Not to mention the conditioning. You can probably play with more intensity when you are only out there for 20 minutes a game. You don't have to worry about foul trouble either.
The 2017-2018 Celtics Will Be a Bizarre Monstrosity
- Thread starter slamminsammya
- Start date
Not to mention the conditioning. You can probably play with more intensity when you are only out there for 20 minutes a game. You don't have to worry about foul trouble either.
Amir and Olynyk excel in advanced stats due to Brad's ability to spot them in advantageous matchups while not overexposing them. Their time on the floor was limited to these positive matchups and in Olynyk's case against 2nd units to avoid the length/athleticism which he struggles against. Morris was a full-time starting-4 who was on the floor regardless of matchup.......had he been utilized in the same manner as Amir and Olynyk his advanced numbers would look much better as he would not have been on the floor in the bad matchups. Yes, Amir started but his role was very limited.....we have seen many years of limited impact 4's playing part-time minutes while grading out erroneously using certain metrics. Call it the Chuck Hayes Rule.
Taking advanced stats out of context results in grading Amir/Olynyk over many players including Morris.
Edit: PKB beat me to it.
Adam Kaufman was on NBA Radio and the host thought that a better route for the Celtics this offseason would have been:
1-Drafting Ball
2-Trading for Paul George
Who are these people that get radio shows?
Kaufman essentially said "Dude, you can't blame Ainge because the Pacers guy was an idiot."
1-Drafting Ball
2-Trading for Paul George
Who are these people that get radio shows?
Kaufman essentially said "Dude, you can't blame Ainge because the Pacers guy was an idiot."
I agree that coaching plays a huge part in these analyses, but this is simply wrong with respect to Amir. As Bowiac already mentioned, he has been a stats darling across different teams and coaches.
This brings us back to a point that someone mentioned upthread. I love Amir, and I think that he was a very useful player last year until injuries took him. But any stat that has him top 25 among all NBA players is FUBAR.
We could take this in a philosophical direction about how statistical models should not be created simply to reinforce our preconceived notions of something - that unusual results are one of the reasons to adopt a model, etc. But this was in response to a claim that Amir was no better than replacement level.
I can grant you he's probably not in the top 25 or top 50 players in the league. But you don't find Amir Johnson's ready to sign 10 day contracts just lying around like your Jordan Mickeys or your who the hell knowses.
I think the problem is people are creating a binary choice between 'hidden stud' and 'complete dud' and neither is true.
ESPN's RPM had him as the 21st best player in the entire NBA last year. I suspect that almost everyone here would agree that's ridiculous (there are a couple hardline stat adherants who might defend that). So the question is not 'is he that good' it is 'how good is he actually?' and in particular 'how good is he actually today, at his age?'
It is true that five years ago, when he was playing in the high 20s in minutes, he still had a very strong RPM score. I think he has always been a solid player and someone better than the eye test (as I alluded to in my prior post, and as I suspect HRB would also agree). That is different than saying he was a legit top 20 in the league guy, though. Today, I believe he's less effective than he was five years ago and his minuts dropping by a third suggest that to me as well. I do not think he's replacement level today (he's better than that) but I also think he's nowhere near the top 20. He's getting paid in part to be a vet presence, but in part because he can still play and smart teams realize he's effective in a specific role. I don't think that his loss is as big as a pure stat look might suggest, because his useful role is one that is declining in value over time.
No doubt. Certainly not a replacement level player. And I appreciate that the advanced metrics call things out that maybe we didn't see, and each generation is a required part of the evolution to something even more useful. I'm just saying...
You're actually just misunderstanding what RPM is. All it does is calculate a players impact per 100 possessions, independent of minutes played. The metric isn't claiming that Amir Johnson was the 21st best player in the NBA, just that he had a positive impact in the minutes he played relative to the replacement level NBA player.
In other words, all Amir's RPM numbers are saying is that defensively he's about 3.58 points better per 100 possessions than a replacement level NBA player.
RPM isn't FUBAR. But the way people interpret it is.
But the sample will be skewed for someone playing short minutes. When they are going good they will play more.
People keep clamoring for Ainge to get rebounding and rim protection but I suspect they are going to be disappointed. I could be wrong but I think they are going to look for a big who can shoot from the outside to round out the roster.
Olynyk/Johson to Morris is a downgrade in terms of overall shooting and while having more scoring from Hayward (and possibly improved shooting from Brown) might offset that, they are going to still need someone to help space when Morris and Horford are resting. This is the style of play that the Celtics and Warriors play (as SRN noted above) - they live and die at the perimeter on both ends of the floor. Its why both teams have forsaken the traditional rim protector for guys who can spread the floor. I would be surprised if they don't take a look at Speights. If so, just watch - if you think Amir or Olynyk is polarizing you haven't seen anything yet.
Olynyk/Johson to Morris is a downgrade in terms of overall shooting and while having more scoring from Hayward (and possibly improved shooting from Brown) might offset that, they are going to still need someone to help space when Morris and Horford are resting. This is the style of play that the Celtics and Warriors play (as SRN noted above) - they live and die at the perimeter on both ends of the floor. Its why both teams have forsaken the traditional rim protector for guys who can spread the floor. I would be surprised if they don't take a look at Speights. If so, just watch - if you think Amir or Olynyk is polarizing you haven't seen anything yet.
Yes, of course. But RPM doesn't require you to ignore that. Literally all RPM is saying about Amir Johnson is that when he's on the floor, he's good. It's not pretending to be anything beyond that, nor is it insisting that anybody ignore the context that those numbers exist in.
A lot of people here are fundamentally misunderstanding RPM. Amir's numbers do nothing more than suggest that he's a strong role player.
This conversation has sort of evolved into a very different monster than what it originally was.
To me, the question is very simple. It was posited early that almost all serious basketball fans think Amir Johnson is currently better than Marcus Morris. Does anyone here (I think we are all pretty serious basketball fans) prefer to have Amir Johnson to Marcus Morris for the next year?
Personally, I would take Morris and it's not even close. I understand that there are certain "advanced" stats tHat prefer Johnson. My opinion is that anyone would prefer Johnson is fucking certifiable and hasn't actually watched a game within the last year
To me, the question is very simple. It was posited early that almost all serious basketball fans think Amir Johnson is currently better than Marcus Morris. Does anyone here (I think we are all pretty serious basketball fans) prefer to have Amir Johnson to Marcus Morris for the next year?
Personally, I would take Morris and it's not even close. I understand that there are certain "advanced" stats tHat prefer Johnson. My opinion is that anyone would prefer Johnson is fucking certifiable and hasn't actually watched a game within the last year
Many here have long used it as a measure of absolute value, and I agree that is a problem.
There are a variety of problems with it even for the limited purpose you describe above, too. This doesn't make it useless at all (I like it as a simple way to see a players impact) but it's a tool with some gaps. One of which is that it absolutely does ignore the context of the players performance, btw.
Thank you for this explanation, I was definitely misinterpreting RPM/RAPM to be absolute impact vs cumulative impact normalized over 100 possessions.
I would prefer Morris for health and salary reasons, but Amir if he's anything approaching healthy is a better basketball player than Morris. Amir Johnson has been a better basketball player year in and year out for their respective careers, it doesn't have to be advanced stats to know that. He's a much, MUCH, better rebounder, a better defender including being much much better as a rim protector, he scores more efficiently, when he's on the floor he performs at a higher level than Morris.
Morris' best ability is that he plays a ton of averagish minutes and he's cheap.
Agreed on it's limitations. My point is about how it's being used here and in other discussions here.
That RPM ignores the context of when a player's minutes were played does not require that we ignore that context in a discussion that includes RPM. The biggest problem here is that a bunch of people keep citing Amir's RPM rating as meaning that "advanced stats" claim Amir is the 21st best player in the league. And that's not what RPM is claiming.
But the fact that he is in the starting lineup limits, to some extent, the ability to spot him into favorable matchups. His matchup is whoever the other coach opts to start. I guess he goes for fewer minutes against less favorable matchups than against more favorable ones, but he is still logging minutes against starting lineups, which to me seems meaningfully different from, say, Olynyk.
I agree with you about scalability. I'm not arguing that Amir is the equivalent of a 35 mpg starter.
I've never disagreed w a post more than this that didn't involve Kelly Olynyk. Or Anything HRB posted regarding Hayward.
Morris plays 30-35 min and is a good defender, much better at switching onto perimeter players and has the size to defend the post and the lateral quickness to handle switches on the perimeter.
Minutes played is a very important stat. Amir being kinda good for 20min is fine, if you have a similar player to take those other 28min at the position.
If you don't, Morris playing 32-34min is much more valuable. Lessens the stress on the bench guys.
Do people see Morris playing 30+ minutes per game? I am assuming hes a 20 minute man max. I can see Morris as first off the bench with certain matchups, but I think Stevens will be going small a lot next year.
Likely no. But was just making a larger point. I think Morris in a 20-25 min role is perfect. He also lends a nessecary toughness that this team has lacked in recent seasons
Well:
1. I noted that minutes is his real skill, though I don't think either guy is playing big minutes for this team.
2. Morris is a 3 who can play small-ball 4, Amir is a 4/5 of course Morris is better on the perimeter, but he doesn't protect the rim even a little, and he's not a particularly good post defender (are you thinking of Markieff?).
3. If you consider Morris a 4, he's the worst rebounding 4 in the league, I mean he's really bad, he's not even particularly good for a 3, he rebounds like a guard. That's a major factor to me when comparing with Amir who is mediocre for a big, but not terrible.
4. Morris plays those minutes because he's been on a Piston's team the last 2 years that was devoid of talent, he was a bench player for Phoenix, it's great that he can play them and stay healthy, but there is nothing to indicate that he'll become a better player by getting less minutes.
5. Amir playing less minutes has been an issue of health and roster, he played 26-29 MPG from 2010-2015 (so more minutes than Morris, on teams better than Morris') and his performance was significantly better than Morris.
I like Morris, he's a decent 3/4 who can defend some and stretch some. I don't think if he stays on this roster that he'll be a significant contributor, he's just not significantly better than the guys he's competing with at any skill. I wonder if he gets moved again once we're back over the cap, he's like a cut rate Crowder for $10M over 2 years, that might get someone who is a 4/5 and therefore a better fit back.
That's cute and all that you want to keep talking about Amir Johnson like this is 5 years ago and he doesn't have a GIANT pitchfork in his back, but even in your bubble where Amir and Marcus are in the same point in their career, Marcus helps MUCH more vs the Cavs and Warriors, and beyond that I don't really care.
He's more versatile defensively,and yes he's not a great rebounder, but I still think when the time is right Ainge will address the DREB issues, also I think the increased size across the board w the loss of Bradley but more mins at the 2 for Jaylen/Hayward and also funky lineups w Smart/Jaylen/Hayward/Crowder/Horford has no traditional big but is above avg size at 4/5 spots will help the rebounding issue as well.
I certainly wasn't claiming that. At least I didn't intend to.
Any stat where 400+ NBA players measure not as well as Amir Johnson and only 20 players measure better than Amir is a jacked stat. IMHO.
I don't think Morris helps much against anybody to be honest, he's mostly just a shitty Jae Crowder. I actually noted I'd rather him to Amir, but overall Amir is a better player, it's not close, and every shred of statistical evidence from the most basic to advanced backs it up. As does common sense and the general opinion of NBA GMs who keep giving Amir big contracts.
Back to the thread title, I'm much less worried about depth now, but can we agree that the 2017-18 Celtics Summer League team is a bit of a bizarre monstrosity? A guard or two might help...
Yes, totally agreed. However, I wonder if they almost did it on purpose to force Brown and Tatum into ball handling/creator roles and get them out of their comfort spots. Brown clearly needs to work on his handle and decision making, and this is a way to do it in a meaningless environment. The results obviously aren't optimal so far though...
One thing is it's probably tough to get guys to play SL for the Celtics because they know there's no shot of making the roster, and there are multiple top 5 picks eating up the possessions. I assume Jackson would have been there if the Avery trade happened sooner, which would have helped.
I don't understand the comment about RPM and context.
As to what RPM is, I am citing the ESPN data, and ESPN explicitly says that its version of RPM is designed to show the value of the players; I get your point that it is reporting what occurs on the court, but at a minimum you need to acknowledge that people who are using it exactly as ESPN suggests are not misstating things.
http://www.espn.com/nba/story/_/id/10740818/introducing-real-plus-minus
We do have, of course, the problem we always face with a rate stat vs a cumulative stat and if your comment is just 'he's the 21st most impactful player per minute' that's fair. But still illustrates the problem with overrelying on the metric because I think few here would agree with that statement, either.
Again, I like it and I use it---but we should be clear it has its problems, too. That is why I typically push back when one of the folks who doesn't really understand what goes into making one of these stats overrelies on it. I don't think you do that, but we see it regularly around here.
Last edited:
I agree it is not as extreme as we see in some cases---as HRB has alluded to, the ratings of backup 4/5s on RPM have become a joke in some circles because of this problem. But I'd still wager that 15 of his 20 minutes are selectively chosen, too. I mean, I am not trying to argue he stinks---I think the 'eye test' folks underestimate him for sure.
I'd like that signing. He's a completely different player than he was 2 years ago. He went from shooting no 3s to half his shots being 3s. He can also rebound and block a shot better than Olynyk. He's exactly what we need, just the poor man's version of it.
Man, I have to say, whether or not advanced stats ever become as popular as they did in baseball, I could cut and paste this discussion onto the 1994 usenet baseball boards and it would fit right in. "this stat doesn't account for this." "I agree there is some value, but since some results don't make sense to my previously established bias, I am rejecting it." "let me make up this way that people have presented this stat so that I can knock it down."
Most importantly, though, the idea of "well I'll concede that this stat might accurately measures what happened, but it doesn't tell me who's better than whom" and the various appeals to authority...with a really vague definition of "better" that usually means "coincides with my own bias." It's pretty nostalgic really.
Most importantly, though, the idea of "well I'll concede that this stat might accurately measures what happened, but it doesn't tell me who's better than whom" and the various appeals to authority...with a really vague definition of "better" that usually means "coincides with my own bias." It's pretty nostalgic really.
I would really like to have a PG playing as well, and I can't imagine that there isn't a functional one somewhere who would give his hightops for a chance to run even an SL team. I wondered about the idea it was forcing Tatum/Brown to handle the ball more, but I think the real value would be in playing some actual celtics sets. Guess not.
Also, KO is/was a decent player (that I don't mind saying goodbye to) but come on people: Amir is on his 5th straight year of decreasing minutes, rebounds, points, steals and blocks. Giant fork indeed.
The point people are making is not about what to expect going forward from Amir or that it's a mistake to let Amir go. It's to refute the claim that Morris is going to be an upgrade over what Amir brought the last two years. I realize it's gotten lost in the stat discussion but the whole reason the stat discussion started is the frankly dumb idea that Amir has been a garbage player for the Celtics. Like, what has actually happened.
I agree we shouldn't have brought Amir back and he appears nearly done. I still think replacing him isn't trivial - he gave us some good value, even for much of last year.
I agree with that wholeheartedly. Amir was a productive player, and he seems to be underappreciated around here. Probably some of that is due to his finishing last year with a limp. And part of this "Yay Morris" phenomenon is likely due to lowered expectations on what the return on the cap-clearing trade was going to be, and ending up pleasantly surprised with the Morris result. If Morris were capable of adding the same value that Amir did the last few years, that would be a huge win.
Apples and oranges my friend. In baseball, stats are highly predictive for offensive results. Pitching has more variation and then defensive metrics get quite a bit more complicated, since you have to account for interaction (i.e. a great CF will take away putouts from a LF).
In basketball, every single possession is an interaction with the other 4 players on your side and the 5 players on the opposing team. Avery Bradley is a good NBA player, but we can all see how he would be better positioned to succeed on a team like Cleveland or Milwaukee (with LeBron or Giannis running point forward), than on a team like Boston where they need to hide IT on defense...
I'm like a broken record on this, I know, but...
Listening to NBA Radio a lot and it seems (and I'm biased here) that the Celtics are essentially being dumped on by a lot of the "experts" for, I don't know, improving their team because there's no way to beat Cleveland/GS, and not one of these experts acknowledges that the Celtics are still IN THEIR REBUILD.
It's like they're penalized for winning 53 games during this period. They're not in their window fella, their window is just barely starting to open.
Four seasons ago they won 25 games, Philly won 19, and yet now these guys think the Celtics are treading water while Philly is in some glorious "trust the process" rebuild.
The host today actually said "Hayward is a marginal all-star and I had no idea they'd have to give up Olynyk and Bradley for him. Did they know this?"
Yes, kind of. KO was gonna be a goner and it's really not the Celtics fault the cap came in lower.
It seemed like this host would be more impressed if the Celtics had won 43 games, and kept KO and Bradley. They simply can't process a team winning 53 games and going to ECF *during* the rebuild.
But, you know, then it was onto Lonzo talk.
Listening to NBA Radio a lot and it seems (and I'm biased here) that the Celtics are essentially being dumped on by a lot of the "experts" for, I don't know, improving their team because there's no way to beat Cleveland/GS, and not one of these experts acknowledges that the Celtics are still IN THEIR REBUILD.
It's like they're penalized for winning 53 games during this period. They're not in their window fella, their window is just barely starting to open.
Four seasons ago they won 25 games, Philly won 19, and yet now these guys think the Celtics are treading water while Philly is in some glorious "trust the process" rebuild.
The host today actually said "Hayward is a marginal all-star and I had no idea they'd have to give up Olynyk and Bradley for him. Did they know this?"
Yes, kind of. KO was gonna be a goner and it's really not the Celtics fault the cap came in lower.
It seemed like this host would be more impressed if the Celtics had won 43 games, and kept KO and Bradley. They simply can't process a team winning 53 games and going to ECF *during* the rebuild.
But, you know, then it was onto Lonzo talk.
"I've been stabbing myself in the balls repeatedly of late, and I have some interesting observations I'd like to impart as a result"
In all seriousness though, I've been hearing a lot of "concern" about the Celtics from friends lately. Most of them are Lakers fans, who want to talk about how worried they are that the Celtics are hoarding assets, not getting Jimmy Butler, not drafting Lonzo Ball, etc.
I take it as an indicator that things are going quite well.
I take it as an indicator that things are going quite well.
Well, yeah, but Christ, I'm driving in the car hoping for signs of life. Not always in the mood for music.
I come here too and it's not all Shakespeare you know?
I actually listen to NBA Radio quite a bit and have noticed a lot of the same concern-trolling on there. I had no idea how many fans of other teams thought that Avery Bradley was so critical to Boston's success. Same with Olynyk.
Or the constant refrain that "eventually Ainge has to use some of these picks, right?"
Or "the Celtics should have traded for Paul George first before they signed Hayward!"
Just gotta laugh it off. I mostly enjoy NBA Radio though, particularly Isola and Scalabrine.
It's only 'jacked' because you're contextualizing it incorrectly. You're arguing that the entire thing is worthless because it's making the following statements:
1) Adjusted to 100 possessions, Amir Johnson is 3.58 points better defensively than a replacement level NBA player.
and
2) Adjusted to 100 possessions, Amir Johnson is basically replacement level on the offensive end (0.22 points better than replacement level per 100 possessions).
For some reason people here look at that ESPN RPM list, and because it filters arbitrary to only include players who played >15 minutes/game, read it as some sort of final ranking.
This is an extreme example obviously, but would you look at NBA.com's rebound rate leaders and conclude that Danuel House's 50% rebound rate makes him the best rebounder in the league? Of course not, because he didn't even play a full minute of NBA basketball this season. Likewise, Jordan Farmar's probably not going to average 3 points per possession in isolation. Does the fact that you need to account for sample size in those rate stats make them pointless? Of course not. You'd simply adjust the filters accordingly.
So why do Amir's possession adjusted RPM numbers make the stat 'jacked' in this case?
Last edited:
No doubt, Celtics and Danny have definitely taken a beating on most mainstream and internet channels. Mostly this comes down to not landing a STAR player like George and Butler. The 'wouldn't give up Rozier' meme is also pretty popular.
I'd argue that renting George would be a dumb short game move (even though from most reports they made a better offer than what the Pacers eventually got) and their offer for Butler was also better than the deal the Bulls eventually made.
Like you said (and thanks to IT) these have been pretty successful Bridge Years and the best is to come.
Yeah, my post wasn't clear on that at all. All I meant is that it's incumbent upon whoever is citing RPM to understand what's actually being stated by the stat. And I don't think ESPN is trying to say -- even in the piece you quoted -- that this metric is the be all, end all of who is a better NBA player. Hell, even in your quote they say that RPM "can help" in the Gibson/Crawford debate; it doesn't read to me that they're presenting it as empirical evidence, just trying to explain a potential use case for the data.
And of course RPM is designed to measure the value of players, I'm not arguing that it doesn't. I'm just pointing out that much of the RPM backlash in this thread seems like it would be better targeted at ESPN's completely arbitrary filtering of that list to players who played >15 minutes/game, and not RPM itself. If that ESPN link was filtered for players who played >25 minutes/per game, do you think we'd have heard anything about RPM being FUBAR and advanced stats being worthless?
All that said though, I think for the most part, we're totally on the same page here. RPM is a rate stat and should be treated as such, and sample size and all manner of other things contribute noise to rate stats. RPM can provide useful insight into players, but shouldn't be viewed as anything more than a rough estimate of their per 100 possession value. The problem here is that many people insist it's more than that, whether they're arguing for or against RPM's data.
This is good. I wandered into baseball at the tail end of that, and this depresses me--at least the sabermetrics cavemen had the excuse of "novelty is scary."
I just busted out laughing. This is great.
It took Belichick years before even the haters acknowledged that he was smarter than they were and they were left to merely speculate at what the actual genius was behind any moves that perplexed them. Popovich/Buford likewise, although San Antonio is a very different sports media environment than Boston. Ainge isn't in that category yet, may be heading there, but he's getting Belichick-circa-2003 kind of second-guessing. If KG doesn't get busted up in 2009 or Perkins doesn't hurt himself in 2010 Game 6, Ainge might be viewed the same way already. If Tatum outperforms Fultz, or the lottery balls bounce right and we end up with two top-5 picks next year, he might get there - but really, if the next window of contention results in a title or two, 3-5 years down the road, he'll start getting the same breathless deference.
Trust the process, indeed.