Knicks Morning News (2018.09.13)

It’s midway through September and Noah is still on the team. Is this just that they’re just taking a long time to negotiate the terms of his buyout? Or is it a sign that Perry sees what we do: that there’s minimal value at best to stretching Noah until we know for sure we will need that cap space?

It’s midway through September and Noah is still on the team. Is this just that they’re just taking a long time to negotiate the terms of his buyout? Or is it a sign that Perry sees what we do: that there’s minimal value at best to stretching Noah until we know for sure we will need that cap space?

I guess either option is good, right? They both would mean we aren’t just stretching him before we know we’ll need the space and paying the full price. Hopefully we’re either trying to save so money or just not dealing with this situation until it’s necessary (or both).

@4 in the absence of greatness go for stability. Is that also the definition of mediocrity?

Old Hippocrates had it right when he said “First…. do no harm” 2500 years ago 🙂

@1 I’m not optimistic the Knicks found religion on this. If I had to guess, the delay is because he’s refusing to take one dollar less on his buyout, and the Knicks are trying to wait him out. I think we just postponed the buyout until the start of training camp or the regular season.

I’m pretty sure the players vote on captains and both Lance and Lee are co-captains. I wouldn’t read too much into it beyond “they’ve been around longer.”

Is KP worth a max contract for his loomin extension?

I’m not so sure every team would offer the max

Burke was unbelievably efficient last year; outplayed both Dragic & Kemba but thats not important. Sure, we can bury PER but why exactly? Also, what do we replace it with?

Sure, we can bury PER but why exactly? Also, what do we replace it with?

https://alexpetralia.com/datablog/2015/6/11/nba-how-do-per-and-win-percentage-correlate

Offense:

The relationship is significant and, here, PER explains about 79% of the variation in a team’s offensive efficiency.

Defense:

This time, the relationship is insignificant with a p-value of .806 and PER explaining only .2% of the variation in defensive efficiency. In other words, the model is extremely poor and PER says nothing about how a team performs defensively.

PER’s flaws are more obvious, but no more egregious than the flaws in the other metrics widely used here over the years, including the ones you cited. It is actually better-aligned with BPM/VORP than Berri’s WP, which was the “gold standard” for some around here.

Tyson Chandler and Allen Iverson are two great examples. Both PER and BPM-based stats conclude that Iverson was consistently a more productive and valuable player than Chandler in their respective primes. Yet Iverson was a punching bag around here for the Berri acolytes, some of whom have now moved on to BPM. Yet you will likely never hear the usual suspects say “hey, maybe I did overrate Chandler and underrate Iverson!” Just like you will likely never hear any of those same people say that PER has its uses as much as any of the others.

Thanks Jowles and Zman; This girl is learning a thing or two about math (statistics) and I have to admit that I kind of like it.

Do I have this straight? In a nut shell, PER is pretty good offensively but is lacking big time defensively? I guess, this explains why Kanter and Timmy are ranked in in top 10 and both Burke and KP are ranked 2nd in the Eastern Conference.

Bruno (btw, love that name): We can agree to disagree about whether 43% of one season is a valid sample size and/or the net valid discount effective rate for team losing most of the games played.

With respect to other things like WORP, BPM or whatever you prefer and based ONLY on their production in 2017-18, where does our starting line up rank with respect to other players in the Eastern Conference?

Kanter
KP
Timmy
Frank
Burke

If you shoot a lot, you’ll have a high PER, even if you shoot and miss a lot. PER absolutely loved Carmelo Anthony.

hard to believe we may be actually living in the golden age of civility:

https://slides.ourworldindata.org/war-and-violence/#/title-slide

Worth a read.

PER has been said to reward inefficient shooting. To quote Dave Berri, the author of The Wages of Wins:

“Hollinger argues that each two point field goal made is worth about 1.65 points. A three point field goal made is worth 2.65 points. A missed field goal, though, costs a team 0.72 points. Given these values, with a bit of math we can show that a player will break even on his two point field goal attempts if he hits on 30.4% of these shots. On three pointers the break-even point is 21.4%. If a player exceeds these thresholds, and virtually every NBA player does so with respect to two-point shots, the more he shoots the higher his value in PERs. So a player can be an inefficient scorer and simply inflate his value by taking a large number of shots.”

Hollinger responded, via a post on ESPN’s TrueHoop blog:

“Berri leads off with a huge misunderstanding of PER—that the credits and debits it gives for making and missing shots equate to a “break-even” shooting mark of 30.4% on 2-point shots. He made this assumption because he forgot that PER is calibrated against the rest of the league at the end of the formula.

Actually, if we took a player that was completely average in every other respect for the 2006–07 season—rebounds, free throws, assists, turnovers, etc.—and gave him a league-average rate of shots, and all of them were 2-pointers, and he shot 30.4%, he’d end up with a PER of 7.18. As long-time PER fans know, that would make him considerably worse than nearly every player in the league.

To end up with a league-average PER of 15.00, the actual break-even mark in this case is 48.5%, which is exactly what the league average is on 2-point shots this season”.

I agree about PER. I think it also correlates pretty well with a players perceived trade value (which can be different from his actual productivity on the court). That is, if a player has a high PER, GMs are willing to give up more to get him. I think this is much more true for PER than for, say BPM or RPM.

@20 and @23

During Melo’s time with Denver, his PER never went above 22.2, which is high, but he was never in the top-10 in the league during that time. He only cracked the top-10 twice (2o12-13 and 2013-14, the same 2 years that his WS48, OBPM and VORP peaked, as did hid USG%, but not his TS%….go figure!) even though he was in the top-10 in usage ten times. He never approached LeBron’s or Durant’s average PER. So PER “liked” Melo, but didn’t “absolutely love” him.

I’m not arguing that PER isn’t flawed. It is. I’m arguing that WP, WS, BPM and VORP are equally flawed, and that if you want to “bury” PER for its flaws, you might as well bury the whole lot of them. For example, Melo had a higher OBPM than Chandler in 2011-2012, even though Chandler’s TS% was .708 and Melo’s was .525. So OBPM “confirms” that Melo was significantly more valuable than Chandler on the offensive end in that year, which was disputed by Jowles and others at the time.

That defense by Hollinger still doesn’t explain why a player as dogshit at Iverson in 2001-02 (at everything but USG% and STL%) could be 11th in the league in PER. So either PER is giving him primary credit for the #4 defense in the league, which would be a laughable argument, or it overvalues volume scorers at virtually any efficiency.

Iverson should not have been ranked #11 in PER if he was making shots less often than the worst offensive team and doing virtually nothing else of value in the box score. And I’m a guy who thinks that steal rate can be attributed to individual talent.

The cold, hard fact is that Iverson used more scoring attempts than any player in the league to create points, and he was worse than the worst shooting team in the league at turning those shot attempts into points. Incredibly, he was #2 in shots missed while playing only 60 games. He was also #2 in FGA.

Of the 32 players who played 1500 minutes on >25 USG%, Iverson ranked only ahead of Jason Williams (.466), Jordan (.468) and MVP-candidate Marcus Fizer (.482) in TS%. Yet again, he was #2 in the league in FGA and FGM, just 20 FGA behind Antoine Walker, who played in 81 games.

This isn’t a speculative argument about how a hyper-efficient scorer (say, Tyson Chandler) inherently makes it harder for his teammates to score. No evidence has ever been given for that claim. But there’s lots of evidence that chuckers are not productive basketball players.

Either PER builds in a “volume scoring is inherently good” facet or the stat is FUBAR.

I really feel the arguments are becoming personal or out of spite when PER merits a defense like this.

I’ve said it in a jokingly manner because I thought we had buried the idea of PER being any relevant years ago when it became clear it was mostly just reinforcing common sense and overvaluing volume shooting sans efficiency.

I agree about PER. I think it also correlates pretty well with a players perceived trade value (which can be different from his actual productivity on the court). That is, if a player has a high PER, GMs are willing to give up more to get him. I think this is much more true for PER than for, say BPM or RPM.

That seems fair to me. NBA GMs, in general, are about where the stat community was roughly 15 years ago. While I mean that as an insult of NBA GMs, I also do mean it as a slight compliment of PER, as PER is a lot better than traditional box score stats. It’s like, say, OPS. MLB GMs eventually became comfortable with OPS. Similarly, NBA GMs are now comfortable with the basics of PER, so that does put them ahead of where they were before the days of PER.

#27 I got to say, i am actually a little surprised by that post. I think I was as deep into the thickets with Wins Produced and PER as anybody back then, and I never read that defense by Hollinger.

I stumbled onto that quote when I was trying to mimic PER with a more basic formula.

Most of these models have what I would call a break even point on shooting efficiency. Berri was wrong about PER’s break even point, but he was generally correct that scoring and usage gets more value in PER than in Wins Produced.

I believe using a fixed break even point for measuring how much a player is contributing to the offense is a flaw. I believe if you want to measure a players value to the offense you have to use a more variable analysis that combines efficiency, usage, role, system, teammates, coaching, the defense the player faces, shot selection etc.. Unfortunately, that’s a complex mess that doesn’t translate into numbers very easily. So it gets very subjective and is also prone to error.

I might say I think Porzingis is a better scorer than his TS% indicates and list a bunch of reasons why I think that’s so. There will be resistance to that view because I can’t convert it into a provable number and people are searching for easy answers that don’t seem to exist (at least yet). But I still feel strongly that the factors that impact efficiency (aside from his talents) have been stacked against KP (especially last year). So with some tweaks, better coaching, and better decision making (which is on him), he can become a valuable high usage scorer quite easily. If he also actually improves his skills, he can become VERY valuable.