Back then, it seemed to make sense to run a team-foul RAPM
How to combine the results of that with RAPM, to again have a single-number metric, wasn't immediately obvious
Flaws with RAPM
Re: Flaws with RAPM
This topic was explored on this very board probably 8 years ago. You don't remember?
Back then, it seemed to make sense to run a team-foul RAPM
How to combine the results of that with RAPM, to again have a single-number metric, wasn't immediately obvious
Back then, it seemed to make sense to run a team-foul RAPM
How to combine the results of that with RAPM, to again have a single-number metric, wasn't immediately obvious
Re: Flaws with RAPM
My memory is going....
Yes, that makes sense. Merging it back together would take some thought. I wonder how much of an impact it would have on player ratings? It seems like, in general, this would help starters a bit?
Re: Flaws with RAPM
It should bump those players who tend to play earlier in the quarters rather than later
It might drop players like Ginobili, who rarely started, but often ended games
Most teams start and end games (halfes) with very similar units so not sure if it's ultimately a big bump to starters
The, to me, more obviously benefitting group should be players that draw more fouls. If you only have BoxScores you'd have to look at FTA
But, going back to Ginobili, maybe he helped teammates get fouled with good passes of his? Nash probably a similar case. That's why it makes sense to run an RAPM type analysis
It might drop players like Ginobili, who rarely started, but often ended games
Most teams start and end games (halfes) with very similar units so not sure if it's ultimately a big bump to starters
The, to me, more obviously benefitting group should be players that draw more fouls. If you only have BoxScores you'd have to look at FTA
But, going back to Ginobili, maybe he helped teammates get fouled with good passes of his? Nash probably a similar case. That's why it makes sense to run an RAPM type analysis
Re: Flaws with RAPM
The Athletic has a great article on CP3 changing his game when his team is in the bonus. He tends to draw more non-shooting fouls in those situations and has a disproportionate amount of his FT's from those situations.
https://theathletic.com/1451155/2019/12 ... -rip-move/
Even if adjusting for bonus has a small overall effect, It will have a strong effect on the Offense vs Defense ratings. Your starter might be a league average player with a +1 O RAPM and -1 D RAPM. If he happens to play a lot in the bonus, his overall overall rating will still probably be close to 0 but his Offense might +0.5 and his defense gets boosted to -0.5 so his Offense vs Defense balance gets shifted.
https://theathletic.com/1451155/2019/12 ... -rip-move/
Even if adjusting for bonus has a small overall effect, It will have a strong effect on the Offense vs Defense ratings. Your starter might be a league average player with a +1 O RAPM and -1 D RAPM. If he happens to play a lot in the bonus, his overall overall rating will still probably be close to 0 but his Offense might +0.5 and his defense gets boosted to -0.5 so his Offense vs Defense balance gets shifted.