Trade Issues

[ra7c7er]

Originally Posted by lejdot

I dont have that much in this game but have extensively played FM2014 and I am a little dismayed I get so many trade requests in OOTP but none are remotely interesting. They are asking for five star potential rookies and offering one star nobodies or 3 star 15m/yr vets.

At a minimum there should be at least one or two apples to apples trade offers a year.

Exactly I get the same thing. Just a few minutes ago I got a trade for a perennial .220 hitting veteran (2.5 star), a 2.5 star and 1.5 star minor league players with no extra potential for my top prospect (7th overall 3.5 star w/5 star potential) and my number 3 starting pitcher (4 star with 4.5 star potential). It's like seriously? are you kidding. Give me those three plus one of your starting pitchers a decent prospect and you better throw in a high draft pick or some cash to cover your overpaid veteran.

What really irks me and I've been seeing more an more is when they AI sends a trade request which isn't remotely something worth considering but you hit discuss to try and make it better but the AI says on the trade screen "I'll have to think about it". Seriously you're already giving me garbage for one of my top players and you "have to think about" even the original trade request. Game really failing in the trade department right now.

Another thing I've realized recently is the AI either doesn't value draft picks at all or the game give such lopsided trades that adding the top draft picks doesn't change the "Yes I accept this trade". No joke just trying to see what would happen I discussed a trade and added the other teams top FIVE picks in the draft to it. The AI considered the trade request so lopsided from the beginning that even after I added the top 5 draft picks, rounds 1-5, the AI still said "I accept this trade". I went further and it took the first 8 picks to get them to say "I'll have to think about it". Again the trade was so lopsided to begin with that the AI was willing to not only give up the original trade but the first 7 picks of draft. That right there shows either a problem with the value of draft picks so shows how stupidly lopsided trading is. And this scenario has happened more than once. Granted not always the top 7 picks but the top 5 or more.


As I said in other posts I'm not looking for completely even trades every time but the trades need to run both sides of the fence. Good, bad, and even.

[PSUColonel]

My issue is this: just because a trade is even...doesn't make it in the best interest of an AI team. What if the team is OK at a certain position?, what are financial considerations?.. are the teams involved rebuilding or trying to improve for the near future? ...there are and should be so many factors involving a trade. Just being "even" is not enough IMO.

[Lukas Berger]

Originally Posted by PSUColonel

My issue is this: just because a trade is even...doesn't make it in the best interest of an AI team. What if the team is OK at a certain position?, what are financial considerations?.. are the teams involved rebuilding or trying to improve for the near future? ...there are and should be so many factors involving a trade. Just being "even" is not enough IMO.

Exactly. Most of the people in this thread are looking at this only from their standpoint. But in real life teams no doubt turn down fair trades all the time because they just don't want the player being offered or don't want to give up their guy or whatever. Maybe they don't want to mess up their chemistry or create change simply for the sake of change.

For those that play fantasy sports, how many times have you turned down an offer that was fair, or even one in your favor because you wanted to keep your player? I've done it often. If we do it in fantasy, how much more are real teams dealing with real people likely to do it?

It's a strength of the game especially compared to console games where you can trade for any player at any time if the offer is good enough. If you don't like the trade offers, just roll up your sleeves and get down to negotiating with the ai. Still no good? Wait a week or a month and try again.

You aren't going to get handed trades that are good for you on a silver platter. You have to work for them.

[RchW]

Originally Posted by PSUColonel

My issue is this: just because a trade is even...doesn't make it in the best interest of an AI team. What if the team is OK at a certain position?, what are financial considerations?.. are the teams involved rebuilding or trying to improve for the near future? ...there are and should be so many factors involving a trade. Just being "even" is not enough IMO.

Well said.

In another thread today somebody complained that the AI wasn't making offers for players on the trading block. To me that's exactly correct. The AI should be looking for players to improve its team not "your" lead weight contracts. Now if needs do match then...

I find many people have a "fantasy league" concept of trades. In fact if you want to obtain a key player for a championship run you should be prepared to overpay or lose the talent equation for the benefit that follows.

[SirMichaelJordan]

Originally Posted by RchW

Well said.

In another thread today somebody complained that the AI wasn't making offers for players on the trading block. To me that's exactly correct. The AI should be looking for players to improve its team not "your" lead weight contracts. Now if needs do match then...

I find many people have a "fantasy league" concept of trades. In fact if you want to obtain a key player for a championship run you should be prepared to overpay or lose the talent equation for the benefit that follows.

That's a good point.

Many people would loose their mind playing with very hard trading with this concept. I personally enjoy the challenge of not being able to always improve my team on the fly or take advantage of the AI. I don't mind feeling like I am on the losing end of a trade because it's so much more satisfying when I feel like I've 1up the AI in a deal I've only won 2 World Series since OOTP 13. (about 20 seasons worth)

...And I am LOVING IT!

[Rembrandt]

i think the biggest problem is the AI won't "overpay" even when they could use the player because they have a weakness and are going all in. An "overpaid" guy for a half a season in a contract year I should be able to move for a 2 or 3 star prospect instead of just being told no...unless I include a top 5 prospect from my organization.

[RchW]

Originally Posted by Rembrandt

i think the biggest problem is the AI won't "overpay" even when they could use the player because they have a weakness and are going all in. An "overpaid" guy for a half a season in a contract year I should be able to move for a 2 or 3 star prospect instead of just being told no...unless I include a top 5 prospect from my organization.

I know this can be frustrating but the trade AI has to be less flexible or it becomes an exploit. In earlier versions of the game we could do just what you say and it led to easy dominance because you could keep all the best prospects.

In v15, I'm trying to move my very good but replaceable pending FA closer with money and the AI wants my hot shot prospect SP who went 12-3 in 21 starts as a mid season call up. I can't argue with that logic.

[hluraven]

I always used to think of it that on average you could make a "winning" trade to 15 out of 29 teams, on hard it was 5 out of 29 and on very hard it was 1 out of 29. But, if you just keep on trading, even on very hard you can exploit the AI. It takes a long time and is quite frustrating (especially as shop a player never returns anyone on very hard) but you can get there with a lot of hard work.

I've not tried the new patch AI setting in a solo league, so I'm guessing it's moved from 15 to 10 out of 29 teams you can get a good trade from on average at a time.

[PSUColonel]

Originally Posted by RchW

I know this can be frustrating but the trade AI has to be less flexible or it becomes an exploit. In earlier versions of the game we could do just what you say and it led to easy dominance because you could keep all the best prospects.

In v15, I'm trying to move my very good but replaceable pending FA closer with money and the AI wants my hot shot prospect SP who went 12-3 in 21 starts as a mid season call up. I can't argue with that logic.

what setting are you using? Markus has said the new average setting is 80 percent of "hard", and it favor prospects a bit more. What seems to be the most realistic setting at this point? Is it still "hard, "very hard"...or "average"? I realize that is a subjective question, but I'd like to see what the consensus is now that it's been re-coded a bit.

[RchW]

Originally Posted by PSUColonel

what setting are you using? Markus has said the new average setting is 80 percent of "hard", and it favor prospects a bit more. What seems to be the most realistic setting at this point? Is it still "hard, "very hard"...or "average"? I realize that is a subjective question, but I'd like to see what the consensus is now that it's been re-coded a bit.

Hard neutral.

[PSUColonel]

I can't decide which setting to use...average seems fair in some instances, but too easy in others. Hard or very hard seem tough...but I do want a challenge. In the end, I think if you are going to use the "shop player" function...it's almost necessary to use very hard.

[harrumph]

I’ve been playing in a fictional world with two leagues that, in format and quality, pretty closely match MLB and *** in real life. It’s been going for over forty years, and I recently switched teams after fifteen consecutive 100-win seasons with my old franchise (got sort of boring). I took over a small-market team with a demanding owner that had gone 74-88 the previous year. In my first year as GM, I won 108 games and the championship (beating my old team in the process and getting the Moneyball achievement).

Either I’m a genius, or something’s too easy. I’ve got trading set to hard, and I’ve generally found it to be a solid challenge, sometimes frustrating, sometimes rewarding. I wouldn’t have said, if you’d asked me yesterday, that the trading system was the reason the game’s too easy (other game settings, if you're curious: scouting accuracy low, actual ratings not displayed, everything else 20-80, injuries normal). But I decided to look at the trades I’ve made and evaluate them in retrospect—there have been, in 43 years, plenty of big ones where everybody involved is retired now.

The analysis of the trades is SUPER LONG, and lots of it involves me reminiscing about all my great old players and whatnot. If you like AARs or dynasty reports or whatever, you might enjoy it; otherwise you should just skip to my conclusions. I've indented it all in lieu of spoiler tags.

I’m going to start by looking at deals where I got high-ceiling guys who hadn’t yet reached their potential. Throughout all this analysis, I’m going to use WAR, flawed though it is, as a rough yardstick for measuring how much the various players were worth from the time of each trade onwards.

Katsunosuki Daikawa: The greatest pitcher in the history of my league. He pitched for twenty years and put up 136 WAR, including seventeen straight seasons of 5.5 or more WAR. At one point, he led the league in strikeouts ten years in a row, winning seven pitching Triple Crowns in the process. I got him for one highly-rated prospect and four guys I thought were going nowhere. When I went back and looked at how the trade panned out, I saw that, while two of them were busts, two had been very successful. One played for fourteen years, mostly as a starter, and put up 20 WAR. The other pitched for thirteen years, made an All-Star game, and put up 27 WAR. The higher-rated guy (#6 in the league) played fifteen years in the outfield, made one All-Star appearance, and won four Gold Gloves; he was worth 28 WAR altogether.

Obviously, I won this trade, but it wasn’t the highway robbery I thought it was. Three league-average players aren’t worth a Hall of Famer, but they’re a much safer proposition. What if my guy had been hurt? He actually had to undergo Tommy John surgery (at age 19) about six months after I acquired him. Luckily for me, he bounced back—but I later had another top pitching prospect I’d acquired in a trade, Chia-Ch’ing Dai (80/70/65 potential, ranked #7 at his peak—Daikawa had 80/60/80 potential and was ranked #1), tear his UCL (he was a little older, 22, which might have made the difference) and never recover his control. He ended up being a serviceable starter, but was also productive when I converted him into an outfielder, an entertaining experiment which is still playing out (6.2 WAR in two seasons as SP; 13.8 WAR in four seasons at LF).

Still, the initial trade was not a great one in retrospect: I got Dai and a journeyman reliever; they got three prospects who developed into journeymen (an outfielder who’s put up 9 WAR and a replacement-level infielder and reliever), an aging outfielder who put up a couple more good seasons, and one prospect (ranked #56 at the time of the trade) who did work out: a starter who’s going into his twelfth year and has put up 20 WAR so far—about the same as Dai, who’s the same sge. If three of the prospects I dealt had turned into 20+ WAR guys, as they did in the Daikawa trade, it would have looked pretty ugly.

Jovica Mladic: The most outrageous theft I perpetrated (in retrospect) was for Mladic, a first baseman who ended up producing 113 WAR in nineteen seasons, including an insane peak from age 26 to 30 when he put up over 10 WAR every season, batted almost .380, and cracked 201 homers. He won eight consecutive MVP awards, which I seriously doubt anybody will ever do again. To get him, I gave up three prospects: an unranked pitcher who bounced around the majors for five years, a #95-ranked shortstop who played ten decent years (18 WAR) and went to one All-Star game, and a #16-ranked outfielder who played parts of eight seasons, all around replacement level. I also shipped them an excellent closer who was getting too expensive—he cost almost $3m that year and got nearly $7m in arbitration the next year.

Manny Coronado: I got a Hall-of-Fame-worthy catcher (88 career WAR) who, at that point, was 21 and had played one full season—his performance that year was all glove, no bat, but my scout (correctly) predicted big things. I gave up three prospects who ended up making the majors: an outfielder (ranked #62) who put up 9 WAR in twelve years, appeared in one All-Star game, and won one Gold Glove; a second baseman (ranked #73) who also made one All-Star appearance and won a Gold Glove, but put up better numbers overall (26 WAR in thirteen years), and a second outfielder who played twelve years, made two All-Star appearances, won three Gold Gloves, and was worth 24 WAR. As with Daikawa, this is clearly a big win for me, but could easily have gone the other way if my guy had gotten hurt (and catchers, like pitchers, do get hurt).

Albert Castro: Another future Hall of Famer I scooped up when he’d made the majors but hadn’t gotten anywhere near his ceiling—ultimately worth 75 career WAR. He’d been called up a little too early and put up one disastrous full season (-3 WAR) and a couple unremarkable ones off the bench. I gave up two journeyman pitchers (worth 10 WAR combined from that point on), two replacement-level backup catchers, and one good prospect, who went on to put up 22 WAR in fourteen years at third and short, appearing in four All-Star games. This one is an unambiguous win for me, and the first trade I’ve looked at where it seems like the Ai just screwed up. The two catchers and one of the pitchers were known quantities with substantial major league service under their belts; the infield prospect was the only guy they could really hope for anything from, and he actually did quite well. Castro had put up truly horrendous numbers at age 20, but at age 22 (his last season before the trade), he’d shown strong improvement, posting an OPS+ of 99 in about half a season’s worth of at-bats. Seems like they were foolish to punt him, especially considering he’d been rated the #1 prospect in baseball two years earlier.

Attis Antonakos: Here’s an interesting trade—I thought I’d won, just based on the performance of the guy I got, but looking back at how things turned out, it looks more like a loss. I got a major-league ready center fielder who plugged a big hole in my defense and went on to produce 56 WAR and make eight All-Star appearances, so I felt pretty good about it. Turns out the guys I gave up eventually did even better: a shortstop who was worth 30 WAR across fourteen years, went to five All-Star games, and was voted MVP once; a catcher who played for six years, won a Gold Glove, and went to one All-Star game (but only put up 3 WAR; he aged really harshly); a center fielder who played eleven respectable years (13 WAR); another center fielder who played for thirteen years, was worth 30 WAR, played in five All-Star games, won a Gold Glove, and was voted Rookie of the Year; and, most gallingly, a corner outfielder named Udo Wacker. He played for fifteen years, mostly in an international league (like *** to my MLB). In my league, he was voted MVP once, went to three All-Star games, and was worth 13 WAR; then he went over there, won Rookie of the Year, appeared in nine All-Star games, won another MVP, and put up 46 WAR. Replacement level is lower there, but I can’t help suspecting that I didn’t even get the best player in this trade. Even converting WAR very conservatively from one league to the other, it looks bad. Better to have one 55-WAR guy than two 30-WAR guys, sure, but two 30-WAR guys and a 13-WAR guy and a 40+ WAR guy? Whoops.

Pistol Hasegawa: In a similar trade decades later, I got a center fielder who’d played four years in that same international league (still had a minimum contract, though; they’ve developed very player-unfriendly rules). He ended up being worth 40 WAR during his career in my league. To get him, though, I gave up five prospects who, again, surprisingly all ended up getting major league playing time: a replacement-level second baseman and left fielder, both of whom bounced around for a decade or so; a closer who ended up throwing fourteen solid seasons in my league (12 WAR); a center fielder who spent his entire eleven-year career abroad, putting up 17 WAR, winning a couple Gold Gloves, and appearing in a couple All-Star games; and a starting pitcher who played for seven years over there, put up 20 WAR, won a Cy Young, appeared in three All-Star games, and then came back to my league and did it all again: eight years, another 20 WAR, and two more All-Star appearances. Once again, I might not even have gotten the best single player in the deal, and I certainly gave up more total value than I got.

Aurélien Chapuis: I got Chapuis, a third baseman who had a major-league contract but almost no playing time, for a good young starting pitcher I owed $15.6m over four years, my starting third baseman, whom I owed $8m over three, and some minor-league filler that never panned out. Chapuis ended up being worth 41 WAR; the previous 3B tanked but somehow managed to accumulate -7 WAR over the rest of his career, because various teams trotted him out for another twelve years. The pitcher threw another fourteen years, became an All-Star, and put up 31 WAR. A successful salary dump and a big return on a prospect for me, but a pretty good deal for them too—they traded their lousy new third baseman before he did much damage.

Urbanus Gogele: A catastrophe in progress. I gave up a pitching prospect who has been worth 27 WAR and an infielder who’s been worth 15 WAR at third and short, both of whom are in their early thirties and still producing. Plus Gogele, a first baseman who’s been worth 54 WAR through age 29 and just won his first, but probably not last, MVP award. I got two pitchers; one ended up as a replacement-level reliever and the other just finally had his first good season as a starter (also age 29). In my defense, I had an aging Hall of Famer (Mladic) at first and an organization stacked with 1B prospects who all basically looked the same: no glove, no speed, 45/50/55. I picked the guy with the best makeup and dealt the others (Gogele has a “very low” work ethic, supposedly). The two pitchers were both top-ten prospects—80/65/65 (#1) and 80/50/60 (#9). Gogele never cracked the top 100 as a prospect. Just goes to show you: there’s no such thing as a pitching prospect.

José García: This one hasn’t played out fully yet, but I’m ready to call it. I scored García for a very good closer, a utility infielder, and three pitching prospects who are now all in their late twenties or early thirties and have thus far produced about 10 WAR combined. García has been worth 33 WAR through his age-25 season. A huge win for me.

Shinji Fukuda: He was considered a middle reliever by his organization, but still ranked very high (#21 the year of the trade). My scout thought (correctly) that he could start; after three years in the pen, he’s given me five solid years in the rotation (25 WAR altogether). I also got an excellent young reliever who immediately donated a kidney to his sister and retired (c’est la vie!) and a less excellent, much younger reliever who finally seems to be hitting his stride now in his mid-twenties. I gave up three middle-tier prospects (#65, #85, and #91), some guy drafted in the ninth-round, and a veteran outfielder making the league minimum who retired within the year (not to donate a kidney). They have collectively accomplished nothing so far. Another failure for the AI, although maybe the problem here was just that they underrated Fukuda.

Michael Gaines: A big whiff for me. I had a 30-year-old pitcher, a borderline starter who was making $5m/year basically just as a hedge against injury. Mid-season, nobody was hurt and I wanted to free up some budget room, so I dealt him for a sharp-looking outfield prospect. The team I traded him to used him out of the pen for a year, then in the rotation for two. He then got a $46m/4 deal in free agency from a third team and spent four years as a closer. Altogether, he was worth 16 WAR from the trade on; he’s retired now. Gaines was worth 42 WAR; he’s now an old bench player and not likely to put up much more. Unfortunately, I also gave up prospects of my own to get Gaines: a solid first baseman worth 23 WAR, a right fielder who was worth 10 WAR, a shortstop who was worth 20 WAR, and a guy named Yano who ended up being a solid mid-rotation starter, worth 59 WAR so far and still going at age 39.

Now some trades for established players.

Kiyomitsu Fujimoto: I got an MVP first baseman who had just, at age 28, had his least-productive season ever. I gave up a replacement-level backup catcher, an infield prospect who went on to put up 16 WAR in thirteen seasons as a regular, and three major leaguers: a middle reliever (owed $1.2m/1), my starting left fielder ($41m/4), and my starting second baseman ($4m/1). Fujimoto was owed $37m over three years, so the money was pretty much a wash. Their salary commitments went up by a couple million in the first year, down a couple million in the next two, and up $10m in year four.

The reliever played another five years, was worth a couple WAR, and didn’t cost much. Fujimoto, the second baseman, and the left fielder were the big pieces. The two guys I was dealing were both thirty, both regular All-Stars, and had both just put together respectable seasons (2.5 WAR apiece). Fujimoto had been worth 4.9 WAR all by himself, but that was a big step down for him—he’d produced 7.6 WAR at the age of 20 and put up 6-10 WAR every subsequent season. The wrinkle in this trade was that he played for my-ex employer. In fact, he was my baby, my favorite prospect. I scooped him up as an international amateur free agent in my first year as a GM, shepherded him through the minors, won a championship with him—and then was cruelly separated from him when my ultra-demanding owner fired me—one year after I’d delivered our second title!—for finishing second in the division (never mind that we won 91 games). I was so bereft at losing this team I’d so carefully built over the course of eleven seasons that I almost quit and started over (in addition to Fujimoto, the team had an eventual Hall of Famer—then only 23—at short, a third baseman who debuted at age 32 and still managed to rack up over 2000 hits, and the best reliever ever, a guy whose career ERA+ ended up at 198, at closer).

I was convinced that Fujimoto was destined to be the greatest of all time and determined to get him back; I’d been plotting it ever since I got the sack. When he put up this great but not elite year, I decided to go for it. It actually seemed like a risky move at the time, giving up two of my starters and a very good prospect for a guy who might be in decline. But it paid off ridiculously well, because I wasn’t just blinded by love/desire for revenge: he really was the greatest. In his first year with our new team, he hit 47 home runs (the best mark of his career) and was worth 8.5 WAR for the season. He went on to produce another 83 WAR for me, setting career records for hits (3664), doubles (827), home runs (701), and about a dozen other things. The left fielder and second baseman both washed out four years later, having produced sub-replacement-level numbers. So it was a massive, massive win for me, but I don’t think anybody would have predicted just what a disaster it would be for my old team (who got exactly what they deserved).

Tudor Vernis: I got five late-career years from a star SP, who put up 15 WAR in that time. He had one $17.4m year left on his contract; I was able to negotiate down to $7m for the next two seasons, then $4m for the two after that. So I paid $40m for 15 WAR—not very good (above-average regulars typically get $3-4m/year in this league; $8m/year is a star player’s pay). They got a bunch of filler; it was basically just a salary dump. One guy turned out to be a decent journeyman SP, worth 11 WAR or so for his career. A win for the AI.

Eustace Vassilopulos: I got an established third baseman who cost $18m over four years and produced about 11 WAR in that time; I gave up a bunch of pitching prospects, one of whom was a bust but the rest of whom had decent careers (6-10 WAR apiece) as relievers. I didn’t get great value, but Vassilopulos filled a need, I could afford to pay him, I had a strong bullpen already, etc. A win-win trade.

Bill MacLeod: I gave up an outfield prospect, Kinji Tanaka, who went on to put up 22 WAR in my league and another 46 abroad. I got MacLeod, who turned out to be a replacement-level catcher, and some filler. A colossal bust.

Sebastian Blunsden: I got him and a prospect (who didn’t work out) just four months after he signed a $25m/3 extension. He’d only pitched four years in the bigs but had established himself as one of the best starters in the game (19 WAR in the three years before I acquired him); he went on to win over 200 games with a sub-3.00 ERA and might have had an outside chance at 300 if injuries hadn’t ended his career at age 36. To get him, I gave up three good prospects—#22, #35, and #61 (two corner outfielders and a third baseman) and a fourth guy who wasn’t very highly rated. Although the AI got three top-100 prospects, this seems like a terrible deal even if you don’t know how it turned out (#22 had a pretty good career, worth 25 WAR, but the other guys were busts). Blunsden wasn’t injury-prone (he’d only been on the DL once, minors or majors), he was crushing hitters, and they’d just signed him to a very team-friendly extension. Stupid trade by the AI.

Feng-Liang Xiang: I needed another big bat on a team that was basically just Fujimoto and a bunch of slap-hitting OBP guys, so I spent a bunch of prospects on this guy. He was 24 and had only just started to hit homers, but my scout graded his power potential 80. He ended up delivering a Jim-Thome-like rate of one homer every 13.9 at-bats and hit .303 in the bargain, giving me 47 WAR in eleven years. I gave up a bunch of prospects: an outfielder who ended up with 15 WAR, a shortstop ultimately worth 22 WAR, an outfielder who was basically a bust, and a first baseman who produced 20 WAR. I also threw in the right fielder Xiang was replacing; he had been an average 2-WAR-per-year guy for a while but only had one decent year with his new team before he slipped down to replacement level. This is a case where, although the total value I gave up ended up being more than what I got, I think it was kind of a steal.

Bob Smith: Another recent one that I’m already ready to pass judgement on. It was four of my guys for three of theirs, but only three of them are of consequence. Smith was 26, had just thrown three excellent seasons as a starter, and was still cheap in arbitration. I sent them a high-ceiling, low-floor young center fielder (the #11 prospect in the league) and a 34-year-old starter who’d been putting up numbers very similar to Smith's (but who was in the last year of a $9.2m/year contract). Six years later, the pitcher’s retired, the center fielder has been hobbled by injuries, and Smith is going strong, earning a very team-friendly $6m/year.

So it’s been a fun trip down memory lane, but I’m starting to scrape the bottom of the barrel—not so many blockbuster trades left to look at. The AI has actually acquitted itself pretty well; of these trades, I’d say ten were wins for me (Daikawa, Mladic, Coronado, Castro, García, Fukuda, Fujimoto, Blunsden, Xiang, and Smith), two were neutral (Chapuis and Vassilopulos), and six were wins for the AI (Dai, Antonakos, Hasegawa, Gogele, Vernis, and MacLeod).

The problem is that the wins are a lot bigger than the losses: Daikawa, Castro, Fujimoto, and Blunsden are in the Hall of the Fame, Mladic and Coronado are locks as soon as five years are up, and Smith and García might make it if they stay healthy. Gogele is the only player I traded away as a prospect who I expect to see in the Hall of Fame, and he, Wacker, Yano, and Tanaka are the only real star players in the bunch (Yano probably would’ve made the Hall had he played for better teams; his record stands at 172-143 at age 39). Most of the “wins” for the AI involved bunches of thoroughly average players—guys who played 10-15 years and were worth 20-30 WAR.

It’s not luck or bad scouting; the AI knows that it’s giving up the best player in the deal (remember the foremost cliche of real-life trading: the team that gets the best player wins). It just doesn’t properly value its strongest assets. Daikawa and Castro were #1 prospects. Mladic was #11. García was #11 at the time of the trade and eventually made it to #2. I traded for some top prospects who didn’t pan out, too. Dai was #7. Chaleev and Riehl—the guys from the Gogele deal—were #1 and #9. It’s crazy that I was able to get so many top prospects while giving up so little. In all these deals, I only gave up three top-twenty prospects; in two cases I was trading up—a #6 to bag Daiwaka and a #16 for Mladic—and in the other I got a young, cost-controlled All-Star starting pitcher. The success of the most productive players I gave up as prospects, on the other hand, has been relatively surprising—a matter of luck, you could say. Yano was ranked #86 at the time of his trade and peaked at #56. Wacker was unranked at the time and peaked at #50. Tanaka was unranked and peaked at #46. Gogele never made the top 100 at all.

When you look at the established players, it’s even crazier. I’ve traded for four Hall of Famers in their prime (Blunsden, Fujimoto, and two guys whose trades I didn’t analyze in detail, Sepúlveda and Sánchez). Each was between 25 and 28 and had been an All-Star at least once. Sepúlveda had already won a Cy Young; Blunsden had been runner-up. Fujimoto had been MVP twice and Rookie of the Year. In cases like these, the AI should simply not be willing to trade. They should be planning around these guys, positioning them to be franchise players, and keeping them off-limits to other GMs.

This doesn’t entirely explain why I was able to turn my new team around so quickly, but it does explain why the game is generally too easy. It undermines the cyclical, equalizing processes in baseball’s structure. If you’re a perennial winner, you’re supposed to perennially miss out on the very best draft picks. You take a 50 or 55 guy in the first round every year. To get young guys with impact potential, you have to spend big bucks on international amateurs. Getting a 75 or 80 prospect as an international free agent should be like Christmas. Instead, all you have to do is take your 55 guy from the first round and your 45 guy from the second round, sign some more 45s and 50s as cheaper international amateurs, and throw five of them at the last-place team. You get a future Hall of Famer; they get two or three guys who might end up being productive regulars. Long-term, they get mediocrity and you get immortal glory.

The craziest thing I noticed in all my poking around BNN player pages and whatnot was that Daikawa and Blunsden—respectively first and fifth all time in strikeouts, first and fifth all time in VORP, first and third all time in WPA, third and fourth all time in QS%—were signed (one as an international amateur, the other the #4 draft pick) in consecutive years by the same team. The players they got from me in those two trades actually did give them a shot in the arm, carrying them to the only three playoff appearances (and two championships) they’ve ever made. The two guys I got from them, though, cemented a dynasty that, from the year of the second trade on, won the division twenty-one times in a row (with nine championships). No AI-controlled team has even had ten consecutive winning seasons yet.

TL;DR conclusions:

Trading is too easy, even on hard, and it's the biggest reason the game is too easy in general.

It boils down to an old baseball cliche: the team that gets the best player wins the trade. The AI basically does a good job of getting total value in trades. It does a very bad job of concentrating that value in individual star players. Generally, it's happy with a trade as long as the total value on its side exceeds the total value on your side by a certain margin (this, I'm guessing, is also the root of problems like the AI accepting trades for guys they didn't want when they were free agents two weeks ago). I think that the AI wants to "win" each trade by an absolute margin rather than one relative to the total value in the trade. This means that low-value trades (say, a good reliever for a middling prospect) are almost impossible to pull off—they simply won't accept any trade unless they're getting > X value from you or dumping > -X value (bad contracts, etc.) on you—while blockbuster trades are often relatively easy.

Your fundamental goal in trading should be to concentrate value: trade a bunch of average players up for a single good one, then trade a bunch of good players up for a great one. The AI doesn’t do this, and it doesn’t in any way obstruct you from doing this. In fact, the AI seems to do no real long-term planning at all. It takes its strengths and weaknesses into account on a trade-by-trade basis, but it doesn’t consider what’s available on the free agent market, it doesn’t ever put its best players off-limits, and it doesn’t seem to care much about contract value (although the AI GMs say “I won’t accept trades for vastly overpaid veterans,” they always do in the end; you just have to give up a little more). Individual trades can be frustrating, but if you're persistent, you can eventually concentrate a preponderance of the league's talent on your team

There was an eleven-year stretch in my league where every single Cy Young award and ten MVPs went to my players (five different guys won the Cy Young, four different guys won the MVP). I had four future of Hall of Famers in my rotation—there was only one other pitcher in the game at the time who was on their level, and I just didn't need him. I could have gotten him if I'd wanted to. That kind of dominance—and that kind of stranglehold on the best players in the game—should not be possible. Part of the problem is that it's too easy to sign all but the greediest players to favorable extensions (another topic for another time), but the trading system, and the fact that the AI never commits to its would-be franchise players, is the heart of the matter.

[Orcin]

Originally Posted by harrumph

Your fundamental goal in trading should be to concentrate value: trade a bunch of average players up for a single good one, then trade a bunch of good players up for a great one. The AI doesn’t do this, and it doesn’t in any way obstruct you from doing this. In fact, the AI seems to do no real long-term planning at all. It takes its strengths and weaknesses into account on a trade-by-trade basis, but it doesn’t consider what’s available on the free agent market, it doesn’t ever put its best players off-limits, and it doesn’t seem to care much about contract value (although the AI GMs say “I won’t accept trades for vastly overpaid veterans,” they always do in the end; you just have to give up a little more). Individual trades can be frustrating, but if you're persistent, you can eventually concentrate a preponderance of the league's talent on your team.

That's a damn good first post!!! The quoted part above is spot on, and the heart of the matter in my opinion. The AI falls for a quantity-for-quality offer every time.

[Honorable_Pawn]

Great first post!!! Welcome aboard! Hope to see a lot more of you.

[harrumph]

Hey, thanks! Hopefully I've gotten the extreme long-windedness out of my system at least for a couple days.

[DCG12]

I don't mean any offense but I have to disagree with Orcin And HP. There are some things that I question here.

First off all these trades are a minimum of 5 for 1 type trades and the trade that you describe as your biggest AI ripoff was a 4 for 1. Do a search of the forums and you will see that this type of trading has been discussed before and most veteran players use some sort of house rule to limit trades to 1-1 or 2-1. Everyone knows that the game uses some sort of balance scale (maybe WAR for example) where if you keep adding players, even junk players, you will eventually tip the scale. I agree that maybe there needs to be some sort of refining of the code to penalize these types of trades in the eyes of the AI.

Now the things I wonder about. You say you're playing on hard trading with low scout accuracy and a quasi stats only yet you took a 78 win team and improved by 30 wins in your first season. I apologize for not being impressed but there is some gaming of the AI going on here somewhere if this is true. I have been playing this game religiously since 2003 and have seen too much go wrong on the best of teams to buy into this. Just last season in my fictional league I had a team that the pre-season predictions said would win the league only to have my whole team slump out of the gate with an 11 game losing streak which we never recovered from. Just wondering...

Edit: upon further thinking I remembered that the game limits trades to 5 for 1 so I went back and re-read the trade that I labelled 7-1 and I had misread it, it was actually 4-1(change made in original post). My apologies. Same point still applies though as these multi-players deals are an issue.

[mking55]

Just a thought and I occasionally do have them.

For many years the trading was terrible. We treated the AI like an idiot and consistently robbed it.

Maybe it's "revenge of the AI"? The AI in trading is good now (never perfect, never can be) and maybe the AI has decided to treat the human player as an idiot. After all, turnabout is fair play.

When I play I set all trading to extremely difficult and over 90% of the time I have to "lose a trade" to make a trade. That's good. It evens the game. And if the AI can rip me off, more power to it.

Power to the AI!

[harrumph]

Originally Posted by DCG12

I don't mean any offense but I have to disagree with Orcin And HP. There are some things that I question here.

First off all these trades are a minimum of 5 for 1 type trades and the trade that you describe as your biggest AI ripoff was a 4 for 1. Do a search of the forums and you will see that this type of trading has been discussed before and most veteran players use some sort of house rule to limit trades to 1-1 or 2-1. Everyone knows that the game uses some sort of balance scale (maybe WAR for example) where if you keep adding players, even junk players, you will eventually tip the scale. I agree that maybe there needs to be some sort of refining of the code to penalize these types of trades in the eyes of the AI.

Now the things I wonder about. You say you're playing on hard trading with low scout accuracy and a quasi stats only yet you took a 78 win team and improved by 30 wins in your first season. I apologize for not being impressed but there is some gaming of the AI going on here somewhere if this is true. I have been playing this game religiously since 2003 and have seen too much go wrong on the best of teams to buy into this. Just last season in my fictional league I had a team that the pre-season predictions said would win the league only to have my whole team slump out of the gate with an 11 game losing streak which we never recovered from. Just wondering...

Edit: upon further thinking I remembered that the game limits trades to 5 for 1 so I went back and re-read the trade that I labelled 7-1 and I had misread it, it was actually 4-1(change made in original post). My apologies. Same point still applies though as these multi-players deals are an issue.

No offense taken!

I can take some screenshots of the transactions if you want to have a look—I don't entirely understand how I pulled it off myself. It was a complete overhaul: fired most of the coaching staff, signed two of the team's best young players to extensions and dealt almost everybody else (the ability to get rid of almost any player, no matter the contract, by trading is another of the big flaws with the system, although in this case there were no real albatross contracts), picked up a bunch of low-cost free agents, and, yes, made a lot of five-for-one trades.

The staff ERA dropped, in one year, from 4.38 to 2.83 (league ERA went from 4.12 to 4.08), so that's obviously the big difference. It was 2.83 again the next year, then over 3.00 six years in a row—so a bit of luck and a bit of BABIP was involved. I replaced the entire bullpen, trading away or releasing the old guys and bringing in new ones mostly as free agents. I rebuilt the rotation with trades, though, and, mea culpa, mostly five-for-X trades.

Seven guys started games that year; only one (in 23 starts) posted an ERA above 3.03. So that, to an extent, is just lucky. Five of those seven guys were new to the team, including the three most productive. Hotmar was a star already and I got him in a five-for-two deal for a good prospect, two middling prospects, and two of the previous year's starters. Kal was a two-for-two trade for an aging All-Star center fielder and a replacement-level first-baseman. Graete had been a below-average pitcher up until this point in his career, but yes, this was another five-for-two, this time for an outfielder, a reliever, a good catcher, and two middling prospects. Those three pitchers combined for 19.2 WAR and didn’t get injured all season. It was a career year for one guy and a top-three year for each of the other two. There's a strong element of luck here.

The other two new starters were also five-for-X trades. Plessis was an excellent reliever whom I converted into a starter midseason; I got him for another aging All-Star (the previous year's starting third baseman), a couple good prospects, and some filler. Graves was a veteran; I got him and a backup catcher for basically half the previous year’s pitching staff—one starter and four relievers.

I got 29.2 WAR from those seven starters (two of whom also pitched in the bullpen and two of whom spent most of the season injured) and 5.8 from the bullpen. The previous year's staff had been worth 17.4 WAR. The hitters also did 11.6 WAR better, although almost half of that was a breakout season by one of the young guys they already had (1.1 WAR one year, 6.8 the next). That's not a 30-WAR improvement, but it's close, and I don't think WAR is supposed to convert directly to wins on the field anyway.

So two of my three best pitchers (and, full disclosure, three of my five best hitters) came aboard via five-for-X trades. I guess that qualifies as gaming the AI by your standards, but should it? Taken individually, these look like realistic trades. Maybe not the Graete trade—that's the one of these where I just sort of piled up organizational depth until they said OK—but the rest are the sort of trades that actually happen in MLB: two decent pitchers and some prospects for a better pitcher; an aging All-Star and two prospects for an excellent reliever. I managed all of this with one of the smallest budgets in the league (half what my previous team had), and it's not like these guys were all pre-arbitration; I got a whopping six 2+ WAR performances from guys under 30.

It's not like taking dubious five-for-X trades is the only thing the AI does wrong in trading. It's not an isolated bug, it's a symptom of the AI not doing any long-term planning. The AI just plugs the players it has into the places it thinks are best for them right now; in trades, it simply assigns each player a value and does some adding and subtracting. What it should do is create goals related to players.

Say an AI GM has two outstanding young pitchers, both potential Hall-of-Famers (see Daikawa & Blunsden above). It should, first and foremost, put them off-limits to trades. Then it should try to nail them down to the best, longest contracts possible (I almost never see the AI sign young players to contracts longer than three or four years before free agency). Only in the event that they're unhappy, refuse to sign, and are approaching free agency or getting prohibitively expensive in arbitration should it consider trading them (or, more to the point, allowing the player to trade for them).

Next, it should set some priorities related to them: it should favor defenders who play to their tendencies (assuming they have complementary tendencies). It should perhaps emphasize defense and pitch-framing over offense when looking at catchers. It should prioritize arm and fatigue skills for team trainers, and make sure to sign a good pitching coach who handles young players well. It should strenuously avoid taking on big contracts, via trade or free agency, that might jeopardize its ability to sign these two to long deals. This is a huge problem right now—again, it's not like you often see an individual trade and say "wow, that was stupid" but you can make one trade after another such that, without even really meaning to, you hobble AI teams with lots of big, bad contracts (they also take on some terrible contracts voluntarily in free agency, of course).

In another example, imagine an AI GM has a really talented young catcher who hasn't reached his ceiling yet—and nobody else at the position. Again, they just shouldn't be willing to trade the guy. Period. Then they should be looking to sign a veteran catcher to a fairly short deal; ideally, he'd have good leadership qualities, the opposite handedness to their prospect, and the ability to move to first when the prospect was fully developed, so they'd give some priority to each of those things, especially if they had plenty of options. They’d also avoid signing a bat-only first baseman to a long deal unless they had room at DH. They should absolutely not be interested in trades for young catchers, catchers with long contracts, etc. The AI should have a little map of what the organization should look like, maybe a five-year-plan, a ten-year-plan: this guy is going to be playing there, these guys are going to be in the rotation, we should add depth here, we should sign a veteran there to bridge that position until our top prospect is ready.

I appreciate that programming this sort of thing fully must be incredibly hard, but it seems like there are a few simple first steps that would make a world of difference. For one thing, the AI should just put its best young players off-limits sometimes. Completely off-limits, no exceptions. It should make a serious effort to sign them to long deals, clearing veteran payroll to do so if necessary, and it should just not trade them. That would make a huge difference all by itself. Teams should also be strictly unwilling to take trades for guys 30 and over who are signed to big, long deals—unless, and only unless, the player serves an immediate need (i.e. is actually the target of the trade). They shouldn't take guys they didn't want in free agency. They shouldn't really ever take non-prospect who aren't going to play unless they're high-value pieces they believe they can move in trades of their own.

It's true that you can impose house rules on yourself. Don't ever put more than three guys in a trade. If you gamble on a long contract and it doesn't pan out, suck it up and call it a sunk cost. But it feels weird to do that—in real life, teams do make those trades. If you keep asking around, you will eventually find somebody who's willing to take your grossly overpaid veteran for the right price, or your "can't-miss" prospects for their established superstar. I'm a Tigers fan (oh, woe is me right now), so I immediately think of the Cabrera/Willis for Maybin/etc. deal and the more recent Kinsler-Fielder deal. Those are definitely both trades that you'd call fishy in OOTP, especially in retrospect. Is Dave Dombrowski gaming the AI? (Signs point to yes.)

As a last note, the AI actually does make these deals too—a fourth-place team in my league just traded its star first baseman (our man Urbanus Gogele, if anybody read all my analysis above) for two good prospects and two middling ones. Again, exactly the sort of thing you'd see in real life, but probably not a great move for the seller.

[SirMichaelJordan]

Originally Posted by DCG12

I don't mean any offense but I have to disagree with Orcin And HP. There are some things that I question here.

First off all these trades are a minimum of 5 for 1 type trades and the trade that you describe as your biggest AI ripoff was a 4 for 1. Do a search of the forums and you will see that this type of trading has been discussed before and most veteran players use some sort of house rule to limit trades to 1-1 or 2-1. Everyone knows that the game uses some sort of balance scale (maybe WAR for example) where if you keep adding players, even junk players, you will eventually tip the scale. I agree that maybe there needs to be some sort of refining of the code to penalize these types of trades in the eyes of the AI.

Now the things I wonder about. You say you're playing on hard trading with low scout accuracy and a quasi stats only yet you took a 78 win team and improved by 30 wins in your first season. I apologize for not being impressed but there is some gaming of the AI going on here somewhere if this is true. I have been playing this game religiously since 2003 and have seen too much go wrong on the best of teams to buy into this. Just last season in my fictional league I had a team that the pre-season predictions said would win the league only to have my whole team slump out of the gate with an 11 game losing streak which we never recovered from. Just wondering...

Edit: upon further thinking I remembered that the game limits trades to 5 for 1 so I went back and re-read the trade that I labelled 7-1 and I had misread it, it was actually 4-1(change made in original post). My apologies. Same point still applies though as these multi-players deals are an issue.

Yup main reason I don't go lower than very hard is because of the multi player trade issue. It is better than previous years (atleast on very hard) and part of that was limiting to offering 5 players.

As for the game being too easy Idk, I've only won 3 WS in 20 season between OOTP 13 through OOTP 15 with the Phillies, Mets, Dodgers and been fired by The Yanks, Dodgers and recently Mia. Two of my championships came while being a wild card team.

I do offer 5-1, 4-1 trades and they usually involve one all star and a bunch of minor leaguers and may or may not involve a top 100 prospect and that is no matter what side of the trade I am on.

I do play with high injuries so I know that adds to the difficulty level. (And TCR @ 200)

Oh and I should mention trading preference is set to heavily favor prospects which is probably why I can't offer ton of min salary vets for a top prospect or a young all star.

I don't find it easy at all to steal away a young player from a team even if the player I am giving them is an established all star. I always get in reply "You are kidding right?"

So I guess having trading preference set to neutral treats everyone the same if they are close in ratings or if you give them a few players that are rated high enough. Neutral is probably why it seem as if the AI is not planning for the future because it doesn't have much of a preference.

Beside that one thing I would like to see is small market teams signing their young prospects to long term deals as early as possible. I would like to see the AI buy out the arb years.

[Orcin]

Originally Posted by SirMichaelJordan

Oh and I should mention trading preference is set to heavily favor prospects which is probably why I can't offer ton of min salary vets for a top prospect or a young all star.

I don't find it easy at all to steal away a young player from a team even if the player I am giving them is an established all star. I always get in reply "You are kidding right?"

So I guess having trading preference set to neutral treats everyone the same if they are close in ratings or if you give them a few players that are rated high enough. Neutral is probably why it seem as if the AI is not planning for the future because it doesn't have much of a preference.

I will try a historical league with the settings at Very Hard and Heavily Favor Prospects. What about using AI evaluation vs. pure ratings... which setting makes the AI more competitive? If AI evaluation, how do you set the ratios? I would guess that AI evaluation with a heavy emphasis on ratings would yield the best results, but that's just a guess. I have tried a few sample trades and it doesn't seem to make much difference.