Make the AI pay attention to stats.

[knockahoma715]

I've become a big fan of Mogul. Overall, I prefer this product now to OOTP-- which is saying quite a lot. A few years ago, I thought OOTP was far ahead.

But I do like OOTP's option of "weighting" the AI's decision-making regarding lineups. For example, you can weight it 60% toward stats, 40% toward ratings. As the year goes along, you simply increase stats percentage to mock a manager saying, " Hey, this just ain't Ferguson's year; I'm gonna play Yeager more."

By weighting the stats to be more favorable, you don't get those guys disappearing into AAA after hitting 20 homers the year before ( I had this happen to OF Charlie Spikes of Cleveland upon starting a 74 league). In 73, Spikes hit 20 roundtrippers only to find himself down in AAA at the start of 74. And it wasn't as if Cleveland had a lot going on in the OF. Spikes was young and I'm sure the AI figured he'd be better off being seasoned than spending the year on the bench in Cleveland. So, not totally illogical. But still, it would be nice to have this sort of a weight system.

Best,

Knocky

[beerchaser]

Hey knocky,

This was being discussed on another thread last week. I think there's a chicken/egg thing going on with Mogul here. I agree with you that stats sometimes get ignored in Mogul as I've seen some good stats put up by players with crappy ratings and vice versa. However, Ohms says that Mogul uses the ratings to generate the stats.

[FRENCHREDSOX]

Hey knocky,

This was being discussed on another thread last week. I think there's a chicken/egg thing going on with Mogul here. I agree with you that stats sometimes get ignored in Mogul as I've seen some good stats put up by players with crappy ratings and vice versa. However, Ohms says that Mogul uses the ratings to generate the stats.

Yep RATINGS determine the stats but ratings themselves are comparative to players at the same position ie a 75 1B & a 75 CF do not mean they will hit the same (75 = league average) & since the patches fielding ratings are "more" weighted than previously.So resumé this:

1) Stats sre determined by the POSITION PLAYER'S RATING ;

2) FIELDING can drag down/increase the rating but doesnt affect his hitting capability;

3) there is random element that can allow a 75 rated player to over perform/under perform

[HoustonGM]

http://forum.sportsmogul.com/showthread.php?t=117356

[6leopard6]

i can't imagine there'd be a game where ratings DON'T generate the stats, if it's a simulation. it doesn't mean that the stats CAN'T BE LOOKED AT by the AI. it just isn't currently programmed to. and judging by certain responses, is unfeasible at this time to do - in my COMPLETELY uninformed opinion

[dolfanar]

I think that some sort of weighting system for the AUTO-SORT would be a great thing, not for performance.

So, for example if you set it to 100% on Ratings, you would have the current format where stats and actual performance are ignored, and predicted stats and scouting reports are used to determine lineups, and so on.

Ofcourse this argument has been beaten into the ground at this point. I'll say it again though, as long as the AI is sending proven 30 HR hitters to AAA just because their ratings are bad without even trying to see what they can do in the majors, then BBM has a serious problem.

[ohms_law]

http://forum.sportsmogul.com/showthread.php?p=644592#post644592

[knockahoma715]

Yes. I agree this is a sorting issue, not a ratings/stats problem.


As I wrote earlier, I loved OOTP's AI weighting system. The User decides what is "realistic" by weighting Ratings vs Stats. You even have the option of how much to rate past years' statistics. For example:

At the beginning of the year 1975, you might weight the AI's roster sort system this way:

1974 stats 50%
1973 stats 25%
Current Rating 25%

In July, you might weight those categories like this:

1975 stats: 35%
1974 stats 35%
Current Rating 30%

In September:

1975 stats: 90%


I think BBM's interface and simplicity make it a superior product for me. But in this one area, OOTP has the "simplest", yet most effective idea. Let the user weight the AI Roster sorter. Then EVERY user is happy with his version of what he believes to be a "realistic universe"-- even if the universe is actually whacked.

Ignorance is bliss. :D