Fair Trade - Page 2

waittilnextyear · 08-03-2018, 06:24 PM

I think it could be well nigh impossible to come up with a quantitative, predictive model for whether a trade should be deemed "fair." For a number of reasons mostly relating to context and league settings as well as the sensibilities of different users. Imagine placing every possible variable (age, salary, team control, performance this year, last year, team goals, positional depth...) into a very large matrix--this would probably be the only truly quantitative way. It would seem horrendous to try that for each deal you're considering.

Also, you can't really judge a trade until some time passes, so something that checks all the boxes up front might end up going sour later. And, of course, if we're going for strict realism and using baseball history as a guide, there should be some non-zero fraction of trades that are not really fair at all. Both in the moment and after the fact.

However, it's not altogether a bad idea to try to come up with some sort of "code" to live by in terms of making your own trades. Because we all surely hate the feeling from taking advantage and the internal pressure to...gasp...restart. I think people finding their own criteria and sense of "feel" is as good as it gets. One idea might be, like we've done on Operation Sports for the MLB The Show series, to start a trade thread where people can propose things and get feedback--getting 2 or 3 people saying "yeah, that seems legit" can make people feel justified in making a particular trade and less likely to feel "dirty" later on.

Some things I do in my own OOTP save to keep myself honest...

(1) I've begun giving myself a maximum of 5 trades per calendar year. This makes me really have to prioritize (I can't salary dump everyone like I otherwise might). Ultimately, if I've used my last trade on July 8th, then I have to pass on every subsequent offer from the AI no matter how much it could help me. Quantity or trade volume is something that is easy for me to control without going above and beyond to devise an elaborate system. As a peripheral benefit, it also helps me hang onto some of my players so they have longer careers with my franchise (which was another issue for me and roster churn).

(2) The settings are helpful. I use "hard" and "favor prospects." I think this combo does a pretty decent job of putting upper bounds on how much advantage I can squeeze out of a deal and how much salary I can dump. Of course, I want to get some advantage from the trade otherwise there's no incentive for me to make it. The question is whether I can live with the magnitude of that advantage. Or not. I would suggest people play with not only the easy/medium/hard and favor prospects/veterans, but to adjust AI evaluation settings/percentages as well. Use a test franchise and season to taste--keep doing what you feel is an unfair deal and change the settings to where you have the toughest time doing it.

(3) I really try to look at the deal from the opposing GM's viewpoint and ensure that there is some logical motive for them. As I balance that with trying to rip them off, lol. It doesn't mean they're getting the better of it necessarily, but it means that I can't have everything. If I am getting the best player, and the cheapest player, and the most players, and on and on, then that feels like an exploit. But if I get the best player and that player makes good money, and his team is over budget, and operated at a loss last year, then I can start to see some logic there. Or, if I send a decent starter to a contending team that has 3 starters on the DL for the forseeable future. Or, if I am taking a veteran and some salary from a bottom-feeder. It helps that there is some reasonable facsimile of a real-life trade.

To the OP, though, you may not really get a truly quantitative, by-the-numbers modus operandi out of this, but maybe people can reflect on what they find to be fair/unfair generally and this thread becomes helpful in that way.

RchW · 08-03-2018, 06:47 PM

Well said. This I particularly liked.

Quote:

Also, you can't really judge a trade until some time passes, so something that checks all the boxes up front might end up going sour later. And, of course, if we're going for strict realism and using baseball history as a guide, there should be some non-zero fraction of trades that are not really fair at all. Both in the moment and after the fact.

Orcin · 08-03-2018, 06:49 PM

Quote:

Originally Posted by waittilnextyear

To the OP, though, you may not really get a truly quantitative, by-the-numbers modus operandi out of this, but maybe people can reflect on what they find to be fair/unfair generally and this thread becomes helpful in that way.

This was a great answer and really helpful. I particularly like the idea of a trade volume limitation. While arbitrary, it does indeed force you to prioritize and not just take every deal that you can find which favors you. Your description of how to analyze the deal from the opposing team's viewpoint also gives me (and hopefully others) some good ideas.

NoOne · 08-03-2018, 11:48 PM

the teams i trade with usually do well afterward or at least better in part due to the players received.

i'm trading talented mlb players by 29-32 more times than not. the way i do contracts typically won't cause a long-term burden to the destination team. if they aren't better off it is invariably because the ai did something dumb or catastrophe occured - tcr or age or injuries, et al. they get 1-5 years of good talent/cost ratio.

Timofmars · 08-04-2018, 03:52 AM

I don't like to handicap myself beyond what the game offers in settings and situations. I'll play for a low budget team with very hard trade difficulty, but I don't like to have to decide cancel trades because the AI is being stupid. That makes it feel less like a game and more like a simulation model. I want a game, not a model.

In fact, I like to mess with the game world by giving the richest teams the worst deals they will accept, and saving my best quality unwanted players for the weakest low budget teams. I like seeing the standings be in reverse from what they usually should be, and I feel happy for the Marlins being at the top.

Most of the bad trades are either a result of the AI seeing value in useless players for some reason, or in their disregard for cash and salary retention. So perhaps you could avoid most bad trades by refusing to add low overall rated players in order to complete trades (possible exception being where the player is an improvement to the team), and not taking cash or asking the AI to retain salaries.

Another area the AI may make strange decisions is in how they value players they don't need. They seem to be happy to take a talented player even if they already have a better player in that position. Although a team that is weak in that position would probably put much more value on the player and give up more to get him. Anyway, I don't think this is really a problem, since a talented player is valuable to someone, so they could trade him for something they actually need. But maybe setting the AI trade frequency higher would result in the AI teams getting more efficient use out of their players by making more trades helpful to both teams, such as trading away that surplus player that would be a starter on another team.

If you have a certain tendency regarding the prospect/veteran strategy, you could set the favor prospect/veterans setting to be tougher for you. For most players, it seems they mostly seek to trade their aging vets in order to acquire prospects to develop and use in the ML, so having the AI favor prospects would make that more difficult. Though I'm not sure if that setting actually is about prospects/vets, because it's unclear to me how that would work. I suspect it may be about how much the AI favors potential ability versus current ability.

rink23 · 08-04-2018, 06:49 AM

Just noticed a major issue as I went through my 2020-21 offseason.
Most of the time, when making trades I have given up some top-line prospects to players with proven MLB value.
Did not think of anything of it.

This offseason, during the free agency period there were some players I missed out on.
As we moved, past the first of the year, there were some 2.5, 3 star minor leaguers who I shopped because they were blocked at the major league level.
The offers I got back were pitchers like James Paxton (who the Reds just outbid me on a month ago), Madison Bumgartner and Gerrit Cole (who another team outbid me on, after I let him get into free agency). The Dodgers also offered a four-star closer they had just signed the day before.
They were giving up 3.5-4 star pitchers for players who have never played in the Majors and would probably never be more than utility players.
I enjoy playing the GM role, but I don't like fleecing people to the point of it being downright stupid.
I played with the trade settings and used Dakota Hudson as bait, I was still getting 3-4 star offers for him.
This is not good.

Westheim · 08-04-2018, 10:27 AM

Quote:

Originally Posted by Timofmars

Most of the bad trades are either a result of the AI seeing value in useless players for some reason, or in their disregard for cash and salary retention.

Sometimes, very occasionally, I feel like the AI does know more than the numbers or scouting reports show. I mean, I get that they have a different scouting department, and some scouts will think that Ling-Ming Poo is the next six-time MVP while everybody else is sure that Poo is just Poo.

But sometimes I get the feeling that they know something that I don't know, which then messes with my head. When they come five times - different teams! - and want THIS mediocre player.

This is also true in other aspects of the game. I have an outfielder. 24 years old, fourth-year player. I was convinced when he was 20 that he was the key piece to my next championship lineup. Since then he has rode the train to AAA and back quite a bit. Career .224 batter. Batted .224 last season. Now batting .262/.349/.436 with 22 walks just before the halfway point. Half of those intentional. He has six homers, 16 extra-base hits. He is not batting eighth. Why is the AI walking him intentionally? The guy has done nothing to warrant this many intentional walks. Behind him I have two reliable singles slappers with higher batting averages. They gain nothing with the intentional walk.

They would also like to trade for him, although his stats say he's never going to be of any use whatsoever.

And all of this makes me think the AI knows more than I do, he will break out any minute now, and so I have to keep playing him so I don't miss his (overdue!) breakout…

Maybe I really *am* outsmarted in doublethink by a glorified calculator...

PSUColonel · 08-05-2018, 01:42 PM

Don’t forget....AI evaluation plays a role as well. I am using hard/numeral trade settings with AI evaluation at 65/20/10/5 and am finding trading to be mostly challenging. Very hard could be an option as well.

Orcin · 08-05-2018, 01:44 PM

Quote:

Originally Posted by PSUColonel

I am using hard/numeral trade settings with AI evaluation at 65/20/10/5 and am finding trading to be mostly challenging.

These are the settings that I use as well.

PSUColonel · 08-05-2018, 04:37 PM

I have to admit...I am still torn between 65/20/10/5 & 55/25/15/5 .....but I just bit the bullet and went more ratings heavy....most teams MLB probably do this as well. They will give highly scouted players a very long leash.

RchW · 08-05-2018, 07:16 PM

I'm heavily stats oriented using 2-8 ratings.Makes it difficult but enjoyable.

Reed · 08-05-2018, 07:35 PM

AI evaluation (ratings, current stats, last year stats, 2 years ago stats) according to the manual is used by the AI for line up construction. Does anyone know for sure if this setting is also used by the game engine trade logic (the manual does not say)?

PSUColonel · 08-05-2018, 07:47 PM

Quote:

Originally Posted by Reed

AI evaluation (ratings, current stats, last year stats, 2 years ago stats) according to the manual is used by the AI for line up construction. Does anyone know for sure if this setting is also used by the game engine trade logic (the manual does not say)?

I believe it does...and also contract negotiation too...which make me curious about what kind of results Rich sees with such a stats heavy evaluation....surely the AI must over/ under value players based on small/ medium sample sizes.

RchW · 08-05-2018, 08:21 PM

Quote:

Originally Posted by PSUColonel

I believe it does...and also contract negotiation too...which make me curious about what kind of results Rich sees with such a stats heavy evaluation....surely the AI must over/ under value players based on small/ medium sample sizes.

That's how it works IRL.

PSUColonel · 08-05-2018, 08:52 PM

Quote:

Originally Posted by RchW

That's how it works IRL.

very true

NoOne · 08-09-2018, 12:30 AM

Quote:

Originally Posted by rink23

Just noticed a major issue as I went through my 2020-21 offseason.
Most of the time, when making trades I have given up some top-line prospects to players with proven MLB value.
Did not think of anything of it.

This offseason, during the free agency period there were some players I missed out on.
As we moved, past the first of the year, there were some 2.5, 3 star minor leaguers who I shopped because they were blocked at the major league level.
The offers I got back were pitchers like James Paxton (who the Reds just outbid me on a month ago), Madison Bumgartner and Gerrit Cole (who another team outbid me on, after I let him get into free agency). The Dodgers also offered a four-star closer they had just signed the day before.
They were giving up 3.5-4 star pitchers for players who have never played in the Majors and would probably never be more than utility players.
I enjoy playing the GM role, but I don't like fleecing people to the point of it being downright stupid.
I played with the trade settings and used Dakota Hudson as bait, I was still getting 3-4 star offers for him.
This is not good.

this is the major trend in mlb -- houston and cubbies made it 'hot' because it worked for them.

that's why it was such a buyer's market this year at trade deadline. everone and their brother was selling if not in playoff race... not just KC to Yankees.

NoOne · 08-09-2018, 12:33 AM

Quote:

Originally Posted by Westheim

Sometimes, very occasionally, I feel like the AI does know more than the numbers or scouting reports show. <snip>.

most defintiely. you can test this by changing a player in editor while in trade screen and the ai will changes its feelings in real-time without a scouting report.

i'm not saying it sees through everything, but at least for some things and in some contexts it does.

jimmysthebestcop · 08-09-2018, 02:43 AM

Often you won't know a trade is fair until years later 3+. I love it when all of a sudden I am facing a team in the playoffs or the WS and I notice they were a frequent trade partner and like 6 of my former prospects are their starters now.

MrDov · 08-12-2018, 12:17 AM

The one idea I always circle back to is every team is going to evaluate players differently. My scout can have a 40 FV on a guy while another team's scout can have a 50 FV on him. Happens in real life, no reason it couldn't happen in OOTP. Just because my scout thinks I'm fleecing the AI doesn't mean I actually am. I've had my fair share of 35 or worse FV guys I give up turn into solid major leaguers or even all-stars. I'll check myself when it comes to the AIs roster construction to make sure they aren't adding a guy they don't need or giving up a guy they do, but as far as player value trading hands, I tend to let the AI make its own decisions.

Orcin · 08-12-2018, 12:38 AM

I can assure you that you will know when you are truly taking advantage of the AI. I can also assure you that the game is far more enjoyable when you don't take advantage of the AI.

I continue to hope for the day when the AI is able to take advantage of me. Until then, I must arrange my own competitive landscape.

08-03-2018, 11:48 PM	#24
NoOne Hall Of Famer Join Date: Apr 2015 Posts: 7,167	the teams i trade with usually do well afterward or at least better in part due to the players received. i'm trading talented mlb players by 29-32 more times than not. the way i do contracts typically won't cause a long-term burden to the destination team. if they aren't better off it is invariably because the ai did something dumb or catastrophe occured - tcr or age or injuries, et al. they get 1-5 years of good talent/cost ratio. Last edited by NoOne; 08-03-2018 at 11:50 PM.

08-04-2018, 06:49 AM	#26
rink23 Hall Of Famer Join Date: Apr 2013 Posts: 18,033	Just noticed a major issue as I went through my 2020-21 offseason. Most of the time, when making trades I have given up some top-line prospects to players with proven MLB value. Did not think of anything of it. This offseason, during the free agency period there were some players I missed out on. As we moved, past the first of the year, there were some 2.5, 3 star minor leaguers who I shopped because they were blocked at the major league level. The offers I got back were pitchers like James Paxton (who the Reds just outbid me on a month ago), Madison Bumgartner and Gerrit Cole (who another team outbid me on, after I let him get into free agency). The Dodgers also offered a four-star closer they had just signed the day before. They were giving up 3.5-4 star pitchers for players who have never played in the Majors and would probably never be more than utility players. I enjoy playing the GM role, but I don't like fleecing people to the point of it being downright stupid. I played with the trade settings and used Dakota Hudson as bait, I was still getting 3-4 star offers for him. This is not good. Last edited by rink23; 08-04-2018 at 04:29 PM.

08-05-2018, 07:16 PM	#31
RchW Hall Of Famer Join Date: Jul 2004 Location: The big smoke Posts: 15,628	I'm heavily stats oriented using 2-8 ratings.Makes it difficult but enjoyable. Attached Images __________________ Cheers RichW If you’re looking for a good cause to donate money to please consider a Donation to Parkinson’s Canada. It may help me have a better future and if not me, someone else. Thanks. “Conservatism consists of exactly one proposition …There must be in-groups whom the law protects but does not bind, alongside out-groups whom the law binds but does not protect.” Frank Wilhoit

08-03-2018, 06:24 PM	#21
waittilnextyear All Star Starter Join Date: Jan 2016 Posts: 1,272	I think it could be well nigh impossible to come up with a quantitative, predictive model for whether a trade should be deemed "fair." For a number of reasons mostly relating to context and league settings as well as the sensibilities of different users. Imagine placing every possible variable (age, salary, team control, performance this year, last year, team goals, positional depth...) into a very large matrix--this would probably be the only truly quantitative way. It would seem horrendous to try that for each deal you're considering. Also, you can't really judge a trade until some time passes, so something that checks all the boxes up front might end up going sour later. And, of course, if we're going for strict realism and using baseball history as a guide, there should be some non-zero fraction of trades that are not really fair at all. Both in the moment and after the fact. However, it's not altogether a bad idea to try to come up with some sort of "code" to live by in terms of making your own trades. Because we all surely hate the feeling from taking advantage and the internal pressure to...gasp...restart. I think people finding their own criteria and sense of "feel" is as good as it gets. One idea might be, like we've done on Operation Sports for the MLB The Show series, to start a trade thread where people can propose things and get feedback--getting 2 or 3 people saying "yeah, that seems legit" can make people feel justified in making a particular trade and less likely to feel "dirty" later on. Some things I do in my own OOTP save to keep myself honest... (1) I've begun giving myself a maximum of 5 trades per calendar year. This makes me really have to prioritize (I can't salary dump everyone like I otherwise might). Ultimately, if I've used my last trade on July 8th, then I have to pass on every subsequent offer from the AI no matter how much it could help me. Quantity or trade volume is something that is easy for me to control without going above and beyond to devise an elaborate system. As a peripheral benefit, it also helps me hang onto some of my players so they have longer careers with my franchise (which was another issue for me and roster churn). (2) The settings are helpful. I use "hard" and "favor prospects." I think this combo does a pretty decent job of putting upper bounds on how much advantage I can squeeze out of a deal and how much salary I can dump. Of course, I want to get some advantage from the trade otherwise there's no incentive for me to make it. The question is whether I can live with the magnitude of that advantage. Or not. I would suggest people play with not only the easy/medium/hard and favor prospects/veterans, but to adjust AI evaluation settings/percentages as well. Use a test franchise and season to taste--keep doing what you feel is an unfair deal and change the settings to where you have the toughest time doing it. (3) I really try to look at the deal from the opposing GM's viewpoint and ensure that there is some logical motive for them. As I balance that with trying to rip them off, lol. It doesn't mean they're getting the better of it necessarily, but it means that I can't have everything. If I am getting the best player, and the cheapest player, and the most players, and on and on, then that feels like an exploit. But if I get the best player and that player makes good money, and his team is over budget, and operated at a loss last year, then I can start to see some logic there. Or, if I send a decent starter to a contending team that has 3 starters on the DL for the forseeable future. Or, if I am taking a veteran and some salary from a bottom-feeder. It helps that there is some reasonable facsimile of a real-life trade. To the OP, though, you may not really get a truly quantitative, by-the-numbers modus operandi out of this, but maybe people can reflect on what they find to be fair/unfair generally and this thread becomes helpful in that way.

08-04-2018, 03:52 AM	#25
Timofmars Minors (Triple A) Join Date: Feb 2018 Posts: 251	I don't like to handicap myself beyond what the game offers in settings and situations. I'll play for a low budget team with very hard trade difficulty, but I don't like to have to decide cancel trades because the AI is being stupid. That makes it feel less like a game and more like a simulation model. I want a game, not a model. In fact, I like to mess with the game world by giving the richest teams the worst deals they will accept, and saving my best quality unwanted players for the weakest low budget teams. I like seeing the standings be in reverse from what they usually should be, and I feel happy for the Marlins being at the top. Most of the bad trades are either a result of the AI seeing value in useless players for some reason, or in their disregard for cash and salary retention. So perhaps you could avoid most bad trades by refusing to add low overall rated players in order to complete trades (possible exception being where the player is an improvement to the team), and not taking cash or asking the AI to retain salaries. Another area the AI may make strange decisions is in how they value players they don't need. They seem to be happy to take a talented player even if they already have a better player in that position. Although a team that is weak in that position would probably put much more value on the player and give up more to get him. Anyway, I don't think this is really a problem, since a talented player is valuable to someone, so they could trade him for something they actually need. But maybe setting the AI trade frequency higher would result in the AI teams getting more efficient use out of their players by making more trades helpful to both teams, such as trading away that surplus player that would be a starter on another team. If you have a certain tendency regarding the prospect/veteran strategy, you could set the favor prospect/veterans setting to be tougher for you. For most players, it seems they mostly seek to trade their aging vets in order to acquire prospects to develop and use in the ML, so having the AI favor prospects would make that more difficult. Though I'm not sure if that setting actually is about prospects/vets, because it's unclear to me how that would work. I suspect it may be about how much the AI favors potential ability versus current ability.

08-05-2018, 01:42 PM	#28
PSUColonel Hall Of Famer Join Date: Feb 2002 Posts: 12,990 Blog Entries: 37	Don’t forget....AI evaluation plays a role as well. I am using hard/numeral trade settings with AI evaluation at 65/20/10/5 and am finding trading to be mostly challenging. Very hard could be an option as well.

08-05-2018, 04:37 PM	#30
PSUColonel Hall Of Famer Join Date: Feb 2002 Posts: 12,990 Blog Entries: 37	I have to admit...I am still torn between 65/20/10/5 & 55/25/15/5 .....but I just bit the bullet and went more ratings heavy....most teams MLB probably do this as well. They will give highly scouted players a very long leash.

08-05-2018, 07:35 PM	#32
Reed All Star Starter Join Date: Oct 2013 Posts: 1,947	AI evaluation (ratings, current stats, last year stats, 2 years ago stats) according to the manual is used by the AI for line up construction. Does anyone know for sure if this setting is also used by the game engine trade logic (the manual does not say)?

08-09-2018, 02:43 AM	#38
jimmysthebestcop Banned Join Date: Jun 2018 Posts: 1,728 Infractions: 0/2 (5)	Often you won't know a trade is fair until years later 3+. I love it when all of a sudden I am facing a team in the playoffs or the WS and I notice they were a frequent trade partner and like 6 of my former prospects are their starters now.

08-12-2018, 12:17 AM	#39
MrDov Minors (Triple A) Join Date: Aug 2012 Posts: 275	The one idea I always circle back to is every team is going to evaluate players differently. My scout can have a 40 FV on a guy while another team's scout can have a 50 FV on him. Happens in real life, no reason it couldn't happen in OOTP. Just because my scout thinks I'm fleecing the AI doesn't mean I actually am. I've had my fair share of 35 or worse FV guys I give up turn into solid major leaguers or even all-stars. I'll check myself when it comes to the AIs roster construction to make sure they aren't adding a guy they don't need or giving up a guy they do, but as far as player value trading hands, I tend to let the AI make its own decisions.

08-12-2018, 12:38 AM	#40
Orcin Hall Of Famer Join Date: Aug 2010 Location: Indiana Posts: 9,798	I can assure you that you will know when you are truly taking advantage of the AI. I can also assure you that the game is far more enjoyable when you don't take advantage of the AI. I continue to hope for the day when the AI is able to take advantage of me. Until then, I must arrange my own competitive landscape.