Game Thread GAME# 33: LAKERS @ BLAZERS - DECEMBER 28, 2019 - SATURDAY, 7:00PM, NBCSNW

CJ has an ISO eFG of 46.5% and scoring percentage of 45.5%. Neither are great. Good call @GrandpaBlaze !

 

I generally stay out of the Stotts debate. For the most part, I think the argument that Portland shouldn't fire Stotts because the replacement could be worse is a hollow argument. I think it was a mistake to extend his contract, but it didn't bother me much. He's played the hands he's been dealt, and some of them have been bad hands....like this season. But the way he's played the hands haven't impressed me either. I think he's a weak defensive coach and his offense is kind of remedial

but I do think the roster has major holes and weaknesses and those aren't the fault of Stotts. And making CJ a featured player is one of those weaknesses

personally, I'd really like to see the house cleaned. Olshey, Stotts and CJ all gone. Get some new thinking while rebuilding the roster and the offense

 

I don't like eFG% as a gauge. It does not account for FT's and a shooting stat that doesn't account for FT's is fatally flawed

 

Hardens PPP on ISO is 1.18 vs Dame 1.14. Not sure if that refers to those player’s individual PPP or the team’s PPP when that player is ISOing. Hopefully the latter. Don’t know what gives but last year Dames ISO PPP was a horrid 0.87 which seems too low to be right.

I have the data to back up the scoring rates after a missed shot from various distances. I always suspected that “3pt shot is bad for your D due to long rebound” was BS just from watching since your team has added seconds to get back when the shot is in the air. And watching 5 on 4s after missed layups and player on his back under the basket. So I got the raw data from NBA play by play and tested it out. I forget the actual numbers but missed near the basket are the worst. By how much, I don’t remember. But I could dig it up.

 

Agreed! Point per possession would be the most ideal stat, but it would require a lot of math on my end. Dame has a free throw frequency of 15.9% on ISO compared to CJ's 9.1%, so that would up Dame's effectiveness even more considering he shoots 90% from the line.

 

I couldn't find PPP on the site I was using, if you don't mind sharing the link I'd love to see it. Great find, I love this stuff!

I'm assuming 1.14 PPP is higher than our team average?

 

So ideally we'd find this shot chart for Dame's ISOs and his misses in the paint would be more of a negative than a three? When you ran these numbers, was this league wide or just for Portland? Some teams are much more aggressive on the O-boards which I assume would impact that number as well.

 

Very likely higher. But last years 0.87 isn’t. Those two numbers seem too far apart. Seems suspect. This year is something like 130 Dame ISO plays. Not exactly a small sample.

https://stats.nba.com/players/isola...ir=1&SeasonType=Regular Season&PerMode=Totals

 

I ran league wide for multiple seasons. Running one player or one team won’t be a big enough sample size.

 

This is a CRAZY difference for Dame. He's not under 1.00 PPP on ISOs in any other season (only goes back to 2015).

 

Yeah I figured it out. That stat and link I gave you I think is useless. It’s Dames and Hardens individual scoring on that play type. What it should be is teams scoring in that play type run by Dame. What if Dame kicks out for a 3 every time.

 

https://stats.nba.com/players/isolation/?sort=TEAM_ABBREVIATION&dir=1

NBA.com breaks down different play types and tallies PPP. You can sort by team to get to Portland.

Hood is actually the best Blazer at 1.42 ppp which is in the 99th percentile. I'm assuming those numbers account for FT's. Now Hood only has 11% of his possession as iso, and he was low usage so there's a little bit of a sample size issue with that crazy number of his. I think you also have to somehow gauge for defensive attention. I see it every game that Dame has much more defensive resistance than CJ does. CJ often will drive around screens and picks and see an open lane. It might not extend all the way to the hoop but it's regularly open into the mid-range. But in the same games against the same teams when Dame rolls around a screen there's almost always another defender waiting. Defenses are stacked against Dame and CJ often drafts off of the attention paid Dame. That will skew some numbers

I also question how NBA.com's data is categorized into play types. Some of the play types they track are iso, PnR ball-handler, off-screen, hand-off, and cut. Those types of plays can look like distinctions without a difference for a team like Portland that runs so much one-on-one action

 

Absolutely... I was actually wondering if last year was so low because Aminu/Harkless were bricking so many thress. So much for that theory!

 

I spent a couple years logging games for Synergy Sports and as you proved your ability to accurately log games, your level would move up. Level 1 would just identify when a possession started/ended. Next guy would do who was on the floor, next guy outcome, result, and eventually play type. While you have to watch a lot of film to get to the play identification level, it certainly isn't an exact science. Even knowing the play call wouldn't promise you would log the action right and sometimes knowing the play call would create a bias in your ability to make a determination.

 

Yeah welll teams probably know what they are doing. Dame is much more efficient than CJ so they pay way more attention to him.

Yes, if they don’t categorize consistently that would skew the data. I’m gonna look at that page some more to see if there is any way to get better numbers. Maybe the raw data I have has play type. I’ll check. NBA data is a pain in the ass to deal with. It’s kind of a mess and some voodoo is needed to even break the plays down into minor and major possessions.

 

Cool sounds like a fun job!

 

It didn't take too long for it to start to feel like work! What was cool is you would have access to the film library and all the games.

For example you could query all Dame/LMA PnR on the left side of the court where the defense jumps and results in a turnover. One click and there's a list of every clip what that exact outcome happened. It's no wonder NBA and college teams loved it. Advanced scouts probably hate it though!

 

Yeah that’s the case with all jobs I guess. That sounds really cool though. The stuff I got was all from NBA stats site and I had to write a .Net program to download, parse and query it. I just checked NBA Stuffer which has similar data but more cleansed and morphed into possessions. They don’t have play type so that means I don’t either since both are based on the same event data.

 

It's almost like he lacks confidence because he's not being used correctly. Terry should be encouraging him to shoot 3s, he's a good shooter, but I've never audibly heard him do so (remember how we'd hear him yell at Meyers to shoot?). A lot of times, Skal gets the ball up top with space and immediately looks for a guard to get the ball to, just like a Whiteside or Steven Adams type would.