Evaluation - Ratings vs Stats

X3NEIZE · 05-23-2016, 03:05 PM

I know, I know this is the eternal debate....

But can someone that has tried both (0/67/22/11 and a more traditional 60/20/10/5) give me the sales pitch as to why one is better than the other?

As I understand, all that does is change the way scouts (or you) see the different players "stars", correct? I would assume that having 0 in ratings leans to increasing the starts on players performing well on current season (or past couple) vs a player that can be hitting .190 and last name being Trout still showing a 5 star..... is my assumption correct?

In other words the "hiding" potential is completely not affected at all by this setting, right?

Statsman1 · 05-23-2016, 03:15 PM

Deleted...sorry about that.

NoOne · 05-24-2016, 12:08 PM

percentages are mostly personal preference. think of what the ramifications are... how do you want to weight the years? the results won't change that much as long as you don't have 100% scouting accuracy... then things will change alot with stats being introduced to the equation.

i figure i want he scout to be nearly 50% of the evaluation, with stats making the rest. i do 40ratings 36/18/6 - so the current year is 2X last year and 4X 3 years ago in importance. there's no magic bullet in regard to the percentages you choose.

the greater the influence of stats, the greater amount of error you introduce to the equation. your scouting accuracy will cause scouted ratings to have errors, too.

so, depending on these settings it makes GMs less perfect, in a sense. it's just a way to make things less clear to your and the AI. the less accurate your scout and ai evaluation, the more difficult it is to make the 'best' decision.

e.g. playing 100% stats and no ratings means you have to rely completely on statistics... which may or may not be a large enough sample size available.

some say this is more realistic, but i think it's missing some feedback that a manager would get in real life. this is why i like introducing error to the scouting ratings and also using stats as part of AI Evaluation. whether somethign is realistic or not shouldn't prevent you from doing anythign that you like, though. don't get caught in that nonsense. this is just entertainment... be entertained.

ai eval only only affects overall/potential - whichever scale you are using won't matter.

05-23-2016, 03:05 PM	#1
X3NEIZE All Star Reserve Join Date: Aug 2012 Posts: 649	Evaluation - Ratings vs Stats I know, I know this is the eternal debate.... But can someone that has tried both (0/67/22/11 and a more traditional 60/20/10/5) give me the sales pitch as to why one is better than the other? As I understand, all that does is change the way scouts (or you) see the different players "stars", correct? I would assume that having 0 in ratings leans to increasing the starts on players performing well on current season (or past couple) vs a player that can be hitting .190 and last name being Trout still showing a 5 star..... is my assumption correct? In other words the "hiding" potential is completely not affected at all by this setting, right?

05-23-2016, 03:15 PM	#2
Statsman1 Major Leagues Join Date: Sep 2002 Location: Marmora, Ontario, Canada Posts: 340	Deleted...sorry about that. __________________ Statsman1 Last edited by Statsman1; 05-23-2016 at 03:17 PM.

05-24-2016, 12:08 PM	#3
NoOne Banned Join Date: Apr 2015 Posts: 7,273 Infractions: 0/1 (3)	percentages are mostly personal preference. think of what the ramifications are... how do you want to weight the years? the results won't change that much as long as you don't have 100% scouting accuracy... then things will change alot with stats being introduced to the equation. i figure i want he scout to be nearly 50% of the evaluation, with stats making the rest. i do 40ratings 36/18/6 - so the current year is 2X last year and 4X 3 years ago in importance. there's no magic bullet in regard to the percentages you choose. the greater the influence of stats, the greater amount of error you introduce to the equation. your scouting accuracy will cause scouted ratings to have errors, too. so, depending on these settings it makes GMs less perfect, in a sense. it's just a way to make things less clear to your and the AI. the less accurate your scout and ai evaluation, the more difficult it is to make the 'best' decision. e.g. playing 100% stats and no ratings means you have to rely completely on statistics... which may or may not be a large enough sample size available. some say this is more realistic, but i think it's missing some feedback that a manager would get in real life. this is why i like introducing error to the scouting ratings and also using stats as part of AI Evaluation. whether somethign is realistic or not shouldn't prevent you from doing anythign that you like, though. don't get caught in that nonsense. this is just entertainment... be entertained. ai eval only only affects overall/potential - whichever scale you are using won't matter. Last edited by NoOne; 05-24-2016 at 12:10 PM.