Getting the AI to play as shrewdly as a human?

[jmknpk2]

I would like to set up a solo fictional league where it seems as if I am playing against humans rather than a weak AI that I can take advantage of.

I recently set up a league where I have trading set on "hard", "favor prospects", and I chose a team with a market size of 1. This gives me a significant challenge, and for the most part, the trading seems to work to my liking. However, there are a couple ways that I can still take advantage of the AI that I don't like.

1) I can sign free agent minor leaguers to contracts with almost no competition. I usually can find one or two young players in the free agent pool with potential ratings above 50 out of 100. I offer these players minor league contracts and they always sign with me. If I was playing with human managers, surely someone else would at least make an offer, such that I occasionally were not able to sign these players.

2) I can trade 10 minor league players who would never have a chance in the majors for one half-way decent player who will almost certainly have a chance in the majors. Or I can trade a decent player plus nine players who will never make it to the majors for one good player.

So, with these two weaknesses, I find that I can take advantage of the AI. I can sign a bunch of these free agent minor leaguers. Then I can trade dozens of them for a few decent players. In essence, I can create a few good players from nothing. This is something I would never be able to do in a league against real human players.

I know that I could set some gentlemen's rules for myself.
a) I could make a rule to never trade more than three players for one, that would solve problem 1. I don't think I want to set trading to very-hard because overall, I think the trading difficulty works to my satisfaction.
b) But I'm not sure how to handle the signing of free agent minor leaguers problem. Would I set a rule that I only sign one minor league free agent per season? Do I roll some dice on the side and only allow myself to sign that good minor league free agent if I roll a six?

I personally struggle, because the competitive side of me wants to exploit every possible angle in building a successful team (like I would against humans). I would rather not create gentlemen's rules that I impose on myself. Because that is kind of like playing a game against a child where I let the child win. I lose the thrill of being competitive. I would rather be restricted by a clever AI.

Anyway, ultimately I would prefer that I find settings that can give me an experience close to what I would find playing against humans. But I'm not sure that's possible now.

I'm curious whether others have found the same two weaknesses above, and how they might deal with them.

[injury log]

The trading thing is messed up. I think things were changed in this version to solve a problem in OOTP8, but it's introduced a new problem. Last version, you could easily stack a trade with the AI, getting ten players in exchange for one, because the AI assigned nil value to anyone below a certain threshold. So you could set up what would be a fair one-for-one exchange, then add on nine semi-good prospects. One talent boost later, and you'd find you'd traded for a superstar. It seems to solve this, the new version assigns value to *every* player (at least if they're not overpaid vets), regardless of how rubbish they are. So you can trade ten 38 year old relievers who couldn't hold their own in Rookie ball and get something decent in exchange. You can get around this issue by imposing a house rule: if the AI is dealing you X players, you can force yourself never to trade the AI more than X+2 players, or X+3, or whatever you like. If the AI needs to deal you nearly as many guys as you deal them, the value of excess players will mostly cancel out.

As for the high quality minor league free agents, I'm guessing you have hidden players turned on. With hidden players on, I constantly find quality prospects in the FA pool, and these guys have no past history. My assumption is that this is how AI teams sign hidden guys - the game decides that an AI team finds a hidden player, then puts the hidden player in the FA pool. I don't know why the game doesn't just assign the player to the AI team's roster. So, if you're signing the hidden players you find, and you're also signing the ones the AI finds who end up in the FA pool, you're getting a huge advantage. I simply don't sign any FAs who don't have any prior history, unless I know they were generated for the draft pool.

[jmknpk2]

Thanks injury_log, that's a good suggestion about the free agents without history.

[jmknpk2]

Originally Posted by injury log

I think things were changed in this version to solve a problem in OOTP8, but it's introduced a new problem.

Ah, so the scales have tipped in the other direction. I can see where this is a tricky thing to program.

[Isura]

I also set trading to hard/low/favor prospects, and haven't seen this issue. My general policy is to only make about 2-3 ML level trades a year, and no minor league trades unless it's offered by the AI (or they ask for one in 'make it work'). When using make this work the AI is generally very specific wanting my 4-5 star players. Shop a player is even worse, I rarely a good trade from that method. If there's a player I want I'll attempt to 'make it work' and offer them whatever they ask if it makes sense for both teams. Also I don't allow draft pick or recently traded player trading in league rules.

Regarding 1), I never bother to even look at the FA pool until the ML free agents are available. I use ghost players and prefer to build my up own farm system through the draft.

Also, what is your player evaluation settings? I use 60/20/15/5 which seems to pretty well.

[Isura]

Originally Posted by injury log

As for the high quality minor league free agents, I'm guessing you have hidden players turned on. With hidden players on, I constantly find quality prospects in the FA pool, and these guys have no past history. My assumption is that this is how AI teams sign hidden guys - the game decides that an AI team finds a hidden player, then puts the hidden player in the FA pool. I don't know why the game doesn't just assign the player to the AI team's roster. So, if you're signing the hidden players you find, and you're also signing the ones the AI finds who end up in the FA pool, you're getting a huge advantage. I simply don't sign any FAs who don't have any prior history, unless I know they were generated for the draft pool.

Interesting theory. If this is true it's a good house rule to avoid signing foreign FAs with high potential.

[jmknpk2]

Originally Posted by Isura

Also, what is your player evaluation settings? I use 60/20/15/5 which seems to pretty well.

I use the default of 30/50/15/5. I kind of use that same evalution scale when I evaluate players.

No draft pick trading. I don't trade recent free agents (what I described above was trading last year's free agents).

I do trade recently acquired (by trade) players, sometimes I need to do this to fill holes and remain within mine and other teams available budget. For example I'll trade for a good catcher with low salary from one team, who I know another team would want and then get a shortstop from that team for the catcher.

[Isura]

That could be the issue. Increase the weighing of ratings. That tends to make the AI smarter

[The Wolf]

Try 5/55/30/10 and turn ratings off so you can't see them. Play just by stats.

[jmknpk2]

Originally Posted by The Wolf

Try 5/55/30/10 and turn ratings off so you can't see them. Play just by stats.

That would certainly make things more challenging. Almost seems too challenging, I guess I should give it a try though. Wouldn't think I'd get bored of winning to many championships in a row.

[pressure]

Playing by just stats is the only way that I've been able to find a good balance. I find it to be more immersive too.

[The Wolf]

If you play with stats only you will find OOTP to be a much harder and much more immersive game. You really have to manage by the seat of your pants.

[The Wolf]

You also need to dial Talent Change Randomness down to around 10 in Game Setup. It cuts out the wild, random talent fluctuations. That way you can watch your players pretty much steadily marching somewhere as they develop, even if you don't know where that destination is.

[markprior22]

Quick question...I recently switched my solo league over to "stats only." In the first amateur draft, I turned the ratings back on for draft purposes (and didn't look at any of the existing players in the league). After the draft, I turned the ratings off again. Just curious as to how you handle this. Seems like, with ratings off, the draft would be a complete crapshoot.

[Isura]

You could let the scout pick. Or just turn on overall potential. Add feeder leagues so there's stats for incoming players.

[pressure]

Yeah drafts are a bit difficult, I try to have as many players coming from feeder leagues and use those stats/what myscout says. But it might be more fun to just turn potential ratings back on for that day (just make sure not to "accidentally" look at your lineup with ratings back on) ; )

[bababui]

Originally Posted by The Wolf

If you play with stats only you will find OOTP to be a much harder and much more immersive game. You really have to manage by the seat of your pants.

This might be fun but its impossible when playing out all your teams games. How would you know who to put on the initial 25 man roster?

[The Wolf]

Originally Posted by bababui

This might be fun but its impossible when playing out all your teams games. How would you know who to put on the initial 25 man roster?

I play every game out and play with just stats, so don't tell me it's impossible. I know better.

There's a lot of ways to do these things. For the draft, you can use your scouts and pick the chioces they recommend. As an alternative, you can turn stars (only! no ratings) on just for the draft.

For your starter team the simplest way is to let the AI handle the initial setup. (The rationale for this is that the team the AI sets up is the team that you inherited.) As an alternative, turn stars (only! no ratings) on just long enough to set your initial roster. Or you can let the AI play a season or two out and then take over, with some accumulated stats you can use to figure out who's who and what's what. All of these methods work.

Playing the game this way is harder, more fun, and more immersive than vanilla OOTP. No, I didn't invent it, I just read about it here on the board and tried it for myself. You won't have a dynasty, you will be scrabbling to figure out who to play and who to promote and demote, you will have epic successes and epic failures, and you will be immersed in the game - because now every single piece of data you can get matters because you don't have any ratings to go by.

Oh, and it makes trades nail-biting, agonizing fun, too.

Want to really feel like a GM? Play it this way.

[injury log]

Originally Posted by markprior22

Quick question...I recently switched my solo league over to "stats only." In the first amateur draft, I turned the ratings back on for draft purposes (and didn't look at any of the existing players in the league). After the draft, I turned the ratings off again. Just curious as to how you handle this. Seems like, with ratings off, the draft would be a complete crapshoot.

Actually, no, perhaps surprisingly- or at least this was true in OOTP8 (doubt it's changed) -- even with ratings invisible, you can still sort any column you want. So if you want to know which guy has the best Stuff Potential in the draft pool, click on that column, and you'll find out. You just won't have any idea how good his Stuff is, but you will know it's better than anyone else's. Fun way to draft.

[The Wolf]

And there's yet another way.