APBRmetrics

Hey guys,

If you guys have been reading my posts then you know Ibaka has been near the top of my rankings for a while, and that I think he is underrated. I recently did a post on him on my blog that I won't attempt to summarize here. You can check it out by going to link in my signature.

Basically I was wondering why you guys and people in general don't regard Ibaka as highly as I do. I thought he should be an all-star lock, yet no one seemed upset when he didn't get it. Do your metrics like RAPM say that he actually a fairly average player? If so, in what categories is he underperforming in?

Sorry to start a whole new thread for this, but for he me he's one of the most fascinating players given the large discrepancy between how my stat ranks him and how other stats rank him. I would love to get your opinion on this.

Thanks

Glancing over his 82games page it's easy to see why RAPM doesn't like him *that* much

Offense is 3 points better without him, defense is almost unchanged with him in or out. OKC grabs a way higher % of offensive rebounds without him, and also a higher % of defensive rebounds. They turn it over more when he's on the floor and they cause more opponent turnovers when he's off.

All of that doesn't scream "scrub", especially not on a team like OKC but he'd have a better case if those numbers were more in his favor

Ibaka currently has a RAPM estimated impact of +0.7 per 100 possessions at JE's site. Down from last season. Only 11 players making about as much or more have worse ratings. By RAPM right he is at best an acceptable performance / cost value and I'd say not that really that acceptable.

I'd like to see RAPM splits for him with and without Perkins and at PF vs C.

With your ratings for Ibaka on Rebounding, Rim Defense and Shooting one might expect him to be very strong overall and he is not on RAPM. His raw on/off data hints that turnovers (own and forced) might be a big part of where he hurts the team some to make him only very modestly positive on overall RAPM estimate. Team free throws attempted are also down with him on court vs off.

his old RAPM 4 factors: https://docs.google.com/spreadsheet/ccc ... PWkE#gid=0


This is an old thread on Ibaka from 2012: viewtopic.php?f=2&t=7937&hilit=ibaka

Thanks for the link J.E, I'll definitely check that out. I wonder if the change in rebounding percentage you mention could be from the way OKC plays with him. He probably operates farther from the basket than his replacements (Perkins, Adams, Collison) who aren't as good of shooter as he is. That could help explain why OKC grabs so many more offensive rebounds without him on the floor.

And Crow you could be right. His steals PAA is low, and although I haven't found a way to include turnovers he is averaging a career high (tied with last year) in that category, so that could be hurting his performance. I'll definitely search for that thread too.

On a side note, I would love to see a Clippers-Thunder playoff series, so we would get a chance to see Ibaka vs Griffin. Despite Griffin's great play of late, I think it would be a great matchup - and, I may be in the minority here, but I would actually expect Ibaka to win it...

It confuses me a bit too, in terms of the eye test. Elite rim protection, elite mid range shooting, great spacing from/for a big, and he's a way improved pnr defender now.

But as JE/Crow said, OKC just doesn't seem to play any better with him than they do without him.

His on/off (and hence RAPM values) could still be a bit of an sss thing, I guess.

I'd say Ibaka is likely the best #3 player on any team. Bosh would be the other possibility (though some like DeAndre Jordan).
He's a marginal all-star level player. And he's in the West, so no allstar game.

Given the number of shots he blocks, does he not rebound so well? Here are players with career Block% > 6, along with their Reb%: Minimum 10,000 minutes.
Serge will likely see his Blk% drop as he ages, perhaps more than his Reb%. But he'll still hold his own among these. None are higher in both measures; only Zo was a bigger scorer; all the others were primarily Centers.

I went ahead and looked at his ratio of what I consider "good blocks" (blocks followed by a defensive rebound) vs. "bad blocks" (..followed by an offensive rebound). His ratio is extremely good - only Bogut has a higher ratio for players with a minimum of 30 blocks this season.

My own SPM ranks Ibaka at ~15, and I assume that's the ballpark where a lot of BoxScore metrics have him. Since the combination of BoxScore metrics+RAPM is better than either on its own his "actual" rank among NBA players is likely to be in the middle (50-70). I can definitely see that

JE, your SPM ranks Ibaka around #15, with or without consideration of his 'good blocks' ratio?
With my (team based retro plus-minus) eWins, he ranks around #30. He'd be the best player on a number of teams -- Bos, Brk (wo Lopez), Den, Mil, Orl, Phl, Utah.

Two years ago, he was blocking half again as many shots. Do you know what his good block ratio was in 2011-12?

This year, OKC is blocking 92% as many shots on the road as they are at home. That's a bit less disparity than the league avg of 90%.

He's #31 right now in IPV (boxscore/demographic informed RAPM) at +2.9, right below Wall, and right above Kawhi.

Though I'm not a big fan of using these as rankings per se (same with RAPM in general) due to how context-specific this stuff is.

JE's good blocks vs bad blocks comment is mad cool.

Mike G wrote:JE, your SPM ranks Ibaka around #15, with or without consideration of his 'good blocks' ratio?

Two years ago, he was blocking half again as many shots. Do you know what his good block ratio was in 2011-12?

My SPM gets fed the raw number of good (bad) blocks per minute, but not the ratio. The coefficient (for defense) for good blocks was ~3.5 times higher than the one for bad blocks, so these seem to be much more valuable (at least they help more in predicting future defensive performance)

His good/bad ratio for this season is 2.5. In '12 it was ~1. That being said, he still led the league in good blocks that year, by a wide margin

JE's good blocks vs bad blocks comment is mad cool.

Thank you thank you

J.E. wrote: His good/bad ratio for this season is 2.5. In '12 it was ~1. That being said, he still led the league in good blocks that year, by a wide margin

JE's good blocks vs bad blocks comment is mad cool.

Yes, Boston sportswriters claimed that Bill Russell's blocks would often result in defensive rebounds by the Celts whereas other guys' blocks often went out of bounds resulting in a team offensive rebound. They might've been right, but who knows? So it's cool to see someone tracking this.

But is it a repeatable skill? The bit of stats we see on Ibaka suggests that it is; i.e. it's not just luck that his blocks tend to be "good blocks", he genuinely consistently does a better job of getting good blocks year after year? And conversely the players who tend to get bad blocks also do so year after year?

mtamada wrote:But is it a repeatable skill? The bit of stats we see on Ibaka suggests that it is; i.e. it's not just luck that his blocks tend to be "good blocks", he genuinely consistently does a better job of getting good blocks year after year? And conversely the players who tend to get bad blocks also do so year after year?

It's a very subjective observation (and small sample size because I don't watch too many games) but to me there are a couple of players in the league that could potentially block opponents in a way so that the defense gets the rebound, but choose not to - I'm guessing because they want to "send a message" or make the block look more "sick" (whatever you want to call it). I'd say McGee and DeAndre Jordan, at least at some point in their career, belong(ed) in that category

Obviously the ratio of "good" to "bad" blocks is somewhat dependent on how good your teammates are at gobbling up loose balls

This is my all-time favorite block btw

The good/bad block ratio goes back to an 82games article -- and obviously it goes back to the days of Russell too.

http://www.82games.com/comm16.htm

There's also this Sloan study on the value of a blocked shot:
http://www.sloansportsconference.com/?p=2811

By the way, NBAWOWY keeps track of this stat now. Go over to the defense tab and refer to BLK-DRB%. Over the past couple years, for example, Ibaka's blocks have been rebounded at 65.5%.

It also includes the DRB% of the player grabbing it after his own block. Duncan appears to be good at this.

Theoretically, the fact that Ibaka's back-up (Collision) plays so well according to RAPM should NOT effect Ibaka's RAPM score right?

Yes but there could be some impact.