Quote:
Originally Posted by andyhdz
Yes I'm seeing this too. AI tends to pull the trigger on these vets who are slipping a bit a lot faster and just puts them on waivers. Yes it does happen in real life but not use to seeing it in ootp
|
It's a tricky balance. On one hand, the AI using ratings more means they'll be more competitive against humans. They'll ditch the vets who are declining and replace them with higher rated players. But it's also not realistic to have teams just waiving a solid veteran because they had a ratings drop.
Hopefully this is taken constructively, but here's a few things I think might improve the AI.
1) Regression feels too linear. Once a player starts declining, they're just going to keep declining. I feel like that's not always the case in baseball as players age. Sure over a long period of time, they will regress, but there are more fluctuations from season to season. Take a look at a player like Yu Darvish. His season last year would never happen in OOTP. The decline would appear to start in 2021 and he'd probably be out of the league by 2023. But in real life, he found something and had a resurgent season. It's not super common, but I think more fluctuations in ratings on a year by year basis would help make veterans more valuable. So maybe a player starts to decline at age 30 for a couple years, has a resurgent season at 33 or 34, then starts that slow decline again.
2) Regression of ratings happen too fast during the season. You can start a season with a veteran player who is rated as a mediocre player at the start. By June, their ratings have completely tanked and they'd struggle to survive in A ball. Players do fall off, but I think it happens too rapidly too often. For the most part, players slowly regress over time and it's rarely that dramatic over the span of a couple months. I don't know how the game handles ratings internally, but maybe a wider range (0-10000 lets say) would allow for more gradual regression. On the human side, we wouldn't see that wide range.
3) I don't think the AI puts enough emphasis on team control. It has improved a great deal when it comes to prospects in the minors. But once a player is in his 3rd year, it doesn't seem to value those next few years of affordable team control as much as it should. It's not something major that needs to be changed, just some minor tweaking to make sure those good young players have a little more perceived value.
4) Prospect for prospect trades. I'd like to see teams who have an abundance of a position in the minors trade for a need they have in the minors with another team. Say team A is loaded with pitching prospects but light on infielders. Team B is the opposite. It would make sense for them to make a trade for say 3 star prospects to level that up a bit. Both teams get more balanced and that should make them more competitive against humans.
5) AI teams don't seem to factor in future regression of a veteran into their dealmaking as much as they should. I feel it's far too easy to dump a decent veteran as soon as they've started to decline but have years left on their deal. Only contenders should be looking to make those trades and they should be more wary of anticipated production in the later years of the contract.
I don't think the AI is too far off. Most of these things feel like minor tweaks (outside of a more dynamic regression). Hopefully this doesn't come across negative and more constructive. Just what I've experienced in the game.