Is Bradley a keeper?

[PedroKsBambino]

 
 
 
To sum all this up:

• At his best skill (catch and shoots) he is average
• Looking at his overall offensive skillset he is below average from a shooting efficiency standpoint and with drawing FTs, which means his non-strength skillset (creating on his own) is a giant black hole in his offensive game
• He is average in some ancillary defensive stats and significantly below average in some others
• Slightly above average defending PGs and average when defending SGs
• Hence, overall defensively I see him as a bit of a mixed bag not a dominant defender even against just PGs

 
Thank you for posting the data---very helpful to see where you are coming from, and appreciate the detail.  That said, even just taking your summary above, your conclusion about him doesn't really follow.  And I don't think your summary is totally fair, either.
 
First, on his shooting...
 
As several people have said, he is not a star, nor a spectacular shooter. I think one of our core disagreements is that you seem to think if someone isn't, they stink and need to go...that just isn't how NBA teams are built.  That repeated, being 'average' or 'above average' in several specific skills (which is what your own data shows) is not a bad thing.  He also has improved this year, as previously discussed, so the idea that he doesn't have any chance of getting better seems insupportable to me.  I think this one comes down to baseline---yours seems awfully high to me given the salary level we're talking about.
 
Second, I do not agree with your description of his defense---even the metrics you list suggest he's a better defender than your summary. I think it's a little unfair to ding him for the FT given that he has an atrocious backline defense, about the worst in the league.  If I thought this were a close call we could look at multiyear data, but I just don't think it really is.  Even if we say he's only a little above average here, still...this is a nice combo.
 
On balance, seems like this data continues to tell a story of a useful (if clearly flawed) player who is very young and has shown improvement.  If you think that's totally fungible that's your prerogative; I just don't think that's true.
 
I do think JakeRae's point is the real one to ask about Bradley; even if we acknowlege he's a solid player now and may have some growth in him, he's not at all likely to be a star, and given that is it worth it?  That's one I think people can still disagree on, but is what Ainge is likely really pondering.

 

[PedroKsBambino]

 
On the study, I owe you a better one, just need to find the time to research it.  However, if you are going to hold me to this burden of proof its probably a waste of my time.  My point was that the general / macro trend for NBA players was to peak within a very early amount of time from the start of their career.  I am not foolish enough to insist that every player will follow that career arc to a T, my only point was that if Bradley is going to be a late bloomer he is going against the grain, is going to be an outlier.  I would hope we could surmise that with a better study hopefully without addressing every single thing you mention here.  If again, you can accept that all I am saying is that its atypical for players to develop in the way you think Bradley might have significant development after year 4 and age 24/25  But if you want to claim that is very normal for players to have development curves like that, I might have to ask you for some evidence of that because I really dont think that assertion would be generally supported here, but I may be wrong.

 
My point was that there's a lot of reason to question the study you posted, and I think the majority of thinking on peaks has it several years later.   As I noted in the analytics thread, we can't expect perfection in any of these studies or new metrics but the one you cited isn't even valid as to the individual players it cites.  Doesn't that give you some pause about it?

 

[Brickowski]

Chris Johnson had 14 points in 14 minutes tonight. That does not bode well for Avery's return to the Celtics next year. BTW, Johnson is a decent defender too, and he's 6-6.

 

[Devizier]

The hope is that you can trade Bradley for something better than anything the Celtics will get in free agency this year, which is a very safe bet in my opinion.

 

[bowiac]

As for peak, that blurb simply doesn't provide enough info on the study to be valuable, and candidly I highly doubt it is accurate. Methodologically, we don't know how he dealt with early-entry; we don't know how he compared players (for example, counting stats can be higher when younger but efficiency stats often are worse---our view of 'value' is different between those today than it was even in 2009).   We don't know how he dealt with players who leave the league (many 22-25 year olds aren't in the league when they are 32. This doesn't mean the player peaked early, though, it likely means teams kept a player thinking there might be growth).    

I think Dave Berri is a hack, but some more details on the study are here. He used wins produced, but notes it doesn't change with respect to efficiency.
 
Here are a couple blurbs about other aging studies: One from Kevin Pelton, finding a peak age of 27, and more importantly, giving you the likelihood of improving based on his metric at any given age. He's one using win shares, by the very smart Neil Paine, finding a peak of age 27, and noting it's different from guards than forwards/centers:
 

I think the guards are the most interesting, because some peak as early as age 25, some peak at the conventional 27, and some even hang around to peak as late as age 32! Among guards who played more than 1 season of at least 2000 MP, here's the distribution of peak seasons by age:

Finally, I know from my contacts with the Rockets (obviously a very analytic friendly team), that they have concluded that age studies for the NBA, unlike MLB, are not especially helpful overall. They've concluded the contextual problems are too severe. The one exception was 3 point shooting, where they have found useful overall trends, although I don't know the specifics of what those trends are.

 

[wutang112878]

 
As several people have said, he is not a star, nor a spectacular shooter. I think one of our core disagreements is that you seem to think if someone isn't, they stink and need to go...that just isn't how NBA teams are built.  That repeated, being 'average' or 'above average' in several specific skills (which is what your own data shows) is not a bad thing.  He also has improved this year, as previously discussed, so the idea that he doesn't have any chance of getting better seems insupportable to me.  I think this one comes down to baseline---yours seems awfully high to me given the salary level we're talking about.
 

 
It might come down to our team building philosophies.  IMO when you are rebuilding retaining average players at their free market value isnt always particularly helpful, especially if you are trying to get under the cap.  Its not that one deal cripples you, but its the collective market value deals that hurt.  Jeff Green making $9M isnt individually crippling, neither is Bass making $7M, or Lee making $5M or Terry making $5M but collectively when they take up ~$26M or 45% of the $58M cap that holds your franchise back.  You have ordinary players who keep you from bottoming out and getting high lottery picks, and you can move them but their money doesnt just vanish off the cap instantly it might take a year or two with the deals you take back.
 
So when I think of retaining Bradley, unless you see big upside (which we know I dont), I dont see the point because you can go get an average SG at a free market price at virtually anytime.  Courtney Lee is a good example of this.  When Houston was in the 'asset accumulating' phase before they got Howard, they didnt want to resign Lee and traded him here and got a 2nd round pick because they didnt see the value in giving him a deal that would clog up their cap a bit.  We wanted him because we needed to add talent to an established core, so we didnt have a problem paying the market value.  But then once we needed to rebuild the core Danny moved Lee because he wasnt particularly helpful.  
 
Which is why personally I would rather save the cap space and try to get flexibility rather than getting a player of average caliber when I can go acquire a player like that at any time.
 
Also, think back to the 06/07 Celts before the Big 3 trade.  The big deals we had on the books were: PP at $15M, Wally at $12M, Ratliff (expiring) at $12M, Baker (albatross) at $5M and everyone else had a salary of $3M or less.  Thats a healthy cap situation where you can make huge splashes and sweeping change, thats where I would like us to go and resigning Bradley isnt a movement towards that direction.  And again, its not that his one deal would be so awful, its that we have other awful deals that we need to work around, so adding a 'not great' deal isnt a step in that direction to me.
 
 

 
Second, I do not agree with your description of his defense---even the metrics you list suggest he's a better defender than your summary. I think it's a little unfair to ding him for the FT given that he has an atrocious backline defense, about the worst in the league.  If I thought this were a close call we could look at multiyear data, but I just don't think it really is.  Even if we say he's only a little above average here, still...this is a nice combo.

 
Wait, here is the narrative to the stats which are just listing rankings:
 
Bradley is 11th out of 22 in steals, which is counter intuitive considering he is thought of as such a great on the ball defender
13th in blocks per game
17th in Opponent FGAs per game (meaning he gave up the 17th most FGAs at the rim)
21st in Opponents FG% at the rim.  And before we say its because the Celts lack a presence inside, look at the list there are a lot of teams without good interior defenses on there as well
 
Bradley ranks 6th, but his 44.6% above the average of 48.9%, but when you look at his PER when guarding PGs and he ranks 16th and his 15.4 is above the average of 14.5  Thats probably because he gives up 5.5 FTs per 18.3 FGAs when guarding PGs and for a great defender you would think he would be better than average in FT/FGA but he isnt.  So its sort of a mixed bag with him guarding PGs
 
There was one stat (eFG% when guarding PGs) where he was 6th, but in every other stat he was average to below average.  There really isnt a debate there, those are just the numbers.  So when I look at the those rankings and then I conclude that he isnt a dominant defender, I dont think thats outlandish.  Every stat but one says he is average, so I know our eyes say he is a great defender because of his intensity, its just not visible in the numbers.  So if you want to say I am incorrect on this one, I think you have to back that up with some numbers yourself to show me the error in my ways.
 
As for his FTs and backline defense, that same logic applies to most of the other guys on that list.  But more specific to Bradley, lets compare 13/14 Bradley to 11/12 Bradley when he had KG playing behind him and the Celts finished 1st in the league in defensive efficiency, he had virtually the same numbers.
 
 

FGA	FTA	FTA/FGA
11/12 vs PG	15.8	5	0.316455696
11/12 vs SG	19.3	4.2	0.21761658
Average			0.267036138
13/14 vs PG	18.3	5.5	0.300546448
13/14 vs SG	16.1	4.1	0.254658385
Average			0.277602417

 
 
 

 

On balance, seems like this data continues to tell a story of a useful (if clearly flawed) player who is very young and has shown improvement.  If you think that's totally fungible that's your prerogative; I just don't think that's true.

 
When you say improvement, do you mean on the catch and shoot numbers or is there anything else that is improving offensively that I dont see?  Because if its just catch and shoot we need to remember that is 1/3rd of his offensive game where he has become average, which is why I just dont get too excited about this improvement.
 

 

[wutang112878]

 
My point was that there's a lot of reason to question the study you posted, and I think the majority of thinking on peaks has it several years later.   As I noted in the analytics thread, we can't expect perfection in any of these studies or new metrics but the one you cited isn't even valid as to the individual players it cites.  Doesn't that give you some pause about it?

 
This is totally a fair criticism, and I havent had the time to find a good study to illustrate my point.  
 
 
I'll stop my crusade on saying that Bradley cant improve, but I will just say if offensively he is going to become average he has a ways to go.