Trades and Transactions

[luckymann]

Dear Lord, how lame this whole process has become in the past few versions of the game. Same daftness season in, season out.

You make a perfectly decent offer to players when FA declare, they come back a while later saying they like the look of it, earning it the dreaded "favours offer" accreditation and then, like clockwork, when Winter Meetings open, no dice, he wants more $$. And so on and so forth.

Even absolute nobody players you offer a minors deal to will come back and start this absolutely bogus to and fro with you.

It is patently obvious that the game is being forced to follow a number of clearly defined and predetermined scripts. Player A follows this one; Player B, another; Player C, a third. No logic to any of it, just randomness and almost something like a quota system is being enforced.

Like I said, lame.

Surely this great game deserves better, especially 26 versions in, don't you think?

How about applying some simple logic to the process.

A club gets a certain score for their recent performance, another score for how good a fit it would be for the player in question. Just like IRL.

How about the terms being offered actually forming a decent chunk of the logic applied, rather than just being incidental to this going through the motions nonsense polluting the process at present, where as many times as not the player ends up signing elsewhere on far worse terms than those you were offering?

I'm the first to admit I have been a pain in the butt over a number of years about getting the AI logic up to speed. But now more than ever, with all the problems being raised regarding prospects being released illogically, can't the devs just admit this whole part of the game needs a complete overhaul and get cracking with one.

Yes, I have heard all the reasons regarding how hard it is etc etc but this is 2025, people. The advances made in this regard over the past 12 months alone make those excuses utter hogwash at this point.

How about for v27 we get a material upgrade in the AI (and, perhaps, in the stats interface as well, which is an embarrassment), rather than the cutesy nonsense of recent releases like the pointless Draft Combine, or any more dead-end upgrades to the iffy at best Dev Lab.

I personally don't need any more 3D advances or new bat colors, regardless of the sales-bait function this nonsense performs.

Pretty sure the time taken for a couple big changes done well would equal out to all the pre- and post-release work required to conceive, implement, and then endlessly fix these other features. And you'll be amazed how popular they are at the cash register.

OK, rant over, I love you guys but you're better than this rubbish.

G

PS Yes, I perhaps should have posted this to the general board but I try and stay away from there wherever possible.

[Pelican]

A lot to process, but I too have found that the script/banter from free agents does not seem to match their eventual decision-making. Including instances where my offer is rejected, but the FA listing still says "favors offer". Huh? That disconnect is disconcerting.

Realistically, all these free agent contracts are going to be negotiated with their agents, which should make the process more businesslike. The dialogue should reflect that. And, rather than wait to sift through offers, the agent should make demands for the player. Negotiation involves proposals each way. The game seems to envision a one-way process, controlled by the teams.

I also get tired of the "you're going to have to match the offer from Cedar Rapids" approach. Could be bluffing. Probably not. But what is the offer? Again, wouldn't an agent say "you've got to do better than $15 million over four years" or something like that?

The software does not seem capable of factoring in either an up-front signing bonus, or a bonus for ASG or Cy Young or MVP or targets in IP or GP. These are important factors that make one offer potentially much better than another. The game needs to weigh those factors.

[jdw]

Originally Posted by luckymann

Dear Lord, how lame this whole process has become in the past few versions of the game. Same daftness season in, season out.

And you aren't even touching on the AI GM Trading issues.

Everytime I return to OOTP, I tend to do Historical Career League runs that start in the early 1970s. The AI GM trades of Joe Morgan always are the breakers of Suspended Disbelief in running there. They happen in at least half the replays that I do, and almost always are worse than the Real Life Terrible Morgan Trade.

I thought I'd test just how bad of a deal the AI GM would make.

I'm the 1970 Red Sox. Import settings were 1+1+1 3 Year Calculations / Recalc, Real Stats, default trading / valuation settings, etc. Tried to be as vanilla as possible.

This is the degree to which the AI GM of the Astros could be fleeced:

Thursday, April 2nd , 1970

Traded 26-year old SS Rico Petrocelli, 32-year old 3B George Thomas, 22-year old LF Joe Lahoud, and 22-year old CF Billy Conigliaro to the Houston Astros, getting 26-year old 2B Joe Morgan, 28-year old RF Jimmy Wynn, 25-year old 3B Doug Rader, 27-year old RHP Mike Marshall, and 22-year old LF Cesar Geronimo in return.

I probably could have pushed harder. But I was laughing too hard when the Sox were ending up with:

1. the best player of the 1970s
2. a near-HOF CF who still had his career year to come in 1974
3. a starting 3B who would be useful through 1977 (8 years!)
4. a RP who would still be useful in 1979 (through the decade!)
5. a trade-bait/back up OF as a throw-in

For Rico having peaked the year before, and... a bunch of garbage.

As mentioned, just about every 1970-72 Joe Morgan trade the AI Astros or AI Reds make is terrible. This is just numbing on default.

Any of us who try to take this seriously and try to avoid building a 140+ win team have to set extreme rules on how we trade. Same if playing under the financials. But it's painful to feel like we need to restrict ourselves for realism, only to see the AI GMs do something stupid.

It's usually the big name trades that catch the eye, it's across the board. Pulling up another replay that I ran recently, these are AI-with-AI GM trades:

Saturday, June 13th , 1970
The New York Mets traded 22-year old SS Ted Martinez and 23-year old LHP Rich Folkers to the Pittsburgh Pirates, getting 30-year old RHP Dave Giusti in return.

Giusti's starting career is over, but he has 8 years of good RP work ahead of him from 1970-1977, with only 1976 being above league average in FIP+ right at 102. Which you could easily live with in a 5-man 1970s bullpen, relative to a lot of other teams' pens. Martinez was a garbage player, while Folkers was spotty and not as valuable as Giusti.

At the time of the trade, Giusti was 5-1 with 5 SV, a 2.55 ERA & 94 FIP+. He's 30 with 1971-77 to come.

Tuesday, July 28th , 1970
The San Francisco Giants traded 26-year old 2B Tito Fuentes and 24-year old 3B Al Gallagher to the Kansas City Royals, getting 27-year old LHP Jim Rooker in return.

Fuentes was at best a mediocre 2B, with some passable non-blackhole seasons. 1977 was good, but almost a dead cat bounce as he was toast the next year.

Gallagher was nothing.

Rooker was a under-100 FIP+ starter through 1976, 7 total years. His ERA+ are rough in 1971-72, and his innings down. But he bounced back for his career best run from 1973-76. His FIP+ fell off in 1977, but it's also one of those seasons where if you had a good defense, you could run him out in the back end of your rotation rather than a blackhole. The Pirates did, and it worked.

You're building a team, there is value in that 1970 + 1973-76, and you can use him in 1971-72 & 1977 depending what you have around him. It's a small deal, but Fuentes isn't really of value.

Friday, July 31st , 1970
The Baltimore Orioles traded 29-year old LF Curt Motton to the Houston Astros, getting 27-year old LHP Jack DiLauro and 24-year old LF Bob Watson in return.

Curt Morton is like a strange Rob Deer before Rob Deer. The import rated him at 42/81/93/100, largely driven by his inflated 1969 season. He was falling apart and unplayable by 1971. Perhaps ahead of his time where he would have been a useful RHB in a very strict platoon combo since he couldn't hit RHP if his life depended on it.

Bob Watson was quite a good hitter whose best seasons were impacted by the Astro Dome, and that he should have been a 1B/DH his whole career.

The 1970 O's? Look them up. They had Boog Powell at 1B, Frank Robinson + Paul Blair + the aforementioned Buford across the OF along with Merv Rettenmund as the roving 4th OF hitting .322/.394/.544. They don't have a spot at the moment for Watson.

OOTP does that all the time. Granted, the Astros have Mayberry coming up and OF as well. In theory someone needed to eventually move. But there are better places Watson could have gone, and for players who have actual careers.

For the AI O's, things *may* sort themselves out. They may make a trade. Buford & Rettenmund fall apart in 1972. Boog instantly falls from his 1970 MVP level downward in 1971, but 1+1+1 along with offense in general being down in 1971-72 may save him from Watson taking the job - Bogg isn't a terrible offensive player at that point.

Anyway, this is just one of those pointless AI GM trades that happens all the time where a good player is traded for garbage, and far too often ended up in a situation where someone gets blocked out of a starting job.

In turn, I can look at the Brewers OF, see a bunch of garbage, and feel bad that Watson didn't end up there where he could be a local hero for the next decade.

Anyway, it usually is the trade AI that sends me into a multiyear break from OOTP. It just a killer.

[luckymann]

I don't disagree with you at all, but we both need to remember that, in this specific context, we are armed with the hindsight that the AI is not. That fact doesn't excuse what is happening as you describe it, but it does mitigate things to a certain degree.

I have been pushing for a while now for some more guardrails for the historical mode of play, a series of options one can apply. Because what you've mentioned is but one of the many landmines to be negotiated between full historical transactions / lineups and the standard form of this mode of play.

The ability to set a player as a "franchise" or "cornerstone" player, thereby making them ineligible to be traded is one such suggestion I have made.

Perhaps the ability to add "weight" to certain players to ensure that, if they are traded, they are traded for the correct return. I'm not so sure about that one, as I feel it should be hardwired into the AI logic (and probably is as it stands, albeit not very well).

Historical minors is another realism buster, and I really feel this needs a comprehensive overhaul, starting with the ability to stipulate that players automatically come onto the same clubs they do in the default non-minors version of the game. Ty Cobb to the Tigers rather than as a FA, this sort of thing.

This is especially so given the incompatibility of historical minors and trans / lineups. There's too much differential between the two and no way to narrow it other than full on orchestration, which I know from experience is a massive onus to bear, beyond what most would be interested in doing.

While some have argued that this is how it happened IRL, I see that as (in almost every case) being beside the point. Virtually anyone who doesn't want rookies allocated in this fashion, would use the Rookie Draft option. If this change I'm suggesting were introduced, simply don't select it if you want them to appear randomly or use the Draft option and go that route.

I'd just like to see some options like these added that actually give the human player of an historical game more meaningful agency.

Not meaning to turn this into a full-on bash of the game, I honestly feel 26 is the best in ages. I just wish some of the core UE / UI stuff that has been ignored for too long finally got some love.

G

[Chubba2Chedda]

Can't AI see potential when making trades? Isn't that based on what a player really did in the future?

[luckymann]

Originally Posted by Chubba2Chedda

Can't AI see potential when making trades? Isn't that based on what a player really did in the future?

Yeah, in a way, but it doesn't get the exactitude the human player does. Potential is scouting-based, scouting is fallible, the AI settings only use this as part of their decision-making process. Whereas we know 1927 Babe Ruth, even a less precise version of him, will belt a bunch of homers, and our certainty is heightened with every setting in place, be it recalc, TCR, whatever.

Like I said, I'm not defending the AI's poor decision-making, just offering a mild mitigation for some of it.

G

[jdw]

Furthering luckymann's response:

The AI GM kinda-sorta sees "future", but also kinda-sorta doesn't. Best example I can give would be the same 1970 Import mentioned above.

At one extreme, the HOU AI GM can't properly calculate the future value of that Morgan+Wynn+Rader+Marshall+Geranimo fleecing mentioned above.

In the other extreme, it see the future value of Bobby Grich enough that it won't even bite at a self-fleece offer of Reggie Smith + Carlton Fisk + Carl Yastrzemski for him.

The only way it could "rate" Grich that high is based on some level of Future Analysis given Grich isn't highly rated for the 1970 season in which he barely played. But it values him so much that it won't take a combo package containing:

1. a Top 10 position player in the 1970s who is very good for the balance of the decade

2. a prospect of no less than equal value at an equally scarce position with a longer career ahead of him

3. a 1970 MVP level player who is an above average-to-good hitter the balance of the decade

So it's pretty obvious the engine see immense value in Grich, which on some level is a very good thing. You don't want the Human GM to fleece the O's for Grich.

But it also values him so extremely high that it doesn't let the Human GM to build what would be a reasonable package for him, which would be Fisk and at most some very nominal player that you often see the AI GM ask for in one of those "You're very close, just a little more" deals.

* * * * *

I will admit that it probably is a very complicated thing to program.

If one is using Remaining Peak as the driver for Potential Ratings, that having an increased significance on AI GM's rating players sounds like a good idea. That would address the Morgan issue. In a 1970 Import in the configuration, no position player in theory should have their Potential Ratings crunch a higher value. There simply is no position player in a 1970 Import that can match 1975 Joe Morgan.

But...

Bobby Murcer and Reggie Jackson in a same scenario in a 1970 Import would have a very close crunch to their Potential Ratings. We as a Human GM know how their careers turn out, and a straight up Bobby-Reggie deal in 1970 or 1971 or 1972 would end up terribly one sided.

We know in Real Life bad deal along those lines happen. It feels like "equal talent" is moving, but over time turns into a wipe out.

How to balance this out to make the GM less likely to make incredibly bad deals... I really don't know.

But as much as I point to stuff like the valuation of elite players (Morgan, Grich, Wynn, etc in 1970), we also need to keep in mind those bad valuations of players like Giusti & Rooker in the examples given above. Also the Watson one. If you're a Human GM who doesn't care, it's no big deal. If you care to a degree, it's painful to see how regularly it happens.

[luckymann]

Originally Posted by jdw

Furthering luckymann's response:

The AI GM kinda-sorta sees "future", but also kinda-sorta doesn't. Best example I can give would be the same 1970 Import mentioned above.

At one extreme, the HOU AI GM can't properly calculate the future value of that Morgan+Wynn+Rader+Marshall+Geranimo fleecing mentioned above.

In the other extreme, it see the future value of Bobby Grich enough that it won't even bite at a self-fleece offer of Reggie Smith + Carlton Fisk + Carl Yastrzemski for him.

The only way it could "rate" Grich that high is based on some level of Future Analysis given Grich isn't highly rated for the 1970 season in which he barely played. But it values him so much that it won't take a combo package containing:

1. a Top 10 position player in the 1970s who is very good for the balance of the decade

2. a prospect of no less than equal value at an equally scarce position with a longer career ahead of him

3. a 1970 MVP level player who is an above average-to-good hitter the balance of the decade

So it's pretty obvious the engine see immense value in Grich, which on some level is a very good thing. You don't want the Human GM to fleece the O's for Grich.

But it also values him so extremely high that it doesn't let the Human GM to build what would be a reasonable package for him, which would be Fisk and at most some very nominal player that you often see the AI GM ask for in one of those "You're very close, just a little more" deals.

* * * * *

I will admit that it probably is a very complicated thing to program.

If one is using Remaining Peak as the driver for Potential Ratings, that having an increased significance on AI GM's rating players sounds like a good idea. That would address the Morgan issue. In a 1970 Import in the configuration, no position player in theory should have their Potential Ratings crunch a higher value. There simply is no position player in a 1970 Import that can match 1975 Joe Morgan.

But...

Bobby Murcer and Reggie Jackson in a same scenario in a 1970 Import would have a very close crunch to their Potential Ratings. We as a Human GM know how their careers turn out, and a straight up Bobby-Reggie deal in 1970 or 1971 or 1972 would end up terribly one sided.

We know in Real Life bad deal along those lines happen. It feels like "equal talent" is moving, but over time turns into a wipe out.

How to balance this out to make the GM less likely to make incredibly bad deals... I really don't know.

But as much as I point to stuff like the valuation of elite players (Morgan, Grich, Wynn, etc in 1970), we also need to keep in mind those bad valuations of players like Giusti & Rooker in the examples given above. Also the Watson one. If you're a Human GM who doesn't care, it's no big deal. If you care to a degree, it's painful to see how regularly it happens.

Great explanation, jdw. It is also very much tied into the AI settings being applied and the value given to each of the four eval categories.

I concede the logic isn't easy to get right. All the same, my OP regarding an overhaul of both process and programming remains very much how I see it.

G

[jdw]

I think we're in agreement, and most of all that it is probably difficult to program without something getting even more broken.

I am loathed to adjust the AI Trade/Transactions Settings much from default, beyond lowering the AI Trade Frequency setting occasionally. I usually end up doing that after:

1. my first run through a Historical League running say the Dodgers
2. finding the AI trades an INSANE amount at the default frequency, include lots of Bad Deals
3. want it toned down in my next run through running say the Red Sox

I'm loathed to change much more in the settings because of things like this:

Mid-70s Red Sox
1B-OF-DH: Yaz, Smith, Lynn, Evans, Rice, Cooper, Oglivie, Manning

Even if I think it might be good to hang onto Manning as a quality back up OF (despite it ruining the realism of his career), that still is 7 other useful-to-good-to-excellent players to cover 5 positions. I need to move at least 2 of them, and would like to get reasonable value back for them. Specifically SP or 2B or 3B help, as the Sox were very light in developing those in the 1970s.

The Dodgers have the same thing. Excess SP while lacking in kicking up OF candidates. One really wants to covert excess into filling those holes.

As a Dodgers Fan of the 70s, I learned a lot on how they filled out their roster spots with a fair amount from their system but also a fair number of smart acquisitions, not a small number of them plugging holes for a good number of year. It actually was helpful in being able to watch the 90s Braves operate in a very similar fashion, though I don't recall many people making the analogy / tie to the styles.

Anyway, I fear that it I adjusted the AI Trade settings to kill off the Morgan Trades, it would kill off my ability to get reasonable value for Excess. Which would result in further Commissioner God actions where I have to fully determine what's Fair and force the trades. Also worry that it might further impact the AI where it's unable to fill their own holes in AI-AI Trades, causing:

1. talent to wastefully stack up to not be used such as a 70s Red Sox team keep all 8 players mentioned above with careers wasted / thrashed

2. in turn a team like the Sox never really dealing with their 2B & 3B issues until Boggs comes in 1982 and the Barrett/Reed coming up.

I just never know which way the dominoes fall if you push them one direction, or the other.

[jdw]

Just to add a note to my comments here:

I wouldn't want needed "good changes" to the Trade/Transaction AI to eliminate the potential for some quirkiness. I think we all enjoy that to a degree in a historical league. Quirkiness has it's place in baseball, in most any era.

An example this is in a recent replay that I ran, again starting with the 1970 Red Sox in a 1+1+1 Real Stats Import / Recalc.

On a lark, I swung a reasonable deal to bring Mark Belanger to the Sox to fill the SS spot with good defense until Burleson comes up in 1974, not really caring about the offensive side.

Belanger's Real Stats for a 1+1+1 import were:

1969: .287/.351/.345
1970: .218/.303/.259
1971: .266/.365/.320

They somehow generated 67/34/1/56 ratings

That's a Contact right in the same range as Reggie Smith's 68, a player who hit .309 & .303 & .283 in those years. Reggie does have a 62 vs 59 BABIP advantage, with a 68 vs 73 Avoid K deficit. We can go down the path of how OOTP's engine deals with the interation of Contact + BABIP + Avoid K along with the massive 73 vs 1 gap Reggie has on those pesky non-BABIP Power. Still... I was surprised that the Contact was remotely close in the Real Stats import of a .259 hitter and a .298 hitter in those three seasons.

But quirkiness is cool, right.

It gets better.

Playing half his games in Fenway, Belanger win the 1970 AL Batting Title with a .317 BA. He topped his former teammate Boog Powell by 2 points, preventing Boog from being the second O's player to win the Triple Crown in 5 seasons.

I've seen some quirky things in the hundreds of historical seasons I've played over the years, and mix in close to a hundred online fictional seasons. I'm not sure if I've ever seen something that quirky.

I'm not going to say I hate it, or that it totally breaks all notions of realism. Glenn Beckert hit .342 the Real following season, roughly 50 points above his second best BA. The .317 Belanger hit here is "only" 30 points above what he hit in the real 1969.

It jumped out because of Belanger's rep as a lightweight hitter which I lived through, and also that he hit .218 in the Real 1970.

I wouldn't advocate wanting these eliminated.

[luckymann]

As much as this is a fair bit off the path of my original post, I could't agree more. It is the overly predictable nature of historical saves that is their major drawback, so these outliers spice things up in a good way, no doubt.

G