AI to AI trade logic is awful

[tonnage]

I'm using 65 - 20 - 10 - 5 with the trading difficulty setting at the hardest and slightly preferring prospects. This is the setting I have found that makes the AI toughest to deal with. I haven't noticed anything different in 24, as trades are still hard to make so I'm sticking with these numbers, at least for now. I think the AI should be using the ratings weight provided by its scouts as the deciding factor on its player's talents, thus the reason I use 65. Of course, there are many other factors involved including age, injury prone, etc.

[jtsly2000]

Originally Posted by tonnage

I'm using 65 - 20 - 10 - 5 with the trading difficulty setting at the hardest and slightly preferring prospects. This is the setting I have found that makes the AI toughest to deal with. I haven't noticed anything different in 24, as trades are still hard to make so I'm sticking with these numbers, at least for now. I think the AI should be using the ratings weight provided by its scouts as the deciding factor on its player's talents, thus the reason I use 65. Of course, there are many other factors involved including age, injury prone, etc.

Small world, I live in Edmonton. Stony is a nice town, been there quite a few times over the years. I'm encourgaged by the data you guys are producing. I'm on the fence. I buy the game every year but hearing about the screen going blank, missing fielders and numerous crashes of the software have made me hesitant to buy the game even though I've bought it every year over the last two decades.

[tonnage]

Originally Posted by jtsly2000

Small world, I live in Edmonton. Stony is a nice town, been there quite a few times over the years. I'm encourgaged by the data you guys are producing. I'm on the fence. I buy the game every year but hearing about the screen going blank, missing fielders and numerous crashes of the software have made me hesitant to buy the game even though I've bought it every year over the last two decades.

A fellow Albertan! Don't let the bugs scare you off. I'm loving 24, even with its blemishes. The devs get them fixed up with each patch. Cheers!

[Matt Arnold]

Thanks for the data above. Getting the trade balance, both in talent matchups and number of deals, is always a tough task. We'll certainly keep watch and see if we need to make some tweaks here or there.

[PSUColonel]

Originally Posted by tonnage

I'm using 65 - 20 - 10 - 5 with the trading difficulty setting at the hardest and slightly preferring prospects. This is the setting I have found that makes the AI toughest to deal with. I haven't noticed anything different in 24, as trades are still hard to make so I'm sticking with these numbers, at least for now. I think the AI should be using the ratings weight provided by its scouts as the deciding factor on its player's talents, thus the reason I use 65. Of course, there are many other factors involved including age, injury prone, etc.

This is what I have used in the past. I think it works good for trades, and definitely will err on the side of caution when it comes to contract offers and/or extensions. I also don't usually ever see players' on waivers that I shouldn't.

Is it as realistic as some people like? Probably not..but it provides the most challenge. As I have said here many, many times....challenge and realism are not necessarily the same thing in OOTP.

I am trying 55/25/15/5 for awhile and I think it's probably just as good with a little more realism (stats) sprinkled in. I have always maintained that the ratings should be over 50%.

The new default of 30/50/15/5 has me re-thinking things however, but I just haven't really had enough time with version 24 to come to any kind of conclusion. I would like to be able to use 50/30/15/5, but what I am using is damn close. Will have to see.

55/20/15/10 could also be a possibility.

[polydamas]

It is really, really bad.

I was just headed here to see if anyone else was seeing this.


The most egregious trade I've seen so far was the literal #1 prospect AND a starting 3B for two unheralded (and garbage) prospects.

It definitely sticks out as the worst trade I've ever seen the AI make in all my years of OOTP.

[shakturi101]

Originally Posted by PSUColonel

This is what I have used in the past. I think it works good for trades, and definitely will err on the side of caution when it comes to contract offers and/or extensions. I also don't usually ever see players' on waivers that I shouldn't.

Is it as realistic as some people like? Probably not..but it provides the most challenge. As I have said here many, many times....challenge and realism are not necessarily the same thing in OOTP.

I am trying 55/25/15/5 for awhile and I think it's probably just as good with a little more realism (stats) sprinkled in. I have always maintained that the ratings should be over 50%.

The new default of 30/50/15/5 has me re-thinking things however, but I just haven't really had enough time with version 24 to come to any kind of conclusion. I would like to be able to use 50/30/15/5, but what I am using is damn close. Will have to see.

55/20/15/10 could also be a possibility.

What exactly makes the AI better when you set AI evaluation ratings higher than stats? Are they just bad, for some reason, at using stats to assess talent? Why would they be bad at that (in comparison to ratings, of course)?

Also, for people generally, who use a high ratings AI evaluation, what are using for your scouting accuracy? Wouldn't using a scouting accuracy less than normal with this tactic completely destroy your AI's ability to compete? You're kinda nerfing your AI hard if you make it use ratings heavily, but then make the accuracy so low.

[PSUColonel]

Originally Posted by shakturi101

What exactly makes the AI better when you set AI evaluation ratings higher than stats? Are they just bad, for some reason, at using stats to assess talent? Why would they be bad at that (in comparison to ratings, of course)?

Also, for people generally, who use a high ratings AI evaluation, what are using for your scouting accuracy? Wouldn't using a scouting accuracy less than normal with this tactic completely destroy your AI's ability to compete? You're kinda nerfing your AI hard if you make it use ratings heavily, but then make the accuracy so low.

The AI can "overreact" to stats at times. Stats fluctuate and are a snapshot in time...ratings are probably more indicative of a players actual worth.

[tonnage]

Originally Posted by shakturi101

Also, for people generally, who use a high ratings AI evaluation, what are using for your scouting accuracy?

I have my scouting accuracy set to "normal". I have never veered from that setting so I can't say how it would affect outside of that. I don't seem to have the bad AI trades others are talking about with the settings I use. Has there never been a bad AI trade? Of course not, I don't think that's possible, but they are far and few between.

[Karl Pagel Blues]

Originally Posted by PSUColonel

This is what I have used in the past. I think it works good for trades, and definitely will err on the side of caution when it comes to contract offers and/or extensions. I also don't usually ever see players' on waivers that I shouldn't.

Is it as realistic as some people like? Probably not..but it provides the most challenge. As I have said here many, many times....challenge and realism are not necessarily the same thing in OOTP.

I am trying 55/25/15/5 for awhile and I think it's probably just as good with a little more realism (stats) sprinkled in. I have always maintained that the ratings should be over 50%.

The new default of 30/50/15/5 has me re-thinking things however, but I just haven't really had enough time with version 24 to come to any kind of conclusion. I would like to be able to use 50/30/15/5, but what I am using is damn close. Will have to see.

55/20/15/10 could also be a possibility.

Do you vary those settings depending on the date? Shouldn't Current Year Stats be weighted at zero until maybe the first month of the season is completed?

And when exactly do Current Year Stats become Previous Year Stats? Once the offseason starts? Once the new season starts?

Maybe I missed it, but I was never able to find an answer to that.

[SirMichaelJordan]

Keep in mind that incorporated stats “yes” is on by default. So the ratings section is also using stats to a degree.

[Dr Naysay]

Not having a pop at anyone here per se and there are definite issues but since moving to a 90% stats only setup in last years iteration I have found out just how biased we can all be looking at ratings/potential ratings. One of my best players in last years game was a guy that I would never have stuck in the Majors had I not switched to stats. I have since seen a ton of whinging about trades on Reddit and YouTube about the AIs thinking but then looking at the trade I'm seeing a player with poor ratings whose overperfoeming those ratings at a 800+ ops throughout his Milb years which is likely why some AI trades happen. Again there are definitely issues but I think some people let their bias towards ratings seriously impact their attitude towards trading.

[Dr Naysay]

Originally Posted by Karl Pagel Blues

Do you vary those settings depending on the date? Shouldn't Current Year Stats be weighted at zero until maybe the first month of the season is completed?

And when exactly do Current Year Stats become Previous Year Stats? Once the offseason starts? Once the new season starts?

Maybe I missed it, but I was never able to find an answer to that.

Originally Posted by SirMichaelJordan

Keep in mind that incorporated stats “yes” is on by default. So the ratings section is also using stats to a degree.

These are two things I'd love to know, what impact does incorporating stats have on trades and shouldn't stats for the current season be at zero till there is a large enough sample size?

[Sweed]

Originally Posted by Karl Pagel Blues

Do you vary those settings depending on the date? Shouldn't Current Year Stats be weighted at zero until maybe the first month of the season is completed?

And when exactly do Current Year Stats become Previous Year Stats? Once the offseason starts? Once the new season starts?

Maybe I missed it, but I was never able to find an answer to that.

The evaluation for current year is weighted "under the hood" until there is a big enough sample size by the game. IDK the answers as to when they kick in at 100% for an on-going season, nor do I know for sure when current moves to previous. My guess is it is when the stats for current turn to zero and the standings show all teams with a 0-0 record. That makes the most sense to me as you literally can't see a player's stats from the last season completed without selecting "last year".

[Thundercrack]

Tigers traded Ty Cobb in his 2nd year while he was hitting .380 for a broken down 36 year old third baseman that I never heard of. Should. Not. Happen.

[Lukas Berger]

Originally Posted by Thundercrack

Tigers traded Ty Cobb in his 2nd year while he was hitting .380 for a broken down 36 year old third baseman that I never heard of. Should. Not. Happen.

Right, because the Red Sox never traded Babe Ruth in his prime for a bunch of washed quad A players.

Real teams make bad trades when trading with each other. The AI should also make bad trades when trading with itself.

The idea that the AI should somehow be far, far better than real, live human GM's, with all the countless terrible trades they've made, and never do anything questionable is just kind of hard for me to get a grip on.

I don't say anyone's wrong to think that, but just speaking personally, it bothers me to see this happen in real-life far more than it does in a game, which is just mimicking real-life, for both better and worse.

[MikeG719]

Originally Posted by Lukas Berger

Right, because the Red Sox never traded Babe Ruth in his prime for a bunch of washed quad A players.

Real teams make bad trades when trading with each other. The AI should also make bad trades when trading with itself.

The idea that the AI should somehow be far, far better than real, live human GM's, with all the countless terrible trades they've made, and never do anything questionable is just kind of hard for me to get a grip on.

I don't say anyone's wrong to think that, but just speaking personally, it bothers me to see this happen in real-life far more than it does in a game, which is just mimicking real-life, for both better and worse.

But why such a high frequency of trades?

[PepeSylvia]

Originally Posted by Lukas Berger

Right, because the Red Sox never traded Babe Ruth in his prime for a bunch of washed quad A players.

Real teams make bad trades when trading with each other. The AI should also make bad trades when trading with itself.

The idea that the AI should somehow be far, far better than real, live human GM's, with all the countless terrible trades they've made, and never do anything questionable is just kind of hard for me to get a grip on.

I don't say anyone's wrong to think that, but just speaking personally, it bothers me to see this happen in real-life far more than it does in a game, which is just mimicking real-life, for both better and worse.

There's absolutely no way you can truly believe the trading engine is fine in its current state if you've simmed more than a season with challenge mode trade settings. Of course you can point to a few crazy trades throughout real life history but there are a bunch of them happening every season in ootp 24. The Ruth trade argument is also disingenuous. It happened over a hundred years ago and the Boston owner was in financial trouble while Ruth wanted to renegotiate his contract.

Nobody is asking for AI perfection, that would be boring. I am asking for realism, though. Trades like Chris Archer for Glasnow/Baz/Meadows should be the exception, not the rule.

In the modern game there is no scenario where a team would trade Bobby Witt, a 5 WAR player in his first arb year, for a 17 year old prospect. Or Andrew Vaughn coming off a 43 HR season in his first arb year for a 30 year old backup shortstop.

Something changed with AI trade logic between 23 and 24. Teams are going haywire near the deadline and near winter meetings and it is breaking immersion.

[Lukas Berger]

Originally Posted by PepeSylvia

There's absolutely no way you can truly believe the trading engine is fine in its current state if you've simmed more than a season with challenge mode trade settings. Of course you can point to a few crazy trades throughout real life history but there are a bunch of them happening every season in ootp 24. The Ruth trade argument is also disingenuous. It happened over a hundred years ago and the Boston owner was in financial trouble while Ruth wanted to renegotiate his contract.

Nobody is asking for AI perfection, that would be boring. I am asking for realism, though. Trades like Chris Archer for Glasnow/Baz/Meadows should be the exception, not the rule.

In the modern game there is no scenario where a team would trade Bobby Witt, a 5 WAR player in his first arb year, for a 17 year old prospect. Or Andrew Vaughn coming off a 43 HR season in his first arb year for a 30 year old backup shortstop.

Something changed with AI trade logic between 23 and 24. Teams are going haywire near the deadline and near winter meetings and it is breaking immersion.

Teams making too many trades and teams making bad trades are kind of different issues, though maybe tangentially related.

I completely agree that on the default settings teams are making too many trades. I know we've already said in a couple spots that we're looking into the amount of trades made and will rebalance things there.

I suppose part of the thing about bad trades is that if there are more trades overall, then there will also be more bad trades. Just stands to reason.

So balancing that out a bit more will likely help bring down the amount of bad trades as well, and allow a more accurate assessment of whether there are too many of those, or just too many trades in general.

[Sweed]

Thanks to those that brought up the issue. Big thanks to billyray1984 for taking the time to build, test, and provide data. Anytime someone takes the time to do this and give context is of great help to all of us.

Thanks to Matt and Lukas for looking into the problem.

For now I've put my league trading frequency to "low", and will have to see if that workaround is effective until the developers figure out the fix.