Using only potential ratings

[PSUColonel]

If you choose to display only potential ratings, without current ones, I do think this can create a bit more realism, in terms of how one may "view" players. (perception) This is because the scouting model in OOTP is nothing like reality unfortunately. My question however....is there a way to get the AI to do the same, in terms of the way it evaluates players?...or will the AI simply always see the underlying current ratings?

[NoOne]

Originally Posted by PSUColonel

If you choose to display only potential ratings, without current ones, I do think this can create a bit more realism, in terms of how one may "view" players. (perception) This is because the scouting model in OOTP is nothing like reality unfortunately. My question however....is there a way to get the AI to do the same, in terms of the way it evaluates players?...or will the AI simply always see the underlying current ratings?

you know that's one of the main reasons i've avoided all stats... or more stats less ratings options... clearly i haven't looked into it enough as to what you can do.

i'm pretty sure the AI sees things as you do... within reason. could it have some exceptions where it it necessary for the game to work well? sure, but i'm confident if that does occur, it's an exception and not the rule. small potatoes (poh- tah - toes).

so, if you turn it on, that's the data teh ai GM's and coaches use to make their various decisions related to players ratings.

anyway, i think you are 100% right, that this is a great way to do it... scouts see potential, and current ability is never known - which it isn't without a suitable track record and even then it is a best guess -- rarely is teh sample and other relevant factors lineup perfectly and only for a short window (Aging and other effects). now you only have to tune one thing in the settings -- how easy it it to see "real" potential while you draft and sign newly created playes (like new intl ama, intl ama fa types in addition to the draft).

you make that with the right amount of randomness, and it's not unlike reality. (or at leass less unlike)... double neg a la douglas adams "...almost, but not quite, entirely unlike tea."

------------------

extra crap i wrote because i am waiting to go make a beer run and head to watch the game at a friend's little gathering... i don't even like watching the super bowl anymore, lol.

this mostly pertains to scouting of amatuers and 'younger talents':

if they made scouts like they were in the real world... you'd(plural) realize how useless scouting info is for amatuers and minor leaguers, yet the only option available. most hot-shot prospects never live up to their billing. some of that is merely public relations (e.g. teams won't say bad things about their prospects publicly), but as you can tell from various trades that happen each year that they really have no clue without a track record with a known* league baseline (like playing years in the mlb - * for simplicity, obviously it's not 100% known).

they are 50/50 at best for the top 20 or so playes drafted or somethign similar to that and the probabilty of hitting on a pick drops precipitously after that.

(likely worse than 50/50, but if i guess i take a conservative guess skewed in a way that does not support my thoughts - worst case scenario type thing) why likely? the nfl draft is more known than mlb draft, and the top gm's hit on 50% of their first round draft picks... or at least that's standard at which you can call yourself decent at your job.... so, in the mlb it's got to be lower than 50% + some of the interesting threads dealing with relevant hard data (WAR by round game research comes to mind).

i tihnk most would be frustrated by reality, even though many, like myself, clamour for it with scouting.

statistical track record is infinitely more useful than what a scout can tell you watchign players play a few games here and there... the problem is you need a long time for that track record to become viable information... in the meantime the scout is the only option.

even mil stats are poorly correlated to mlb success. it's not the same game - different environement... some leagues you only see fastballs, some you have guys working on secondary pitches, others are skewed in ways i can't come up with off the top of my head, but i know it has everythign to do with developing a player and less to do with purely winning games, as it is in the MLB. stats in mil are very restricted in what they offer.

[PSUColonel]

I'll tell you what I have been playing around with this....and I would submit, that even if someone wanted to use potential ratings, that zeroing out ratings in the AI evaluation seems to have a profound effect on the AI's behavior. I think it DOES allow for a much more realistic acting AI....this may be why stats only players are so hung ho.

I also believe you could use both current & potential ratings on this setting, since the AI values ratings on some level anyway.


gotta run...more later

[NoOne]

Originally Posted by PSUColonel

I'll tell you what I have been playing around with this....and I would submit, that even if someone wanted to use potential ratings, that zeroing out ratings in the AI evaluation seems to have a profound effect on the AI's behavior. I think it DOES allow for a much more realistic acting AI....this may be why stats only players are so hung ho.

I also believe you could use both current & potential ratings on this setting, since the AI values ratings on some level anyway.


gotta run...more later

i'd bet from the ai's point of view it is basically similar to 100% stats ai evaluation when you turn off ratings. it basically calculates them off of stats as you are expected to do... and conveniently there's a bit of code already made for exactly that.

i forgot about that part... you and i have differeing opinions on the ai behaviour results being positive when 100% stats.... if i'm right about the previous paragraph, then i may not like it so much. still worth the look

[PSUColonel]

I can tell it is built into the code...when you zero out ratings...not one manager, GM, or coach will ever value them....not the same for any of the stats....so now I am not sure

[PSUColonel]

I am guessing if you were to use "potentials" only....you would need to go Nearly all stats. Obviously with potentials, there is still an idea of how a player COULD perform...and you would think that would need to play into the formula on some level.

My question is this: how much? A loaded question for sure, but not an easy one to figure out either. I am currently using 50/30/20/10...but that is with both current & potentials activated, on normal scouting accuracy. (BTW changing scouting accuracy is not a good idea IMO...but that is a discussion for another times..unless someone wants to bring it up)

Anyway, You would think something would ned to go into ratings....but maybe not as much as if you are using current. How much is perception/potential worth? Any ideas?