Discussion of Pierce and Ray vs other elite SGs by the numbers

[drza44]

Instead of re-posting the whole article (I wrote it in the fan posts section of the blog), I'll just link to it and hit the highlights:

http://www.celticsblog.com/2011/2/28/2018520/pierce-and-ray-vs-other-elite-wings-by-the-numbers

*The article compares Paul Pierce and Ray Allen with 25 other good wing players in the NBA according to 5 advanced stats: PER, win shares, wins produced, adjusted +/-, and Roland Rating (82games.com).

*I show the top-15 wings for each of the advanced stats, along with their "value" in that stat.

*At the end, I summarize by looking at all 27 players as a group.  I average their rank order in all 5 of the stats, and use that as a rough-and-ready ranking tool.

*Based on that last composite table, I group players into tiers.  Among NBA wings...

Tier 1 is essentially LeBron James.  He was first in all 5 stats.  The only person even close is Dwyane Wade, who finished 2nd or 3rd in every stat to also edge onto tier 1.

Tier 2 is Kevin Durant, Manu Ginbobili, Paul Pierce and Kobe Bryant.  While Kobe and Durant are super-duper stars and Ginobili has gotten some quiet acclaim as a dark horse MVP candidate, it's interesting that the Captain measures right with them this year without nearly the fan fair.

Tier 3 is a group of 8 guys, highlighted by Ray Allen (who landed just behind Kobe according to the numbers).  This tier is an interesting mix of big names and guys that may be having a bigger impact than their names would suggest.  New Knick Carmelo Anthony is on this tier with Ray, along with several young scorers (Gay, Gordon, Martin) and a couple of versatile/glue/defensive type players (Iguodala, Deng and Fields).

I'm not going to spend much time with the rest, though it is interesting that guys like Joe Johnson and Monta Ellis are maybe further down the list than you might expect.  The take-away for me is that while Miami (obviously) is built around 2 super-wings, the Celtics (and I guess, now, the Knicks) are the only other teams with two wings from among the top 3 tiers.  This continues what I would suspect will be a common theme for the Celtics in these statistical measurements...quality and quantity combining to form a potent whole.

Pierce and Allen may not be quite as good as the Miami boys, but they're right there with Durant, Kobe, Manu, Melo and the other wings with an argument for top 5 in the league.  Sounds about right to me.

Any thoughts?

(I made this thread on the message board because it seems like often the folks that discuss on the boards are a different group from those that discuss on the fan posts page.  There is some overlap, but often it's an entirely separate discussion.  But if you have a comment or suggestion, feel free to comment at either place.)

[Celtics4ever]

So far, we 've fared well against the MIA lads.  However advanced stats are nothing more than numbers.  I would take Kobe over LeBron any day of the week.  People are still stat drunk over LeBron and he was swept in the finals, his teams don't beat good teams and he finds a way to choke when games matter.   Last night for instance, he missed a three to win the game and had a layup blocked a few seconds earlier.  PP is pretty clutch.

[droponov]

Dave Berri's controversial stat (most likely to be trashed on an APBRmetric board) is also the one seemingly growing fastest in popular usage

Really, is this true? I don't have much use to any linear-weighted boxscore stat (or for any of these rating stats, for that matter, including the +/- based ones, especially with small samples), but that's depressing. If the Celtics sign Murphy, Berri will probably advise Doc Rivers to start him ahead of Garnett. Or at least alongside Garnett. Berri is a very bad influence on the collective knowledge of the game.

I think playing alongside Garnett and Rondo have helped Pierce and Allen to do well with these metrics. They're having great seasons though.

I suppose Joe Johnson is being penalized for his lack of health/form before he had the surgery, but it's amusing how he's below a role-player like Fields. 

[drza44]

Dave Berri's controversial stat (most likely to be trashed on an APBRmetric board) is also the one seemingly growing fastest in popular usage

Really, is this true? I don't have much use to any linear-weighted boxscore stat (or for any of these rating stats, for that matter, including the +/- based ones, especially with small samples), but that's depressing. If the Celtics sign Murphy, Berri will probably advise Doc Rivers to start him ahead of Garnett. Or at least alongside Garnett. Berri is a very bad influence on the collective knowledge of the game.

I say that from my impressions.  It seems to me that among more casual fans, PER is the advanced stat used as the default (makes sense since it's ESPN) but whenever I see a new article in the New York Times or some other public venue that is trying to embrace these radical new-fangled numbers it's almost always wins produced that they use.  Gladwell made news by embracing and discussing wins produced.  Then, I would say win shares are probably the most en vogue among the various message boards that I go to.

The family of +/- stats still aren't embraced as much as I would like (I think they're vitally important, especially as a balance for those that rely on the box scores).

I think playing alongside Garnett and Rondo have helped Pierce and Allen to do well with these metrics. They're having great seasons though.

I suppose Joe Johnson is being penalized for his lack of health/form before he had the surgery, but it's amusing how he's below a role-player like Fields. 

One of my motivations for doing this was to explore the idea that players on good teams are sacrificing their "stats" for the sake of the team.  Obviously, if you look only at volume counting numbers sacrifices are being made.  But across the advanced stats, it seems to me that you get a more accurate idea of worth.  For Pierce and Allen, even now, if they were solo missions on bad teams they'd probably be scoring in the 20s per game which would look better in a casual perusal.  But even in that eventuality I still don't think they'd grade out higher than the 4th - 7th range among wings that they measure out currently in their "sacrificed stats" situation.  In other words, I think an analysis like this does a better job of gauging a player's actual level.

So, to your points, I don't doubt that playing with Rondo and KG help Pierce and Ray to be more efficient.  This is the trade-off to them not having as much volume.  But in the end their net effectiveness is similar in either instance, so I'm glad that we have measures that can capture that.

And re: role player vs main option, it's hard to compare across roles but one way to look at it could be that Fields has been a better role player this year than Johnson has been a main option.  As you point out, injuries could have played a part in that, but maybe measures like these could help us to find the line between where a first option guy is doing his job so poorly that, in the scheme of things, he's being less productive than a good role player.  Just a thought.

[drza44]

So far, we 've fared well against the MIA lads.  However advanced stats are nothing more than numbers.  I would take Kobe over LeBron any day of the week.  People are still stat drunk over LeBron and he was swept in the finals, his teams don't beat good teams and he finds a way to choke when games matter.   Last night for instance, he missed a three to win the game and had a layup blocked a few seconds earlier.  PP is pretty clutch.

This is more about rating individual player contributions, not the team itself.  Miami going 1-2 on this list is fine, but they had nobody on the PG list (where Rondo was one of the best) and (though I haven't done it yet) I suspect their big man will fall behind KG as well.  LeBron and Wade are sick players, but the Celtics are a better team.