Kevin Hollis
Former Thunder GM for 7 years
All Star
Posts: 2,838
Dec 16, 2022 11:27:40 GMT
|
Post by Kevin Hollis on Dec 6, 2018 14:51:45 GMT
Just thought I would put this out here. I personally feel like those that post the initial thread for a player should not get a vote (if the player is your own). Lets be honest, the person writing the thread is usually biased towards that player (for the most part, not always true) since in most cases it is their own player. Also, I feel that a lot of people see the initial rating proposal and throw up a similar number without even looking at the player's work (again, definitely not true for everyone). I could write a lot more on the subject, but wanted to see how others feel. Again, just a thought.
|
|
|
Post by Charles Barkley on Dec 6, 2018 19:18:40 GMT
Just thought I would put this out here. I personally feel like those that post the initial thread for a player should not get a vote (if the player is your own). Lets be honest, the person writing the thread is usually biased towards that player (for the most part, not always true) since in most cases it is their own player. Also, I feel that a lot of people see the initial rating proposal and throw up a similar number without even looking at the player's work (again, definitely not true for everyone). I could write a lot more on the subject, but wanted to see how others feel. Again, just a thought. I don't know. On one hand, GM's making threads for their own players means that the initial rating would be biased. But, on the other hand, I do not pay attention enough to other people's players to really feel comfortable making a thread. I do feel comfortable, however, looking at the stats for the player and making a judgment based on those stats. The problem is, GM's all too often just post the thread with a current rating and a proposed rating without any justifications for the proposal, which is pretty bullshit. That, in my opinion, is where we have the problems with persuasion of other GMs. If Josh or Ian, or somebody who is well respected among the league, posts a number, most people gravitate towards that number because of their reputation. But, you also have people who disagree on numbers just solely because they do not like that person. There's no perfect system and inevitably there's going to be problems. Without thinking too much into, I can't think of a better way to go about it. Discarding the initial vote isn't the best thing to do in some scenarios. And, if my vote is not going to count, there's less incentive to make a thread if it is not my player, so there may likely be situations and outcomes where only threads are going to be of GMs recommending their own player, which then just enhances the persuasion because there's more incentive to give an inflated proposal. The problem with our system is two fold: some GMs never vote, some of the GMs that do actually vote just do so without thinking.
|
|
|
Post by Ian Noble on Dec 6, 2018 22:38:42 GMT
The method we use to formulate the eventual rating of a player excludes the most extreme ratings by default and I think that's enough.
E.g. my guy Jaylen Brown is set to increase to an 82 after receiving this set of votes:
79-80-80-80-81-81-82-82-82-82.5-82.5-83-83-83-84-85
|
|
|
Post by George Gervin on Dec 6, 2018 23:15:33 GMT
The method we use to formulate the eventual rating of a player excludes the most extreme ratings by default and I think that's enough. E.g. my guy Jaylen Brown is set to increase to an 82 after receiving this set of votes: 79-80-80-80-81-81-82-82-82-82.5-82.5-83-83- 83-84-85The way Brown has played so far, an 82 seems generous— but agreed if we’re looking at averages on votes tallied it’d be in that range
|
|
Kevin Hollis
Former Thunder GM for 7 years
All Star
Posts: 2,838
Dec 16, 2022 11:27:40 GMT
|
Post by Kevin Hollis on Dec 6, 2018 23:48:28 GMT
The method we use to formulate the eventual rating of a player excludes the most extreme ratings by default and I think that's enough. E.g. my guy Jaylen Brown is set to increase to an 82 after receiving this set of votes: 79-80-80-80-81-81-82-82-82-82.5-82.5-83-83- 83-84-85True, I honestly forgot about the highest deviations being thrown out.
|
|
|
Post by James Kay on Dec 7, 2018 0:07:35 GMT
The method we use to formulate the eventual rating of a player excludes the most extreme ratings by default and I think that's enough. E.g. my guy Jaylen Brown is set to increase to an 82 after receiving this set of votes: 79-80-80-80-81-81-82-82-82-82.5-82.5-83-83- 83-84-85What’s the formula we use? For some reason I thought it was simply the highest and lowest being discarded
|
|
|
Post by Charles Barkley on Dec 7, 2018 4:11:39 GMT
The method we use to formulate the eventual rating of a player excludes the most extreme ratings by default and I think that's enough. E.g. my guy Jaylen Brown is set to increase to an 82 after receiving this set of votes: 79-80-80-80-81-81-82-82-82-82.5-82.5-83-83- 83-84-85What’s the formula we use? For some reason I thought it was simply the highest and lowest being discarded The method we use to formulate the eventual rating or a player excludes the most extreme ratings by default... Highest vote and lowest vote get tossed out, mean of the remaining votes
|
|
|
Post by James Kay on Dec 7, 2018 5:07:14 GMT
What’s the formula we use? For some reason I thought it was simply the highest and lowest being discarded The method we use to formulate the eventual rating or a player excludes the most extreme ratings by default... Highest vote and lowest vote get tossed out, mean of the remaining votes Then why did Ian cross out the highest and lowest three numbers
|
|
|
Post by Ian Noble on Dec 7, 2018 9:59:49 GMT
The method we use to formulate the eventual rating or a player excludes the most extreme ratings by default... Highest vote and lowest vote get tossed out, mean of the remaining votes Then why did Ian cross out the highest and lowest three numbers I know when I did it, it was just arbitrary and depended on the number of votes. So something like 25% of votes to pluck a number out of thin air. Walt Frazier might be able to tell you differently.
|
|
|
Post by Ian Noble on Dec 7, 2018 10:39:40 GMT
The method we use to formulate the eventual rating of a player excludes the most extreme ratings by default and I think that's enough. E.g. my guy Jaylen Brown is set to increase to an 82 after receiving this set of votes: 79-80-80-80-81-81-82-82-82-82.5-82.5-83-83- 83-84-85The way Brown has played so far, an 82 seems generous— but agreed if we’re looking at averages on votes tallied it’d be in that range Sorry what? I can't hear you over the thunderous sound of Jaylen's 21/3/2 in 25 minutes he logged last night
|
|
|
Post by Walt Frazier on Dec 8, 2018 0:15:41 GMT
We've always just removed the very top and very bottom rating. But, we could move to a % possibly. I would say (at a guess) that 50% or more of our votes end up with anywhere from 4-8 votes. Going into the 8-12 vote range I would say we have now accounted for 85% of our threads. So, just taking out the top and bottom pretty much takes out 20-25% on most of our votes, I guess is what I'm saying.
Which means I'm also saying we could turn it into a % and then on the really big votes we would end up taking out 2 or 3 on each end. Having said THAT, once we get to those larger votes, it really tends to average out anyway.
|
|