RAPTOR CORP Leaders

Moderators: Clyde Frazier, Doctor MJ, trex_8063, penbeast0, PaulieWal

Special_Puppy
Assistant Coach
Posts: 3,872
And1: 2,595
Joined: Sep 23, 2023

RAPTOR CORP Leaders 

Post#1 » by Special_Puppy » Sun Jan 19, 2025 1:21 am

I took Historical RAPTOR found here and created a metric called weighted RAPTOR which is just Total RAPTOR WAR with playoff WAR weighted twice as much. I then did a logistic regression where weighted RAPTOR WAR was the x variable and whether that player won the championship that year was a binary 0 or 1. Using that I was able to create a CORP (Championship over Replacement Value) curve that mimics Thinking Basketball's CORP curve (the Constant ended up being -3.7860 and the x variable ended up being 0.1222 BTW). The thing that stands out is how steep the curve is. Players with 10 weighted WAR (think mid 80s Kareem) have a 8-9% chance of winning a championship in any particular year. Players with 20 Weighted WAR (Think Clyde Drexler) have about a 20% chance of winning the championship in any particular season. Players with 30 WAR (GOAT level seasons) have about 45-50% chance of winning the championship in any particular season). Players with a 0 WAR in a season ended up having a 2.2% chance of winning a championship in any given season just by virtue of happening to ride the bench of the championship team so I subtracted that from the CORP values of every player in every year. So if a player had a 10% CORP in a particular year. That was subtracted down to 7.8%. With all that out of the way, here are the CORP Leaders going back to the merger

1. LeBron James 4.61
2. Michael Jordan 4.50
3. John Stockton 3.21
4. Magic Johnson 2.72
5. Chris Paul 2.39
6. Larry Bird 2.28
7. Stephen Curry 2.15
8. Tim Duncan 2.12
9. Kobe Bryant 2.08
10.Scottie Pippen 2.01
11. James Harden 1.82
12. Kevin Garnett 1.76
13. David Robinson 1.72
14. Charles Barkley 1.71
15. Jason Kidd 1.70
16. Hakeem Olajuwon 1.65
17. Clyde Drexler 1.65
18. Karl Malone 1.56
19. Shaquille O'Neal 1.56
20. Dirk Nowitzki 1.53

The biases of RAPTOR are going to show up here. So RAPTOR loves Stockton so RAPTOR CORP loves Stockton. Overall though I am pleasantly surprised by how sensible the results are here.

https://github.com/fivethirtyeight/data/tree/master/nba-raptor
DraymondGold
Senior
Posts: 590
And1: 764
Joined: May 19, 2022

Re: RAPTOR CORP Leaders 

Post#2 » by DraymondGold » Sun Jan 19, 2025 5:26 pm

Special_Puppy wrote:I took Historical RAPTOR found here and created a metric called weighted RAPTOR which is just Total RAPTOR WAR with playoff WAR weighted twice as much. I then did a logistic regression where weighted RAPTOR WAR was the x variable and whether that player won the championship that year was a binary 0 or 1. Using that I was able to create a CORP (Championship over Replacement Value) curve that mimics Thinking Basketball's CORP curve (the Constant ended up being -3.7860 and the x variable ended up being 0.1222 BTW). The thing that stands out is how steep the curve is. Players with 10 weighted WAR (think mid 80s Kareem) have a 8-9% chance of winning a championship in any particular year. Players with 20 Weighted WAR (Think Clyde Drexler) have about a 20% chance of winning the championship in any particular season. Players with 30 WAR (GOAT level seasons) have about 45-50% chance of winning the championship in any particular season). Players with a 0 WAR in a season ended up having a 2.2% chance of winning a championship in any given season just by virtue of happening to ride the bench of the championship team so I subtracted that from the CORP values of every player in every year. So if a player had a 10% CORP in a particular year. That was subtracted down to 7.8%. With all that out of the way, here are the CORP Leaders going back to the merger

1. LeBron James 4.61
2. Michael Jordan 4.50
3. John Stockton 3.21
4. Magic Johnson 2.72
5. Chris Paul 2.39
6. Larry Bird 2.28
7. Stephen Curry 2.15
8. Tim Duncan 2.12
9. Kobe Bryant 2.08
10.Scottie Pippen 2.01
11. James Harden 1.82
12. Kevin Garnett 1.76
13. David Robinson 1.72
14. Charles Barkley 1.71
15. Jason Kidd 1.70
16. Hakeem Olajuwon 1.65
17. Clyde Drexler 1.65
18. Karl Malone 1.56
19. Shaquille O'Neal 1.56
20. Dirk Nowitzki 1.53

The biases of RAPTOR are going to show up here. So RAPTOR loves Stockton so RAPTOR CORP loves Stockton. Overall though I am pleasantly surprised by how sensible the results are here.

https://github.com/fivethirtyeight/data/tree/master/nba-raptor
Cool!

For reference, I made a post comparing a variety of stats’ all-time career ranks using a variety of methods https://forums.realgm.com/boards/viewtopic.php?t=2393827. Comparing this list to Raptor’s published JAWS list (average of career total and best 7 years):
-Both have LeBron and Jordan out in front. CORP seems a little higher on Jordan than JAWS. Given how much more longevity LeBron has / how many years Jordan missed, it’s pretty impressive how small the gap is. Raptor’s a little higher on Jordan’s peak / prime, so given how steep the CORP curve is, it makes sense that Jordan would be boosted up.
-Both have Stockton 3rd, followed by Magic and Chris Paul. It’s pretty high on guard creation it seems, and credits Stockton a lot for his playmaking, efficiency, defensive, and available plus minus stats. Stockton’s ranking here is probably the biggest deviation from people’s ranking of him.
-Bird, Curry, Duncan, and Kobe are 4/5 of the remaining top 10 in both versions. CORP’s a little higher on Curry (over Duncan — wow!), presumably due to that peak ranking.
-Garnett drops from JAWS to CORP, putting him below Pippen and Harden. 7th all time is a bit high for Garnett, but I’d certainly rather have him over Pippen and Harden.
-Kidd, Robinson, Barkley, are common in the next group. CORP is a little lower on Karl Malone (18th vs 15th) and Kidd (15th vs 11th). There’s that high weighting for peaks again, and the high weighting for playmaking.
-Raptor in general is lower on Hakeem and Shaq’s career value compared to the general opinion. Both coasted in the regular season (and Hakeem is usually rated lower than the general opinion across most impact metrics out there), so it makes sense a career value approach that weights regular season games equal to playoff game would be lower on them. If people could somehow mentally separate the regular season and playoff reputation in their heads, would they also be this low on regular season Hakeem and Shaq? Maybe for Hakeem, but Shaq still feels a touch low even with the regular season weighting.

When calculating this, how did you handle playoff WAR? Did you use full-season WAR or just regular season WAR?
If you included playoff WAR, the unequal length playoff runs might bias things slightly. Seasons that have longer playoff runs will naturally accumulate more WAR (more games played), and will also naturally have higher chances at winning the championship (since these seasons have deeper playoff runs).

If we don’t correct for this, this might increase how steep the best-fit curve is (since higher full-season WAR in also measuring deeper playoff runs, not just better quality players).

The simple solution would be to just use RS WAR, but that seems unsatisfying, a performance in the playoffs is pretty critical for CORP.

One way to correct this might be to project everyone’s WAR to be on the same full-season length, accounting for injury. So, e.g., calculate WAR per game, then project onto some length (e.g. an 82 game season, regardless of how deep the playoff run was), then multiply by the percentage of games played to account for injury.
So: WAR_new = (WAR_full_Season / games played) * (82) * (games played / total games they could have played if healthy all season).

It would be interesting to see this CORP calculated for other stats -- WS, Basketball Reference VORP, and PIPM are all available and go back a long time, although all these are of course not as accurate as plus minus era stats -- to see how much the CORP equation changes across stats. Are they all this steeply curved, and if so does that give us a handle and how much we should weigh e.g. peak vs prime vs non-prime years (or more accurately, GOAT vs all time vs MVP vs all-nba etc. years) when comparing players' careers?
Special_Puppy
Assistant Coach
Posts: 3,872
And1: 2,595
Joined: Sep 23, 2023

Re: RAPTOR CORP Leaders 

Post#3 » by Special_Puppy » Sun Jan 19, 2025 7:33 pm

DraymondGold wrote:
Special_Puppy wrote:I took Historical RAPTOR found here and created a metric called weighted RAPTOR which is just Total RAPTOR WAR with playoff WAR weighted twice as much. I then did a logistic regression where weighted RAPTOR WAR was the x variable and whether that player won the championship that year was a binary 0 or 1. Using that I was able to create a CORP (Championship over Replacement Value) curve that mimics Thinking Basketball's CORP curve (the Constant ended up being -3.7860 and the x variable ended up being 0.1222 BTW). The thing that stands out is how steep the curve is. Players with 10 weighted WAR (think mid 80s Kareem) have a 8-9% chance of winning a championship in any particular year. Players with 20 Weighted WAR (Think Clyde Drexler) have about a 20% chance of winning the championship in any particular season. Players with 30 WAR (GOAT level seasons) have about 45-50% chance of winning the championship in any particular season). Players with a 0 WAR in a season ended up having a 2.2% chance of winning a championship in any given season just by virtue of happening to ride the bench of the championship team so I subtracted that from the CORP values of every player in every year. So if a player had a 10% CORP in a particular year. That was subtracted down to 7.8%. With all that out of the way, here are the CORP Leaders going back to the merger

1. LeBron James 4.61
2. Michael Jordan 4.50
3. John Stockton 3.21
4. Magic Johnson 2.72
5. Chris Paul 2.39
6. Larry Bird 2.28
7. Stephen Curry 2.15
8. Tim Duncan 2.12
9. Kobe Bryant 2.08
10.Scottie Pippen 2.01
11. James Harden 1.82
12. Kevin Garnett 1.76
13. David Robinson 1.72
14. Charles Barkley 1.71
15. Jason Kidd 1.70
16. Hakeem Olajuwon 1.65
17. Clyde Drexler 1.65
18. Karl Malone 1.56
19. Shaquille O'Neal 1.56
20. Dirk Nowitzki 1.53

The biases of RAPTOR are going to show up here. So RAPTOR loves Stockton so RAPTOR CORP loves Stockton. Overall though I am pleasantly surprised by how sensible the results are here.

https://github.com/fivethirtyeight/data/tree/master/nba-raptor
Cool!

For reference, I made a post comparing a variety of stats’ all-time career ranks using a variety of methods https://forums.realgm.com/boards/viewtopic.php?t=2393827. Comparing this list to Raptor’s published JAWS list (average of career total and best 7 years):
-Both have LeBron and Jordan out in front. CORP seems a little higher on Jordan than JAWS. Given how much more longevity LeBron has / how many years Jordan missed, it’s pretty impressive how small the gap is. Raptor’s a little higher on Jordan’s peak / prime, so given how steep the CORP curve is, it makes sense that Jordan would be boosted up.
-Both have Stockton 3rd, followed by Magic and Chris Paul. It’s pretty high on guard creation it seems, and credits Stockton a lot for his playmaking, efficiency, defensive, and available plus minus stats. Stockton’s ranking here is probably the biggest deviation from people’s ranking of him.
-Bird, Curry, Duncan, and Kobe are 4/5 of the remaining top 10 in both versions. CORP’s a little higher on Curry (over Duncan — wow!), presumably due to that peak ranking.
-Garnett drops from JAWS to CORP, putting him below Pippen and Harden. 7th all time is a bit high for Garnett, but I’d certainly rather have him over Pippen and Harden.
-Kidd, Robinson, Barkley, are common in the next group. CORP is a little lower on Karl Malone (18th vs 15th) and Kidd (15th vs 11th). There’s that high weighting for peaks again, and the high weighting for playmaking.
-Raptor in general is lower on Hakeem and Shaq’s career value compared to the general opinion. Both coasted in the regular season (and Hakeem is usually rated lower than the general opinion across most impact metrics out there), so it makes sense a career value approach that weights regular season games equal to playoff game would be lower on them. If people could somehow mentally separate the regular season and playoff reputation in their heads, would they also be this low on regular season Hakeem and Shaq? Maybe for Hakeem, but Shaq still feels a touch low even with the regular season weighting.

When calculating this, how did you handle playoff WAR? Did you use full-season WAR or just regular season WAR?
If you included playoff WAR, the unequal length playoff runs might bias things slightly. Seasons that have longer playoff runs will naturally accumulate more WAR (more games played), and will also naturally have higher chances at winning the championship (since these seasons have deeper playoff runs).

If we don’t correct for this, this might increase how steep the best-fit curve is (since higher full-season WAR in also measuring deeper playoff runs, not just better quality players).

The simple solution would be to just use RS WAR, but that seems unsatisfying, a performance in the playoffs is pretty critical for CORP.

One way to correct this might be to project everyone’s WAR to be on the same full-season length, accounting for injury. So, e.g., calculate WAR per game, then project onto some length (e.g. an 82 game season, regardless of how deep the playoff run was), then multiply by the percentage of games played to account for injury.
So: WAR_new = (WAR_full_Season / games played) * (82) * (games played / total games they could have played if healthy all season).

It would be interesting to see this CORP calculated for other stats -- WS, Basketball Reference VORP, and PIPM are all available and go back a long time, although all these are of course not as accurate as plus minus era stats -- to see how much the CORP equation changes across stats. Are they all this steeply curved, and if so does that give us a handle and how much we should weigh e.g. peak vs prime vs non-prime years (or more accurately, GOAT vs all time vs MVP vs all-nba etc. years) when comparing players' careers?


"When calculating this, how did you handle playoff WAR? Did you use full-season WAR or just regular season WAR?" I used weighted WAR which was just Regular Season WAR+(Playoff WAR*2)
Special_Puppy
Assistant Coach
Posts: 3,872
And1: 2,595
Joined: Sep 23, 2023

Re: RAPTOR CORP Leaders 

Post#4 » by Special_Puppy » Sun Jan 19, 2025 7:34 pm

DraymondGold wrote:
Special_Puppy wrote:I took Historical RAPTOR found here and created a metric called weighted RAPTOR which is just Total RAPTOR WAR with playoff WAR weighted twice as much. I then did a logistic regression where weighted RAPTOR WAR was the x variable and whether that player won the championship that year was a binary 0 or 1. Using that I was able to create a CORP (Championship over Replacement Value) curve that mimics Thinking Basketball's CORP curve (the Constant ended up being -3.7860 and the x variable ended up being 0.1222 BTW). The thing that stands out is how steep the curve is. Players with 10 weighted WAR (think mid 80s Kareem) have a 8-9% chance of winning a championship in any particular year. Players with 20 Weighted WAR (Think Clyde Drexler) have about a 20% chance of winning the championship in any particular season. Players with 30 WAR (GOAT level seasons) have about 45-50% chance of winning the championship in any particular season). Players with a 0 WAR in a season ended up having a 2.2% chance of winning a championship in any given season just by virtue of happening to ride the bench of the championship team so I subtracted that from the CORP values of every player in every year. So if a player had a 10% CORP in a particular year. That was subtracted down to 7.8%. With all that out of the way, here are the CORP Leaders going back to the merger

1. LeBron James 4.61
2. Michael Jordan 4.50
3. John Stockton 3.21
4. Magic Johnson 2.72
5. Chris Paul 2.39
6. Larry Bird 2.28
7. Stephen Curry 2.15
8. Tim Duncan 2.12
9. Kobe Bryant 2.08
10.Scottie Pippen 2.01
11. James Harden 1.82
12. Kevin Garnett 1.76
13. David Robinson 1.72
14. Charles Barkley 1.71
15. Jason Kidd 1.70
16. Hakeem Olajuwon 1.65
17. Clyde Drexler 1.65
18. Karl Malone 1.56
19. Shaquille O'Neal 1.56
20. Dirk Nowitzki 1.53

The biases of RAPTOR are going to show up here. So RAPTOR loves Stockton so RAPTOR CORP loves Stockton. Overall though I am pleasantly surprised by how sensible the results are here.

https://github.com/fivethirtyeight/data/tree/master/nba-raptor
Cool!

For reference, I made a post comparing a variety of stats’ all-time career ranks using a variety of methods https://forums.realgm.com/boards/viewtopic.php?t=2393827. Comparing this list to Raptor’s published JAWS list (average of career total and best 7 years):
-Both have LeBron and Jordan out in front. CORP seems a little higher on Jordan than JAWS. Given how much more longevity LeBron has / how many years Jordan missed, it’s pretty impressive how small the gap is. Raptor’s a little higher on Jordan’s peak / prime, so given how steep the CORP curve is, it makes sense that Jordan would be boosted up.
-Both have Stockton 3rd, followed by Magic and Chris Paul. It’s pretty high on guard creation it seems, and credits Stockton a lot for his playmaking, efficiency, defensive, and available plus minus stats. Stockton’s ranking here is probably the biggest deviation from people’s ranking of him.
-Bird, Curry, Duncan, and Kobe are 4/5 of the remaining top 10 in both versions. CORP’s a little higher on Curry (over Duncan — wow!), presumably due to that peak ranking.
-Garnett drops from JAWS to CORP, putting him below Pippen and Harden. 7th all time is a bit high for Garnett, but I’d certainly rather have him over Pippen and Harden.
-Kidd, Robinson, Barkley, are common in the next group. CORP is a little lower on Karl Malone (18th vs 15th) and Kidd (15th vs 11th). There’s that high weighting for peaks again, and the high weighting for playmaking.
-Raptor in general is lower on Hakeem and Shaq’s career value compared to the general opinion. Both coasted in the regular season (and Hakeem is usually rated lower than the general opinion across most impact metrics out there), so it makes sense a career value approach that weights regular season games equal to playoff game would be lower on them. If people could somehow mentally separate the regular season and playoff reputation in their heads, would they also be this low on regular season Hakeem and Shaq? Maybe for Hakeem, but Shaq still feels a touch low even with the regular season weighting.

When calculating this, how did you handle playoff WAR? Did you use full-season WAR or just regular season WAR?
If you included playoff WAR, the unequal length playoff runs might bias things slightly. Seasons that have longer playoff runs will naturally accumulate more WAR (more games played), and will also naturally have higher chances at winning the championship (since these seasons have deeper playoff runs).

If we don’t correct for this, this might increase how steep the best-fit curve is (since higher full-season WAR in also measuring deeper playoff runs, not just better quality players).

The simple solution would be to just use RS WAR, but that seems unsatisfying, a performance in the playoffs is pretty critical for CORP.

One way to correct this might be to project everyone’s WAR to be on the same full-season length, accounting for injury. So, e.g., calculate WAR per game, then project onto some length (e.g. an 82 game season, regardless of how deep the playoff run was), then multiply by the percentage of games played to account for injury.
So: WAR_new = (WAR_full_Season / games played) * (82) * (games played / total games they could have played if healthy all season).

It would be interesting to see this CORP calculated for other stats -- WS, Basketball Reference VORP, and PIPM are all available and go back a long time, although all these are of course not as accurate as plus minus era stats -- to see how much the CORP equation changes across stats. Are they all this steeply curved, and if so does that give us a handle and how much we should weigh e.g. peak vs prime vs non-prime years (or more accurately, GOAT vs all time vs MVP vs all-nba etc. years) when comparing players' careers?


538 made a CORP like curve for WS a while back https://fivethirtyeight.com/features/lebrons-odds-of-catching-jordan/#:~:text=But%20four%20others%20%E2%80%94%20Jordan%2C%20Bill,13%2C%20or%20about%2030%20percent.
Special_Puppy
Assistant Coach
Posts: 3,872
And1: 2,595
Joined: Sep 23, 2023

Re: RAPTOR CORP Leaders 

Post#5 » by Special_Puppy » Sun Jan 19, 2025 7:37 pm

DraymondGold wrote:
Special_Puppy wrote:I took Historical RAPTOR found here and created a metric called weighted RAPTOR which is just Total RAPTOR WAR with playoff WAR weighted twice as much. I then did a logistic regression where weighted RAPTOR WAR was the x variable and whether that player won the championship that year was a binary 0 or 1. Using that I was able to create a CORP (Championship over Replacement Value) curve that mimics Thinking Basketball's CORP curve (the Constant ended up being -3.7860 and the x variable ended up being 0.1222 BTW). The thing that stands out is how steep the curve is. Players with 10 weighted WAR (think mid 80s Kareem) have a 8-9% chance of winning a championship in any particular year. Players with 20 Weighted WAR (Think Clyde Drexler) have about a 20% chance of winning the championship in any particular season. Players with 30 WAR (GOAT level seasons) have about 45-50% chance of winning the championship in any particular season). Players with a 0 WAR in a season ended up having a 2.2% chance of winning a championship in any given season just by virtue of happening to ride the bench of the championship team so I subtracted that from the CORP values of every player in every year. So if a player had a 10% CORP in a particular year. That was subtracted down to 7.8%. With all that out of the way, here are the CORP Leaders going back to the merger

1. LeBron James 4.61
2. Michael Jordan 4.50
3. John Stockton 3.21
4. Magic Johnson 2.72
5. Chris Paul 2.39
6. Larry Bird 2.28
7. Stephen Curry 2.15
8. Tim Duncan 2.12
9. Kobe Bryant 2.08
10.Scottie Pippen 2.01
11. James Harden 1.82
12. Kevin Garnett 1.76
13. David Robinson 1.72
14. Charles Barkley 1.71
15. Jason Kidd 1.70
16. Hakeem Olajuwon 1.65
17. Clyde Drexler 1.65
18. Karl Malone 1.56
19. Shaquille O'Neal 1.56
20. Dirk Nowitzki 1.53

The biases of RAPTOR are going to show up here. So RAPTOR loves Stockton so RAPTOR CORP loves Stockton. Overall though I am pleasantly surprised by how sensible the results are here.

https://github.com/fivethirtyeight/data/tree/master/nba-raptor
Cool!

For reference, I made a post comparing a variety of stats’ all-time career ranks using a variety of methods https://forums.realgm.com/boards/viewtopic.php?t=2393827. Comparing this list to Raptor’s published JAWS list (average of career total and best 7 years):
-Both have LeBron and Jordan out in front. CORP seems a little higher on Jordan than JAWS. Given how much more longevity LeBron has / how many years Jordan missed, it’s pretty impressive how small the gap is. Raptor’s a little higher on Jordan’s peak / prime, so given how steep the CORP curve is, it makes sense that Jordan would be boosted up.
-Both have Stockton 3rd, followed by Magic and Chris Paul. It’s pretty high on guard creation it seems, and credits Stockton a lot for his playmaking, efficiency, defensive, and available plus minus stats. Stockton’s ranking here is probably the biggest deviation from people’s ranking of him.
-Bird, Curry, Duncan, and Kobe are 4/5 of the remaining top 10 in both versions. CORP’s a little higher on Curry (over Duncan — wow!), presumably due to that peak ranking.
-Garnett drops from JAWS to CORP, putting him below Pippen and Harden. 7th all time is a bit high for Garnett, but I’d certainly rather have him over Pippen and Harden.
-Kidd, Robinson, Barkley, are common in the next group. CORP is a little lower on Karl Malone (18th vs 15th) and Kidd (15th vs 11th). There’s that high weighting for peaks again, and the high weighting for playmaking.
-Raptor in general is lower on Hakeem and Shaq’s career value compared to the general opinion. Both coasted in the regular season (and Hakeem is usually rated lower than the general opinion across most impact metrics out there), so it makes sense a career value approach that weights regular season games equal to playoff game would be lower on them. If people could somehow mentally separate the regular season and playoff reputation in their heads, would they also be this low on regular season Hakeem and Shaq? Maybe for Hakeem, but Shaq still feels a touch low even with the regular season weighting.

When calculating this, how did you handle playoff WAR? Did you use full-season WAR or just regular season WAR?
If you included playoff WAR, the unequal length playoff runs might bias things slightly. Seasons that have longer playoff runs will naturally accumulate more WAR (more games played), and will also naturally have higher chances at winning the championship (since these seasons have deeper playoff runs).

If we don’t correct for this, this might increase how steep the best-fit curve is (since higher full-season WAR in also measuring deeper playoff runs, not just better quality players).

The simple solution would be to just use RS WAR, but that seems unsatisfying, a performance in the playoffs is pretty critical for CORP.

One way to correct this might be to project everyone’s WAR to be on the same full-season length, accounting for injury. So, e.g., calculate WAR per game, then project onto some length (e.g. an 82 game season, regardless of how deep the playoff run was), then multiply by the percentage of games played to account for injury.
So: WAR_new = (WAR_full_Season / games played) * (82) * (games played / total games they could have played if healthy all season).

It would be interesting to see this CORP calculated for other stats -- WS, Basketball Reference VORP, and PIPM are all available and go back a long time, although all these are of course not as accurate as plus minus era stats -- to see how much the CORP equation changes across stats. Are they all this steeply curved, and if so does that give us a handle and how much we should weigh e.g. peak vs prime vs non-prime years (or more accurately, GOAT vs all time vs MVP vs all-nba etc. years) when comparing players' careers?


I'm very interested in calculating a CORP curve for VORP and PIPM. Not super easy to collect the data for that though
User avatar
eminence
RealGM
Posts: 16,857
And1: 11,694
Joined: Mar 07, 2015

Re: RAPTOR CORP Leaders 

Post#6 » by eminence » Sun Jan 19, 2025 8:07 pm



I originally used Fibonacci for my rough CORP tier list, and it seems to match up reasonably well to what 538 came up with there.

1 - Deep Bench
2 - Bench
3 - Starter
5 - Allstar
8 - AllNBA
13 - MVP Contender/1st Team Guy
21 - ATG

More recently have switched to Caterer to flatten the curve somewhat (general belief that there's value in adding wins to a team prior to the team entering championship contention levels). Generally rewards longevity more.

1 - Deep Bench
2 - Bench
4 - Starter
7 - Allstar
11 - AllNBA
16 - MVP Contender/1st Team Guy
22 - ATG
I bought a boat.
Special_Puppy
Assistant Coach
Posts: 3,872
And1: 2,595
Joined: Sep 23, 2023

Re: RAPTOR CORP Leaders 

Post#7 » by Special_Puppy » Sun Jan 19, 2025 8:13 pm

eminence wrote:


I originally used Fibonacci for my rough CORP tier list, and it seems to match up reasonably well to what 538 came up with there.

1 - Deep Bench
2 - Bench
3 - Starter
5 - Allstar
8 - AllNBA
13 - MVP Contender/1st Team Guy
21 - ATG

More recently have switched to Caterer to flatten the curve somewhat (general belief that there's value in adding wins to a team prior to the team entering championship contention levels). Generally rewards longevity more.

1 - Deep Bench
2 - Bench
4 - Starter
7 - Allstar
11 - AllNBA
16 - MVP Contender/1st Team Guy
22 - ATG


Using JAWS or weighing the best season 100%, the 2nd best season 95%, 3rd best season 90%, matches the rank order of the CORP list pretty well viewtopic.php?p=114869305#p114869305
penbeast0
Senior Mod - NBA Player Comparisons
Senior Mod - NBA Player Comparisons
Posts: 30,168
And1: 9,780
Joined: Aug 14, 2004
Location: South Florida
 

Re: RAPTOR CORP Leaders 

Post#8 » by penbeast0 » Mon Jan 20, 2025 4:40 am

How far back do the years you are looking at go?
“Most people use statistics like a drunk man uses a lamppost; more for support than illumination,” Andrew Lang.
Special_Puppy
Assistant Coach
Posts: 3,872
And1: 2,595
Joined: Sep 23, 2023

Re: RAPTOR CORP Leaders 

Post#9 » by Special_Puppy » Mon Jan 20, 2025 11:44 pm

penbeast0 wrote:How far back do the years you are looking at go?


1976

Return to Player Comparisons