|
|
Post by Walt Frazier on May 8, 2018 16:30:44 GMT
So, since we've figured out that the Dunk rating has a negligible effect on the sim, we've been putting all players at a 60 for their dunk rating.
Since this is a rather big redistribution of ratings for some players, I'm going to begin correcting these around the league. Rather large undertaking, of course.
I'm starting by taking the best player from the top 10 teams in each conference and making sure they have a 60 dunk. Then I'll move to the 2nd highest rated player from each of those teams, etc.
No promises on how far I'll get with these, but it's a worthwhile endeavor with the playoffs quickly approaching.
Ian will put up a similar post like he did today, I'm sure, where he pointed out the FG rating fixed I did the other day (all of those players also got Dunk fixes as part of that process.)
Just want people to be aware that we're aware and are working on it.
|
|
|
|
Post by Hanamichi Sakuragi on May 8, 2018 22:13:11 GMT
wait! Did we prove that already? I think we did not.
|
|
|
|
Post by Walt Frazier on May 9, 2018 2:08:36 GMT
Ian posted it somewhere.
|
|
|
|
Post by Hanamichi Sakuragi on May 9, 2018 2:31:03 GMT
Yes. The result of the statistical analysis, but I don't think it was conclusive. |
|
|
|
Post by Walt Frazier on May 9, 2018 2:49:58 GMT
Yes. The result of the statistical analysis, but I don't think it was conclusive. He posted in the Ratings Committee, it actually looks pretty conclusive. |
|
|
|
Post by Jeremiah Hill on May 9, 2018 3:00:17 GMT
Count me among those against making sweeping changes right before the playoffs
|
|
Chris Mullin
 Golden State Warriors
Starter
Posts: 1,303
Feb 19, 2024 21:58:28 GMT
|
Post by Chris Mullin on May 9, 2018 5:31:23 GMT
Count me among those against making sweeping changes right before the playoffs So, what exactly is happening here? We’re like 8-9 games from the playoffs starting and we’re lowering everyone’s dunk rating to 60 now? So let’s take James Harden for example who has a 99 dunk rating. You’re going to lower his dunk rating to 60? Then what is going to happen to those other 39 rating points? |
|
|
|
Post by Hanamichi Sakuragi on May 9, 2018 6:06:27 GMT
Yes. The result of the statistical analysis, but I don't think it was conclusive. He posted in the Ratings Committee, it actually looks pretty conclusive. I am counting on my memory here. Ian himself made a "caveat" about it in that thread. It was not yet conclusive to do this league-altering adjustment Walt. If my memory is wrong, kindly correct me. |
|
|
|
Post by Andrei Kirilenko on May 9, 2018 6:42:12 GMT
Count me among those against making sweeping changes right before the playoffs So, what exactly is happening here? We’re like 8-9 games from the playoffs starting and we’re lowering everyone’s dunk rating to 60 now? So let’s take James Harden for example who has a 99 dunk rating. You’re going to lower his dunk rating to 60? Then what is going to happen to those other 39 rating points? Basically everyone who previously had a high dunk is getting a 2-3 point overall rating increase, and everyone who had a low dunk (lower than 60) is getting a decrease, because those dunk ratings are being redistributed to ratings that actually matter. One overall rating is about 15 rating points, so a 39 point swing is about 2.5 overall rating increase for a guy like James Harden. |
|
|
|
Post by Ian Noble on May 9, 2018 7:55:14 GMT
Embiggened explanation: For a few months now we've been setting all Dunk ratings to a standard 60 on players who receive rating changes. I discovered (and then later proved more definitively) that the Dunk rating is not a measure of a player's "At Rim" scoring skill but instead is used by the sim engine to determine how flashy a player can dunk in-game, it has no impact when I sim results. Players with higher Dunk ratings were actually at a disadvantage. The higher Dunk rating made no difference to player performance in the sim, and actually took ratings away from a player's other skills since Dunk contributes to a player's overall rating. For instance I noticed with the recent decrease to Jabari Parker that on balance he actually gained in-game effectiveness because when Jabari's Dunk rating was decreased from 90-something to 60 those 30 points were redistributed across his other skills, skills that actually impacted the sim. My team's undergone a lot of rating changes this year, which means all my core players have had their Dunk reduced to 60 (except Jamal Murray, whose Dunk increased to 60), which means my team will receive a slight advantage against other teams that are yet to receive the same treatment. This is obviously a concern because there's a high likelihood that I match up with Charlotte, Milwaukee or Toronto in the First Round of the Playoffs and those teams have a few players with high Dunk ratings (especially Charlotte). It's obviously not fair that a team like mine, with this slight advantage, is matched up with a team like Charlotte (an extreme case) who has Westbrook (99 Dunk), Nerlens (92), Ibaka (85) and Paul George (87) in their starting lineup (see: full player ratings). That's why Walt is going to focus first on teams that have qualified for the Playoffs. |
|
|
|
Post by Jeremiah Hill on May 9, 2018 10:23:13 GMT
Mitchell won a dunk contest and I assume is getting a boost. I'm back in. Chris I guarantee LBJ turns into a god even more based on this lol.
|
|
|
|
Post by Jeremiah Hill on May 9, 2018 10:24:30 GMT
NVM Mitchell has 60 already I'm back to being against it lol.
|
|
|
|
Post by Walt Frazier on May 14, 2018 2:34:52 GMT
I sent the top group to Ian tonight. Trying hard to get at least one more group done for everyone before the playoffs begin. 2 more would be amazing but that'll be a lot of time, we'll see!
|
|
|
|
Post by Ian Noble on May 14, 2018 7:20:12 GMT
|
|
Minnesota Timberwolves
Cleveland Cavaliers
Brooklyn Nets
Boston Celtics
