player eval AI settings

[ThePride87]

I'm weird, I like 100/0/0/0. More of a fan of process > results, feel like it gives the AI a better chance to beat me not over-reacting to anomalies. Plus, who cares what someone did 2 years ago? Irrelevant info IMO.

[SirMichaelJordan]

Originally Posted by ThePride87

I'm weird, I like 100/0/0/0. More of a fan of process > results, feel like it gives the AI a better chance to beat me not over-reacting to anomalies. Plus, who cares what someone did 2 years ago? Irrelevant info IMO.

Sweed explained it perfectly. Different strokes for different folks. If AI releasing a performing player because of rating drops is optimal. 100/0/0 is great.

[ThePride87]

Where was this convo last year? You and Sweed were starting to really go deep into the topic, then just quit or something....it was actually one of the best forum topics of 21. I remember 25/25/25/25 being discussed, and I was starting to buy-in until it was just dropped. What Sweed is saying makes sense to me, maybe I should try it.

[RchW]

Originally Posted by SirMichaelJordan

Sweed explained it perfectly. Different strokes for different folks. If AI releasing a performing player because of rating drops is optimal. 100/0/0 is great.

It seems clear that 50/50/0/0 would temper that overreaction which is why I use 40/45/10/5 now. I have used 45/45/10/0 and 45/40/10/5 too.

I run two fictional MLB clones that are 75 (above) and 100 seasons into play. The longer one is currently 45/40/15.

[oomm]

Originally Posted by RchW

I don't understand the bold. What is this "data" if it is not the internal 1-250 ratings, "under the hood"?

Stats.



And "just" ratings, too. And interpretation of same by ingame decision-makers.

[Sweed]

Originally Posted by ThePride87

Where was this convo last year? You and Sweed were starting to really go deep into the topic, then just quit or something....it was actually one of the best forum topics of 21. I remember 25/25/25/25 being discussed, and I was starting to buy-in until it was just dropped. What Sweed is saying makes sense to me, maybe I should try it.

Well to be fair I was just following the trail SMJ was blazing. I could have sworn I went back into the thread to post I had gone to the 4-25's settings. I see I left it off where I was a week or two away from a new season but was thinking I would go to the 25's. I think the timing ended up badly because I bought a new computer, a patch came out, and OOTP kept crashing to a system reboot on my new rig. A long story in itself but it ended up being the machine and not OOTP. Acer sent me a new rig to replace it as they couldn't find the cause after two trips to their service center

In any case that is where I ended up and haven't switched off it yet.

[oomm]

Originally Posted by RchW

It seems clear that 50/50/0/0 would temper that overreaction which is why I use 40/45/10/5 now. I have used 45/45/10/0 and 45/40/10/5 too.

I run two fictional MLB clones that are 75 (above) and 100 seasons into play. The longer one is currently 45/40/15.

...and?

[SirMichaelJordan]

Originally Posted by RchW

It seems clear that 50/50/0/0 would temper that overreaction which is why I use 40/45/10/5 now. I have used 45/45/10/0 and 45/40/10/5 too.

I run two fictional MLB clones that are 75 (above) and 100 seasons into play. The longer one is currently 45/40/15.

That’s a good one, I’ve played around with that a lot this year as well.

[RchW]

Originally Posted by oomm

...and?

what?

[ThePride87]

Originally Posted by Sweed

Well to be fair I was just following the trail SMJ was blazing. I could have sworn I went back into the thread to post I had gone to the 4-25's settings. I see I left it off where I was a week or two away from a new season but was thinking I would go to the 25's. I think the timing ended up badly because I bought a new computer, a patch came out, and OOTP kept crashing to a system reboot on my new rig. A long story in itself but it ended up being the machine and not OOTP. Acer sent me a new rig to replace it as they couldn't find the cause after two trips to their service center

In any case that is where I ended up and haven't switched off it yet.

Well hey, I respect it....when I say 100/0/0/0, that's just me talkin' **** because that's my initial mentality....not like I got evidence to back it up. I just assume it's a battle of ratings vs results....but as I read your info and thought hey, maybe I should've read before I talked.....maybe SirJordan had a point. I get where you are coming from and can understand why 25/25/25/25 is good.....it's kinda where I was thinking myself last year. Cheers my friend.

[Reed]

If I am playing historical, I go 100/0/0/0 to minimize lopsided trades. Playing fictional I just leave it at default..

[oomm]

Originally Posted by RchW

what?

well, from over 175 seasons perhaps there are a few things you've noticed that are interesting. Debate is one thing but results and observations are better.

[oomm]

Originally Posted by Reed

If I am playing historical, I go 100/0/0/0 to minimize lopsided trades. Playing fictional I just leave it at default..

Have you used historical imports into fictional rosters?

[NoOne]

Originally Posted by oomm

You don't think 4/25 turns it off? If, with player eval settings one can impose order of emphasis from say most to least, then it follows that without player eval settings one could not impose order of emphasis. If the player eval settings are used in a way so as to not impose order of emphasis isn't that another way of saying one is neutralizing player eval settings, thus turning them 'off'...........?

I'm going to try 4/25 regardless but am not getting the logic others are insisting on.

I have chosen this emoticon to express my perplexity

I think no matter what you choose, it won't be a crazy difference between any suggestion above.

as far as how it works, simply think of how the inputs are derived... stats and ratings are different beasts. 1 is a product of the other, which potentially changes constantly over time.

ratings are, assuming scouting accuracy, more up to date than stats are, on average. plenty of exceptions. I don't think they go through too many scouting updates before the degradation shows at least in part, where as a player could get lucky with poor ratings for an entire year.. but it would have been a very bad bet with accurate information in hindsight.

stats always lag behind reality - they must accrue in a fairly common context to be worthwhile, too.... 991AB to clearly see BA... any one year can be drastically wrong indication of "contact"

stats eval is limited to the last 3 years, so .. kinda helps, but still same boat. more volatility from looking at small sample sizes, inherently. possibly creates more opportunities for playing time for younger players? depends on exact reasons why the real players get replaced by fictionally created players... could be distribution of ratings difference does something unique with player eval that wouldn't be seen in any fully fictional league.

[brotherblues]

Do you have to be in commish mode to change this? Or in preseason? It's always grayed out when I check.

[brotherblues]

And if it's a commish mode thing... is there a way to turn off challenge mode?

[progen]

Originally Posted by RchW

It seems clear that 50/50/0/0 would temper that overreaction which is why I use 40/45/10/5 now. I have used 45/45/10/0 and 45/40/10/5 too.

I run two fictional MLB clones that are 75 (above) and 100 seasons into play. The longer one is currently 45/40/15.

Here's the original post on this subject. It's a great read, and one I visit quite often.
https://forums.ootpdevelopments.com/...settings+stats

I'm currently playing a fictional universe, and have been using 5/55/25/15, because I am playing "stats only." I'm using HS and College feeder leagues, so for the draft, you do have players with multiple years of playing time to help with your decision making. I haven't seen anything too "wonky" to where I feel I need to tweak things. There are always some "head scratchers" with the AI's draft selections, but overall, nearly all of the first round selections, I could justify as legit. I haven't taken over a team yet, so just observing the first five seasons that I simmed. Some players I may have selected later in the first round then where the AI selected them, but to me, they were all first round selections.

*And the thing I really like with the AI is, yes, it can see the ratings/potential ratings, which plays a part into their draft choices, but when you look at say a pitcher and his WAR, ERA+, BABIP, and WHIP, which in one case were all great, despite his 3-4 record in HS, you could see why the AI chose that player in the first round.

I'm trying to get the AI to look more at stats, but as Sir Michael has mentioned in the thread I posted, the AI still looks at ratings regardless. In the original thread, MizzouRah was playing fictional, and used 25/50/20/5, or 20/50/20/5, although in the end, he uses 30/50/15/5 and said that was working fine for his fictional universe. Any suggestions on your own experience with fictional?

For the questionable draft choices, the AI can "see" the ratings/potential ratings, which is something I can't. I only have "other ratings" checked, so I can see speed/stealing/running/fielding and pitchers stamina and hold runners, but that's it.

I'm now in the process of taking over a team in dire straits(Tigers or Brewers), or expand the league, and take over an expansion team, so will be fun to see how things play out. Been playing this game since OOTP4, and still find this a fascinating subject.

[Syd Thrift]

I was doing 0/67/22/11 but was not enamored with the AI swapping out guys based on one semi bad season so I changed things to 0/34/33/33. Way too early to tell if the results are better…

[brotherblues]

Damn. This is the kind of thing I like to toy with and I'm stuck in challenge mode. Starting over would stink so bad.

[progen]

Originally Posted by Syd Thrift

I was doing 0/67/22/11 but was not enamored with the AI swapping out guys based on one semi bad season so I changed things to 0/34/33/33. Way too early to tell if the results are better…

Thanks for the update. Keep us posted on your results.