AI Line-up Issues/Realism

[Syd Thrift]

Originally Posted by uruguru

I am 100% on board with the high innate variability of batting statistics in the real world, so in that regard I try to replicate that with as much smoothing as possible (5-year averages) to get as close to the "true" performance of the player and then let OOTP do the RNG for the ups and downs.

The AI issue for me is that ratings don't exist in the real world.. there is just performance. So when a Tony Armas has an extended drop in performance, the team can't tell if its due to just random variability or reflects a physical performance degradation. At some point they pull the trigger and either a) move him down in the lineup, b) bench him, or c) demote him to the minors, or maybe even d) trade or release him.

When OOTP evaluates players by ratings (like the default setup), then it doesn't realistically react to performance drops and just keeps plugging their .160 hitter in the #3 slot like nothing is out of the ordinary. That's why I don't use ratings for historical sims, but something like 0/60/30/10.

Ratings are sort of a cheat code, imo, and should be hidden or obfuscated.

I'm in around 90% agreement with this. To me - and I've put this forward as a way of (re)doing scouting but I can understand how people wouldn't like it - scouting ought to be a collection of snapshots a scout has of a player that, taken in aggregate, give them a rating. Any one "snapshot" really ought to be wildly inaccurate, on the level of grading a guy with 60 speed anywhere from, say, 30 to 80. They'd tend to converge on the actual ratings but only over time. So... a college kid who's only been scouted a few times might have a lot of "error bars" around their actual ability, a high school kid might have only been looked at once or twice by a scout and could be all over the place, and so on.

Your veterans, then, would be scouted super well - of course, by the time a guy has lots of stats built up, you don't even really need the ratings per se. That said, if a 26 year old suddenly starts hitting for power (in OOTP terms, getting a TCR boost), a scout looking at him might have 5 years of accumulated data saying that he's got 40 grade power. If that scout sees 60 grade, he'll take it into account... but he's not going to suddenly say he's got 60 power. What if he just caught the guy on a good day/week/month? What's likely to happen is the player in question will start hitting more HRs *before* the scout changes his grade, not after.

And more importantly at least to me, if a vet with 60 Contact suddenly can't hit the fastball and falls to a 30, you should see the results on the field in most cases before your scout can definitively tell you he's toast. I hate to say this because he's one of the nicest guys in the sport but take Eugenio Suarez for example: the raw numbers indicates that his bat speed is down, his whiff rate is up, and he just plain seems due for a cliff year. The Mariners still started him all last year and the Diamondbacks look poised to do the same in 2024 (and watch, now that I've said this, he'll prove me wrong).

FWIW too, I should point out that in addition to turning ratings off I also smoothed out aging quite a bit because look: a 38 year old can go into a slump the same as a 28 year old can. However, a 38 year old is just plain not going to be given all the same opportunities to break out of it that the 28 year old is unless he's like Albert Pujols or something. There's just plain a greater chance that the dip in ability from an older player is permanent and at the same time the chances of positive development are far, far lower, so all things being equal (which they are not) you're waaaay more likely to play the 28 year old coming off a bad year than the 38 year old.

I also made development happen more quickly so that the league isn't dominated by 30+ year olds. And then on top of that I have TCR set to 150 because IMO random stuff just plain happens in baseball a loooot, whether it's Kolten Wong mysteriously turning into a pumpkin last year or, like, Brian Downing becoming a power hitter with a great eye after he turned 30. I do make all the major transactions in the league which is waaaaay too much work for most people but i have to say, the league sure doesn't "feel" like it's overloaded by vets or anything. Top draft picks I think tend to move up a bit more quickly but I'm not actually sure that young *players* move up more quickly than in real life, if that makes sense.

[uruguru]

Originally Posted by Syd Thrift

And then on top of that I have TCR set to 150 because IMO random stuff just plain happens in baseball a loooot, whether it's Kolten Wong mysteriously turning into a pumpkin last year or, like, Brian Downing becoming a power hitter with a great eye after he turned 30.

Every skill improvement or degradation in the real world is a talent change, though.

Nolan Ryan was literally a few months from washing out of the majors until his new pitching coach at the Angels straightened him out. If that Fregosi trade had never happened, "Nolan Ryan" would just be another of the hundreds of forgotten MLB promising teenagers that never developed. But because of that completely unpredictable change in coaching and the talent change it caused, the OOTP version of Nolan Ryan is a 20/80 teenager that every AI GM lusts after in the 1966 draft like he's some kind of can't miss Hall of Famer.

The movie Moneyball had it right about the real world. There are no potentials. The scouts don't really know how anyone is going to develop. There are just actuals of how good the player is now, how good his coaches are, and how hard he works to improve himself (and avoid injury of course).

If there was a way to completely hide potentials from the player and AI and have a player's development be not based on tracking towards some predetermined value but instead on the quality of the player and his coaches, that woud be great for a fictional game (obvsly not a historical sim)

The value of scouting would then lie in measuring how good the player is now (radar gun, bat speed, etc) and what his work ethic is like (big guess there). GMs then draft based on that and toss them to the coaches and sacrifice their chickens to the baseball gods.

[Brad K]

Originally Posted by uruguru

Nolan Ryan was literally a few months from washing out of the majors until his new pitching coach at the Angels straightened him out. If that Fregosi trade had never happened, "Nolan Ryan" would just be another of the hundreds of forgotten MLB promising teenagers that never developed. But because of that completely unpredictable change in coaching and the talent change it caused, the OOTP version of Nolan Ryan is a 20/80 teenager that every AI GM lusts after in the 1966 draft like he's some kind of can't miss Hall of Famer.

The movie Moneyball had it right about the real world. There are no potentials. The scouts don't really know how anyone is going to develop. There are just actuals of how good the player is now, how good his coaches are, and how hard he works to improve himself (and avoid injury of course).

A similar thought on potential. Bobby Bonilla was in the Pirates minor league organization. After the 1985 season he was not protected by the Pirates from the Rule 5 draft and was selected by the White Sox. The Pirates traded and got him back during the 1986 season after the Sox put him in the majors. An interesting question with the answer lost in history is who did the Pirates put on their 40 man roster instead of Bonilla which left Bonilla exposed to the draft.

In the trade the Pirates gave up SP Jose DeLeon who in four years with the Pirates had a 3.7 WAR. So while that shows value greater than someone being allowed to be Rule 5 eligible it still far under rates what Bonilla ended up doing.

[Humboldt KA]

Originally Posted by Matt Arnold

The issue with fatigue subs is that, generally speaking, the AI tries not to be too aggressive at shifting around your lineup. When your cleanup hitter is sitting out, it will naturally take the lineup with the replacement swapped into that spot. Now, usually if they are a worse hitter like a Counsell or Sanders, it should swap them down in the lineup. But if there's something preventing that (do you have a force lineup spot on any of those other guys lower down?), it might not find a suitable swap. When it subs in for defense it doesn't try to rebuild your lineup from scratch necessarily, since most teams tend to stick with their normal lineup but just take a few small tweaks, when someone sits.

This must be dependent on manager or strategy settings. My MLB manager moves the clean-up hitter to 3rd and puts the replacement 8th--always as far as I've noticed..