wins and Point differential), and instead focus on the fact that Portland regressed to the mean (scientific code for smaller sample size) during February and March, when Aldridge and Freeland were hurt. We are told that Houston was good all season, yet we are supposed to ignore the small sample size at the end of the season when Houston won only 10 of their last 19.
These are the two most evenly matched teams in the playoffs. Three of the 4 have gone to OT, and the one that didn't was decided in the last minute. The large sample size season record, and point differential though should be ignored in favor of the small sample size of mid-February through March.
Aldrige's play has forced the Rockets to change their lineup. At the NBA level a player is considered a Star on a team (usually there are 2 to 3 per team) when the opponents team organizes their defense to handle that player. So Aldrige is a star on the Blazers and Houston's actions have publicly acknowledged that. The more interesting question is how much of a Star is he and how does he compare to Stars on other teams.
@Ziggy

Why would Aldridge shooting more but shooting worse be a good thing? Luckily for him his teammates ended up being much better (there's your story! of course if has been covered here before), but saying that he was worse but the team was better as a defense of Aldridge is just ludicrous.
@Dodgson ,
Don't put words in my mouth. I did not try to justify nor rationalize anything about LaMarcus Aldridge this season. I did not try to defend LaMarcus' poor shooting efficiency, nor his shot selection. At no point did I say, nor imply, nor infer that LaMarcus Aldridge shooting more, but shooting worse to be a good thing.

I did make the point that I find it fascinating that Portland has improved so much offensively, while LaMarcus Aldridge has shot a lot more shots, and done so at a career low efficiency. I happen to think that is far more interesting than whether or not LaMarcus Aldridge is a "Star".
"If you want a narrative, how about digging into the numbers, and understand why Portland went from 15th in the league in Offensive efficiency (105.8) to 2nd this season (111.5)"

It's not particularly difficult. In rough order of importance:

1) Portland went from a middling team on the boards to an outstanding team on the boards. Last year they were a merely average team on the boards; this year, they are crashing the glass and one of the best teams in the league at rebounding. The biggest factor? Robin Lopez, who is absolutely killing it (6.0 ORB/48); the team has also improved in this area across the board (much better DRB numbers, many of which are going to LMA). It may not be intuitive, but good defensive rebounding improves your offensive efficiency by increasing your possessions, and rebounding has, by far, had the biggest effect on Portland's performance.

2) Free throws. I haven't heard this talked about too much, but it's a huge huge difference showing up in the data. Both Matthews (+1.5/48) and Lillard (+2.1/48) are getting to the line a ton more this year, which translates to a couple extra points a game. In addition, they've replaced JJ Hickson (60% FT shooter) with Robin Lopez (80% FT shooting) in their starting lineup, which has pushed the team's FT% up to the top of the league.

3) Improved 3 point shooting. Yes, only 3rd. Lillard and Matthews both had great years from beyond the arc; however, Matthews was a fantastic 3 point shooter last year as well (39+% both years), and while Lillard was sizzling at 39% this year he came in just shy of 37% last year. The big difference this year? Mo Williams. Williams is an above average 3 point shooter, and if you dig in the lineup level data a bit it's clear that unlike last year, Portland's 3 point game is now a 48 minute threat.

All three factors have a large effect size, responsible for well over a point per game in efficiency.
BPS,
First off thanks, though to be honest you are speaking to the choir with me. The points you raise are all very valid. I also like that you acknowledged Mo Williams, because even though parts of his game are inefficient he clearly has made important contributions. So often in the NBA the focus is on stars, but deep bench that can contribute is critical to success. Finding players like Troy Daniels can have a huge impact on teams, and is so often minimized.

Several months back I made a comment about 3 point shooting and offensive rebounds. Portland finished the season in 3rd in team offensive rebounds (essentially tied with Minnesota for second 1025 to 1022). Detroit finished first. Neither Detroit or Minnesota made the playoffs. Detroit shot 32.1% on 3's, Minny 34.1%, and Portland 37.2%. While 3 point % and OffRbs are independent of each other, I do believe that this indicates that there is a non-linear effect on wins relating to both. In other words, not all offensive rebounds are the same, some have a greater impact on wins than others, and I mean that in the sense of two separate and independent teams, and as such the players on those two teams. In other words, I think it is highly probable that, Robin Lopez's offensive rebounds are worth more wins than Andre Drummond's offensive rebounds, because Portland has offensive players that are more able to convert those offensive rebounds into points.

I also want to acknowledge your very pertinent point on FT% as well. All season Portland's actual record has exceeded it's Expected Win/Loss. The conventional wisdom is that it is just luck, and that it will eventually even out. I think there is a component of luck in this, but I also believe there is a component of skill as well. Portland's excellent FT shooting means that they are less likely to lose games where they lead, though I certainly don't know how much less likely. Same thing with being able to make 3's when you trail. Portland was also 5th best in the league in turnovers, meaning they are less likely to turn the ball over in crunch time, whether ahead or behind. All of those impact actual wins differently than expected wins. I believe Portland's superior FT shooting has had a clear impact in it's 3-1 advantage over Houston to this point, especially considering how close this series has been.

Finally I find it very interesting that Portland is dead last in turnovers generated. Their opponents turned the ball over 90 fewer times than last year, and yet Portland's defensive efficiency improved from 109.2 last year to 107.4. Obviously Robin Lopez's shot blocking impacted that, and there are a lot of components to why this happened. At the same time, Portland structured it's defense completely differently this year, and chose to attempt fewer steals, and it's defense clearly improved. Once again a non-linear effect, not all steals have the same value.
A few ideas:

- I wouldn't take it as a given that ORB and 3PT% are independent, or even that ORB% and 3PT% are independent. I would not be surprised at all to learn that there's a significant correlation between the two (I seem to remember an article showing that 3 point shots tend to result in a higher percentage of offensive rebounds than 2 point shots). In a more detailed model you'd want to include that interaction.

- You're right that the marginal value of an offensive rebound depends on the quality of the shooters on your team; if you have a team of efficient shooters, an additional possession is more valuable to you, in terms of expected points, than on a team with poor shooters. This is what the previously mentioned interaction term would capture.

- Teams that shoot a lot of 3's will have a lot of variance in their points per game; if you can turn your 3 point shooting on and off, you'll have more of an opportunity to catch up in games where you are behind (as we saw in the game the other night). If your team has a lot of great free throw shooters (both at getting to the line and making), then you have a low variance, but high efficiency offense - it will make small leads very resistant to hacks. Both have strategic implications, particularly towards the end of games, that aren't accounted for by per minute models that treat all minutes equally, but could potentially have a substantial effect on winning and losing in close games.
BPS,
I did not mean they are independent of each other, in the sense there is no correlation between them, there may be in fact a correlative relationship between them. I just meant how they impact wins, ie wins is dependent upon certain independent variables like 3 pt % and off Rebs, and they have independent correlative relationships to wins.

Sign in to write a comment.