Again, this amounts to another "my stats are better than your stats" argument which is beyond pointless in the context of this conversation. I was attempting to explain to bajaden how I can value defense highly and
still want a lazy defender on the team next season. Maybe you don't like the way I use defensive rating, maybe you think Rondo is indefensible, maybe you'd rather we start Collison at PG next season. That's fine. It's all good. I don't have a problem with any of that. But if you think there's
objective evidence that Rondo is a terrible defender -- whether that means one of the worst on the team, one of the worst at his position, or one of the worst in the league -- on that we disagree.
As far as stats are concerned I want something objective, I want something repeatable and falsifiable, and I want something which works. That's it. If either of you wants to undertake a comprehensive analysis of defensive statistics, which ones are worth caring about, and how to contextualize them more power to you. I'll read it with interest. But I'm not going to justify my process for every off-hand comment I make. I've done it once for you Amit149 until it became clear that it was going nowhere. Now I've done it again here for twslam07 and your inaccurate paraphrase of my comments to bajaden indicates to me that this is going nowhere too. I've put more than enough effort into explaining
my point of view on this topic already and you're still not getting it. What else do you expect me to do if I can't even get you to understand what I've already said?
I'll try though. Here's the long and short of it...
ASSERTION: Rondo was a terrible defender this season.
MY REBUTTAL: Find me an objective stat which shows me he was a terrible defender. I think he played lazy this season too and yet he still wasn't terrible -- either relative to his teammates or other PGs in the league.
This is where we go off on a long tangent where twslam07 explains that Collison might be slightly better than Rondo for several reasons and then that Collison
last season was better than Rondo
this season -- even if I concede both of these points, what do they have to do with my rebuttal? Isn't it possible that neither one of them is terrible? Then twslam07 tells me that I don't understand how defensive rating works and proceeds to explain how points allowed per 100 possessions is skewed because backups and starters play with different units (a point which I not only agree with by the way, I made it myself in
this post in reference to all on/off or plus/minus derived stats). I've already made it abundantly clear that I'm not using nba.com's points allowed per 100 possessions, I'm using basketball-reference.com's defensive rating which is calculated completely differently and is not influenced one way or another by which teammates a particular player shared the court with. But just in case I quoted the exact formula in my response so there shouldn't be any further confusion right? Nope, you still assume I mean points allowed per 100 possessions and somehow
your argument about comparing Collison and Rondo has morphed in your head into
my argument about comparing Collison and Rondo.
You're saying "Look at how much better Rondo's defensive rating is compared to Collison or that his opponents score 2% higher than their average when Rondo guards them (which isn't a positive by the way). His lazy defense must be paying off!"
Not even close. I'll again direct you to MY REBUTTAL paraphrased above and
this post which you're (inaccurately) attempting to paraphrase. I was comparing Rondo to all of the guards on the team -- Collison, Curry, McLemore, Belinelli, and Anderson and pointing out that there's no objective evidence that Rondo was the worst of them by far. Then I was comparing him to other starting guards in the league this season and asking how it's possible for a terrible defender to rank 19th out of 53. You're the one who brought up points allowed per 100 possessions not me so I don't see what you're hoping to prove by arguing on behalf of it's flaws. If you prefer that stat to Dean Oliver's Defensive Rating, than
doesn't this show that Kosta Koufos, Seth Curry, and Willie-Cauley Stein were the worst defenders on the team? If
that's the stat you're using to show that Rondo was terrible at defense this season, then weren't they even worse? Now you have an objective stat that shows Rondo was at least one of the worst defenders on the team, but your problem is that it requires you to say that Kosta Koufos, Willie Cauley-Stein and Seth Curry are all terrible defenders too. I'll let you argue on behalf of that one if you'd like to. I've said all along that I don't trust that stat and I don't like to use it.
And since you tucked a little shot in there at Rondo for allowing opposing players to shoot 2% better against him than league average,
here are the other guards who allowed their opponents to shoot 2% better than league average against them this season: Jimmy Butler, Damian Lillard, Chris Paul, James Harden, Brandon Knight. All of these guys are terrible defenders too? My point all along has been that "terrible defense" is a relative term. For him to actually be terrible he would have to be either the worst defender on the team by far or one of the absolute worst in the league at his position. I have yet to find any objective stat which says that either of these is true. So it must be hyperbole then right?