I thing that bugs me about computer baseball games...

robc · 01-04-2010, 09:06 PM

Sorry about the initial title...

I'm not sure anything can be done about this or not without making the game screwy.

It bothers me that computer baseball games are completely based on randomness. Yes a players ratings may be based on data but because those are used to generate randomness, some suspense or maybe I should say uncertainty is taken away and reduces some strategic elements.

For example, because player X has a contact of 100 / 100, I know that odds are pretty good he is going to be excellent. If he has an 0 for 20 streak (based on random chance) I know I should keep playing him because his odds of getting a hit the next game is just as good as it ever was. It isn't really a slump that is based on his mechanics getting messed up or something else. Since his ratings are good I know I should keep playing him.

Again, I have no idea how to improve upon this because you don't want to make the results of the game less believable or contrived. Does this lessen the strategic elements for anyone else or take away from some of the enjoyment?

I guess some people will say just turn ratings off or really reduce the scale so you can't get a lot of info from them. Maybe that would help, but I'm not sure if it is exactly what I'm talking about.

rutri01 · 01-04-2010, 09:29 PM

I think what you describe is pretty much what happens in real life sor stars. You described not wanting to sit a 100 contact player and the same goes for todays stars in ML baseball. They get sit ti get rested and usually play through slumps. In OOTp the players get tired too and need to be rested.

ctorg · 01-04-2010, 11:20 PM

Your complaint is pretty much exactly what got the whole scouting thing to be the way it is, with unreliable ratings that differ from scout to scout. If a guy goes into a slump, it might well be that his ratings have declined and the scout hasn't realized it yet. Or it might just be an unlucky streak. You don't actually know, if you play with scouts on.

jbergey22 · 01-04-2010, 11:23 PM

You want the "fog of war" concept and so do I

lighthousekeeper · 01-04-2010, 11:29 PM

Quote:

Originally Posted by robc

Sorry about the initial title...

I'm not sure anything can be done about this or not without making the game screwy.

It bothers me that computer baseball games are completely based on randomness. Yes a players ratings may be based on data but because those are used to generate randomness, some suspense or maybe I should say uncertainty is taken away and reduces some strategic elements.

For example, because player X has a contact of 100 / 100, I know that odds are pretty good he is going to be excellent. If he has an 0 for 20 streak (based on random chance) I know I should keep playing him because his odds of getting a hit the next game is just as good as it ever was. It isn't really a slump that is based on his mechanics getting messed up or something else. Since his ratings are good I know I should keep playing him.

Again, I have no idea how to improve upon this because you don't want to make the results of the game less believable or contrived. Does this lessen the strategic elements for anyone else or take away from some of the enjoyment?

I guess some people will say just turn ratings off or really reduce the scale so you can't get a lot of info from them. Maybe that would help, but I'm not sure if it is exactly what I'm talking about.

Yeah I hear ya, but keep in mind:

1. You could use scouting as you indicated above.
2. Remember that it is more complicated than the just "100 rating in contact". Hundreds of other factors work in concert to influence a particular random outcome. (Is the batter a pull hitter, does the opponents 3B have poor range, is the pitcher a flyball pitcher, park factors, ad infinitum...)
3. I do believe ootp models slumps. i could totally be wrong about this, it's just something i suspect. *shrug*

highandoutside · 01-04-2010, 11:41 PM

Turning off ratings and just relying on statistics for my evaluations of talent changed everything for me.

robc · 01-04-2010, 11:45 PM

Quote:

Originally Posted by highandoutside

Turning off ratings and just relying on statistics for my evaluations of talent changed everything for me.

Is the AI then forced to just use statistics then? Do you go and modify the AI evaluation weights to just use statistics?

robc · 01-04-2010, 11:48 PM

Quote:

Originally Posted by ctorg

Your complaint is pretty much exactly what got the whole scouting thing to be the way it is, with unreliable ratings that differ from scout to scout. If a guy goes into a slump, it might well be that his ratings have declined and the scout hasn't realized it yet. Or it might just be an unlucky streak. You don't actually know, if you play with scouts on.

I may be mistaken but typically do the ratings change significantly during the season (barring and injury)?

robc · 01-04-2010, 11:50 PM

Quote:

Originally Posted by jbergey22

You want the "fog of war" concept and so do I

Do you use scounting, turn the ratings off, or keep the rating scale low (like 1-5)?

jbergey22 · 01-04-2010, 11:53 PM

Quote:

Originally Posted by robc

Do you use scounting, turn the ratings off, or keep the rating scale low (like 1-5)?

Its something that is in Worldwide Soccer Manager where you dont get every rating of every player. Id like to see it in OOTP.

You might get 10 out of the possible 30 ratings on a particular player.

akw4572 · 01-05-2010, 09:59 AM

This why I started using Talent only, no real ratings. I pretty much have to use stats only.

Malleus Dei · 01-05-2010, 10:16 AM

Quote:

Originally Posted by highandoutside

Turning off ratings and just relying on statistics for my evaluations of talent changed everything for me.

IMO it's the only way to play. Evaluating players based on just stats makes the game really challenging. You want fog of war? This will give you all that you want and more. Is that new stud in AAA really talented, or is he just on a hot streak? Do you take a chance and promote him or do you leave him in AAA to see if he can maintain the pace? You pay your money and you take your chances.

TribeFanInNC · 01-05-2010, 10:49 AM

Quote:

Originally Posted by Malleus Dei

IMO it's the only way to play. Evaluating players based on just stats makes the game really challenging. You want fog of war? This will give you all that you want and more. Is that new stud in AAA really talented, or is he just on a hot streak? Do you take a chance and promote him or do you leave him in AAA to see if he can maintain the pace? You pay your money and you take your chances.

Totally agree. It takes more time, but I find it a lot more enjoyable as well.

highandoutside · 01-05-2010, 10:57 AM

Quote:

Originally Posted by robc

Is the AI then forced to just use statistics then? Do you go and modify the AI evaluation weights to just use statistics?

Usually the plurality of the AI's evaluation points are distributed to using ratings. Over the course of the season I do change the distribution of these points, but I would say that the plurality is always given to ratings. I do this because I want to make sure the AI is giving me its best game. I haven't noticed anything too strange with this approach. The AI doesn't radically change the role of a bad player who is just having a good month, nor does it relegate to the bench a good player who just is having a rough stretch.

jarmenia · 01-05-2010, 01:07 PM

Quote:

Originally Posted by jbergey22

You want the "fog of war" concept and so do I

Play stats only. That will give you the fog of war you are looking for.

Malleus Dei · 01-05-2010, 02:30 PM

If you set the AI to evaluate players based on 0% ratings, 67% this year's stats, 22% last year's stats, and 11% the year before that's stats when you are playing stats-only you will be pleasantly surprised at the results.

Malleus Dei · 01-05-2010, 02:31 PM

Quote:

Originally Posted by TribeFanInNC

Totally agree. It takes more time, but I find it a lot more enjoyable as well.

It's also a lot harder. Did we mention that?

NHSOXFAN · 01-05-2010, 02:55 PM

The one place I find the "no ratings," approach VERY DIFFICULT is in trade negotiations. Does the AI know the potential ratings of the players being proposed in a trade? I hate to be outsmarted by a Windows PC :-)

EDIT - Never mind. Just had to read two posts above mine to get my answer.

duh

spitfire · 01-05-2010, 03:05 PM

Quote:

Originally Posted by Malleus Dei

If you set the AI to evaluate players based on 0% ratings, 67% this year's stats, 22% last year's stats, and 11% the year before that's stats when you are playing stats-only you will be pleasantly surprised at the results.

If you use that distribution, how well does the computer do early in a season before there are many "this year's stats" to go on? Do you advocate changing the allocations at given points of the season, and if so what is your preferred method?

BMW · 01-05-2010, 03:08 PM

Quote:

Originally Posted by spitfire

If you use that distribution, how well does the computer do early in a season before there are many "this year's stats" to go on? Do you advocate changing the allocations at given points of the season, and if so what is your preferred method?

Good question. I'm itching to know too. I could be itching from other reasons too.

Assume it's because I want to know, however.

01-04-2010, 09:06 PM	#1 (permalink)
robc All Star Starter Join Date: Aug 2008 Posts: 1,660 Thanks: 412 Thanked 251x in 153 posts	A thing that bugs me about computer baseball games... Sorry about the initial title... I'm not sure anything can be done about this or not without making the game screwy. It bothers me that computer baseball games are completely based on randomness. Yes a players ratings may be based on data but because those are used to generate randomness, some suspense or maybe I should say uncertainty is taken away and reduces some strategic elements. For example, because player X has a contact of 100 / 100, I know that odds are pretty good he is going to be excellent. If he has an 0 for 20 streak (based on random chance) I know I should keep playing him because his odds of getting a hit the next game is just as good as it ever was. It isn't really a slump that is based on his mechanics getting messed up or something else. Since his ratings are good I know I should keep playing him. Again, I have no idea how to improve upon this because you don't want to make the results of the game less believable or contrived. Does this lessen the strategic elements for anyone else or take away from some of the enjoyment? I guess some people will say just turn ratings off or really reduce the scale so you can't get a lot of info from them. Maybe that would help, but I'm not sure if it is exactly what I'm talking about. Last edited by robc; 01-04-2010 at 09:15 PM.

01-04-2010, 09:29 PM	#2 (permalink)
rutri01 Major Leagues Join Date: Dec 2007 Posts: 322 Thanks: 15 Thanked 18x in 14 posts	ratings I think what you describe is pretty much what happens in real life sor stars. You described not wanting to sit a 100 contact player and the same goes for todays stars in ML baseball. They get sit ti get rested and usually play through slumps. In OOTp the players get tired too and need to be rested. __________________ Rich Commish: www.ahiba3.com GM: Brooklyn Yard Dawgz

01-04-2010, 11:20 PM	#3 (permalink)
ctorg Global Moderator Join Date: Nov 2002 Location: Brooklyn, NY Posts: 9,701 Thanks: 38 Thanked 91x in 51 posts	Your complaint is pretty much exactly what got the whole scouting thing to be the way it is, with unreliable ratings that differ from scout to scout. If a guy goes into a slump, it might well be that his ratings have declined and the scout hasn't realized it yet. Or it might just be an unlucky streak. You don't actually know, if you play with scouts on. __________________ THE VERY US ARTISTS - A project for musicians and visual artists My music Currently reading: Thirteen by Richard Morgan "When the trees blow back and forth, that's what makes the wind." - Steven Wright Fjord emena pancreas thorax fornicate marmalade morpheme proteolysis smaxa cabana offal srue vitriol grope hallelujah lentils

01-05-2010, 02:55 PM	#18 (permalink)
NHSOXFAN Minors (Single A) Join Date: Jun 2008 Posts: 54 Thanks: 5 Thanked 7x in 2 posts	The one place I find the "no ratings," approach VERY DIFFICULT is in trade negotiations. Does the AI know the potential ratings of the players being proposed in a trade? I hate to be outsmarted by a Windows PC :-) EDIT - Never mind. Just had to read two posts above mine to get my answer. duh Last edited by NHSOXFAN; 01-05-2010 at 03:03 PM.

Thread Tools
Show Printable Version Email this Page
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

01-04-2010, 11:23 PM	#4 (permalink)
jbergey22 Hall Of Famer Join Date: Apr 2005 Location: Minnesota Posts: 3,078 Thanks: 151 Thanked 137x in 105 posts	You want the "fog of war" concept and so do I

01-04-2010, 11:41 PM	#6 (permalink)
highandoutside Major Leagues Join Date: Mar 2004 Posts: 394 Thanks: 34 Thanked 20x in 16 posts	Turning off ratings and just relying on statistics for my evaluations of talent changed everything for me.

01-05-2010, 09:59 AM	#11 (permalink)
akw4572 Hall Of Famer Join Date: Jul 2002 Posts: 2,312 Thanks: 120 Thanked 32x in 25 posts	This why I started using Talent only, no real ratings. I pretty much have to use stats only.