dougthonus wrote:...
I've been away from a computer for the last few days. Let me address your post:
This seems like you have taken really small sample sizes to try and extract positive movement.
The overall trends I discussed in his shooting is from his overall season numbers. I did narrow it down to show you the differences between the sets but when I compare 2018 to 2019 to 2020, those are overall numbers. That's the data set we have.
Overall, he shot below the league average from the three point line this year. You remove data you don't like by saying it doesn't count because he's hurt and then extrapolate out the positive data. Given he's below league average overall, it seems pretty likely going through the same exercise of finding reasons to exclude negative data on other players would also raise their averages considerably.
It's been three years in the league, it's hard for me to ignore: 36.2%, 36.1%, 34.4% as his numbers from the three point line and pretend he's improving. If you want to say the injuries caused him to lower his percentages, sure, probably by about 2%.
Okay, again we come back to analyzing timeseries data. Using overall averages is flawed. You keep going back to overall numbers but again it is more nuanced than that. You can get the same overall numbers with radically different performances.
A simple (and current) example of this is the stock market. When looking at trends, you don't just put out the average and call it a day. You do this with the stock market and you'll lose a lot of money. The current average for nasdaq composite is 8,92.47. Does that number tell you about the record highs at the beginning of the year? Does it tell you anything about the covid influenced dip (analogous to an injury performance dip)? Does it tell you about the recovery in recent weeks to back to what it was before? No. it doesn't. All it tells is a jumping off point.
As for the bolded, I am not "removing the data". It's called organizing it into subsets which happens when a set of data has significantly different subsets. I am not picking and choosing negative games to remove. I am detrending the data. That's how you approach data analysis. The "injured" subset is still valid data. The point of it is a branching point. You can ask the question "Is Lauri injured?" If he is, he falls into the injured subset (which is trash performance) and if he isn't he falls into the healthy subset (which is usually good to great performance)
I've said this before: It's perfectly valid to be on the pessimistic side and say "Lauri is always injured so we can only expect 11 ppg on 30% 3pt shooting from him". I choose to take the optimistic take and say "If only Lauri can remain healthy, he could be at 20 ppg at 40% 3pt shooting". <- those become points of opinion, and everyone has their own. It's just disingenuous to label him at 35% and call it a day.
As for applying the same strategy to other players, I want to! If you can remove the injured subset from all the players, it would give a more accurate view of "true shooting" ability. It's the same reason why we discount Lavine's 2017-2018 poor shooting display, it was injury recovery from an ACL. It doesn't get lumped with his last two years. Zach isn't a 36% 3 point shooter (average of 34 during 2018 and 38% last year) He is a 38% shooter when healthy and 34% when coming back from injury. (I think Zach's true shooting ability is much higher easily 40%+ but he takes too many tough ass shots...that's a different discussion though)
Another player that I did apply this to was Jimmy. 2013-2014 he played through the turf toe injury. Many people wrote him off because he shot 29% from 3 that year. But it was pretty obvious the influence of the turf toe. Same thing is happening for Lauri. Except this time around, we didn't find out about Lauri's injury until after the fact (after the team was already imploding) which locked in people's opinion of him as "regression".
The only problem with this method is there isn't an easily managed dataset to sort through. I can do this for the bulls because I try to closely watch all 82 bulls games. I promise you analytics departments look at the data they have in the same way also instead of overall averages.
The other thing is his three point shot type:
https://stats.nba.com/player/1628374/shots-dash/95.6% of his 3s are catch and shoot, assisted, 0 dribbles, which shows very little versatility. An elite three point shooter can shoot threes off the dribble and shoot in tight spaces.
56% of his attempts are considered wide open
38% of his attempt are considered open
6% of his attempts are considered tight
0% of his attempts are considered very tight
As far as "An elite three point shooter can shoot threes off the dribble and shoot in tight spaces." That is true. I guess if that's your qualifications for an elite 3pt shooter, then yeah Lauri is a bit off from that. However, there are only like 10 players (if that many) who qualify for that (and only 1 big man in KAT). Those kind of shots are tough to take and make. Most teams don't condone shots like that (we need Lavine to reduce those shots).
My qualifications for an elite 3pt shooter is more of the traditional sense a la Korver/Ray Allen/Klay. These guys wouldn't qualify with your "elite shooting" criteria. Compared to the traditional guys, I do think Lauri is "almost there".
For the shot distribution, you see that as a negative, I see it as a positive. Wouldn't you want your player to take open shots and avoid hard/contested shots? Do you want to see Lauri start chucking instead of passing it to the open man?
What you SHOULD have listed is his shooting percentages on open vs tight shots. Lauri regressed on open shots (overall) and I agree. It was disappointing. That's when you dig into the numbers more and see the injury trend.
This is a different discussion as you state. I think it is disingenuous to compare him to KAT because KAT just had the best shooting season for a big man
ever. He is the gold standard to get to...not the average. I do think Lauri can match it if he was the focal point that KAT is on his team but let's let Lauri get back to 15 fga per game before we start discussing best seasons ever. But already, your link shows even with shooting troubles/injuries/whatever excuse you want to make, Lauri is still top 3 in percentages for big men (and true shooting ability is a lot higher when healthy).
Anyway, back to the assisted vs unassisted rate. You need to break that down into 2 point shots vs 3 point shots. Lauri took a lot more 3s compared to KAT (shot distribution wise) so his assisted rate will automatically be a lot higher:
2 point assisted -- unassisted:
KAT: 50.5 -- 49.5
KP: 69.5 -- 30.5
Lauri: 53.5 -- 46.5
Dirk: 59 -- 41 (2011 championship year)
Bosh: 64 -- 36 (2012 championship year)
3 point assisted -- unassisted:
KAT: 81.6 -- 18.4
KP: 96.8 -- 3.2
Lauri: 99.1 -- 0.9
Dirk: 97 -- 3 (2011 championship year)
Bosh: 100 -- 0 (2012 championship year)
I think you are unfairly holding Lauri to a much higher standard than he should be held at. Like I said above, KAT is a gold standard but we shouldn't look down at Lauri if he can't match his shot creation abilities. Lauri is actually closer to KAT with his shot creation vs KP.
His assisted rates compare pretty favorably to other big men. It's his USAGE which was wayyy down. Which again, the reason it was down was due to injury. That's why I split the shot charts up between november and december to show a little bit more nuance.
You don't want to give a hurt/soft/passive Lauri who is finishing at 41% at the rim more shots. That's understandable. But when a dude is at 70% finishing at the rim, he should be utilized more. You can't say that's just a hot/cold streak. This was a clear distinction. And players don't have a 30% difference in finishing at the rim during hot/cold streaks.
Outside of actually shooting much better, the most useful thing Lauri could do would be to become a solid defensive center, he'd then provide a huge competitive advantage with his shooting. At PF, I'm not sure he really provides any competitive advantage vs starting a guy like Otto Porter at PF.
I strongly disagree with this. Playing Lauri at C is setting him up to fail. He has the defensive fundementals to play good defense but he doesn't have the strength for the thic bois or the length for the long bois. A "healthy" Lauri who isn't stiff hobbling around the court can play positive defense at the PF position.
His most value will come from consistent shooting (pray for health) and his bread and butter is pick and pop drives (which Bulls went away from this season). On those drives, he can bully smaller players and outrun faster players. He won't be a traditional post up threat (but that is extinct anyway).
But again, if history repeats itself, and Lauri gets injured and relegated into a spacer role (during which he will shoot poorly due to injury) then yeah, Lauri is not worth it at all.
---
Finally one last chart to illustrate my point of "I think he got better as a shooter this year":
This is overall data including injury data from all his seasons. You don't have the excuse of "taking out negative data".
CT on this graph is corners + top of key 3pt attempts. year to year he improved his efficiency on those attempts. His next steps are to improve the wings. The left wing seems stable but do you see the outlier on the right wing? Do you expect him to continue to shoot 25% from there?
Then you can break it down further and compare healthy Lauri vs injured Lauri trends year to year.
Healthy Lauri went roughly 36 - 38 - 40 -> a consistent trend upwards including his troubles with the wing
Injured Lauri was consistent...consistently bad... roughly 30-34%. That's also a pretty clear trend.