Thoughts on trade logic

[ExpansionManager]

July trades usually end up making sense, with young players leaving contending teams, who get veterans in return, but sometimes the "rentals" don't make sense at all, guys having bad years, or guys who haven't played in the majors all year or sometimes even last year. My evaluation settings are 10 ratings/55 current year/30 last year/5 two years ago, trade difficulty 2 away from hardest and 2 ticks away from completely favor prospects(do those affect the computer teams at all?).

Also there's almost no trading before July, or in spring training, my frequency is set to average. How crazy does it get if I tick it up to high? Also are my AI eval settings right, or should I more remove some weight from ratings?

[omg_pwnasaurus]

I generally think it's pretty good, although I still see the weird stuff in July like the CPU team that's getting the desirable rental player also getting a really good young prospect in the deal like everytime (often times better than the prospect package they send out) instead of just getting the rental player for a prospect or prospect package which never makes sense to me.

Also the CPU to CPU trade always gets better trade return than a the human player trying to make a deal imo. Which is ok since I prefer the challenge, but it is still kind of odd. (Default sliders)

[Sweed]

I'd agree with omg_pwnasaurus, ie pretty good overall. I play slowly managing every inning of every game. Every day in my season I read the complete transaction report and review all trades. I see some "huh?" trades but most an argument can be made for it being viable for both teams. This is a broad stroke statement and not meant to say the game's trade AI is perfect. Exceptions can and always will be found. It's a computer game.

Also there's almost no trading before July, or in spring training, my frequency is set to average. How crazy does it get if I tick it up to high? Also are my AI eval settings right, or should I more remove some weight from ratings?

As I said I play slowly so not a lot of v26 data to look at. With that in mind I'll post what I am seeing and what I saw in v25. This is a game that I import from version to version, so the example is from the same save in v25 and v26.

v26: Offseason trades, Nov. to end of ST: 39. Not sure how it compares to real life but for OOTP I think one wants to look at offseason and not just ST.

v25: Offseason 43 trades.
Opening day to trade deadline, 56 trades. 50 of those trades made in July. That is 99 trades from start of offseason to the trade deadline in v25.

Comparing v25's 43 in the offseason to v26's 36 offseason trades would suggest that v26 trades are running at about the same rate in both versions of the game.

This is average frequency and default settings that lean towards hard and favor prospects. Not sure how many clicks that is but looks about 75% towards hard and 60% towards prospect on the setting line.

Evaluation? I use the 4-25's. IE 25/25/25/25. If you do a search you'll find the evaluation topic gets discussed in depth almost every year. I think this year is the first I haven't seen a thread about it.

Here's a link to last years discussion. https://forums.ootpdevelopments.com/...ght=evaluation
In that thread there are links to previous year's threads on the topic. At the end of the day there are valid arguments for why individual users like the setting that they use. There is no wrong answer.

Afraid I can't help on how changing evaluation settings might impact your trading in either quality or amount. I've been using the 4-25's for a few versions now, I'm happy with the results, so have no experience in how making changes may or may not impact trading. Hopefully someone comes along that has tested this and can make some relevant comments.

[ExpansionManager]

Thanks for the in depth reply, I'm curious about those 56 in season trades, how many of them occured before July 1st? My current 2031 season(started in ootp 25, converted at tail end of 2030 after world series), saw 3 trades between opening day and the end of June(2 by me, and a minor league shuffle where the Expos send 2 rookie/A ball pitchers of 2 star potential to the Braves for a 31 y/o AAAA first baseman), then 56 in July, 9 on the 31st. I don't have time right now to compare that to real life data, and it does compute that most trading occurs in July, but I still feel like there should be a little more early activity based on spring training position battles and early season ending injuries.

[Sweed]

Originally Posted by ExpansionManager

Thanks for the in depth reply, I'm curious about those 56 in season trades, how many of them occured before July 1st? My current 2031 season(started in ootp 25, converted at tail end of 2030 after world series), saw 3 trades between opening day and the end of June(2 by me, and a minor league shuffle where the Expos send 2 rookie/A ball pitchers of 2 star potential to the Braves for a 31 y/o AAAA first baseman), then 56 in July, 9 on the 31st. I don't have time right now to compare that to real life data, and it does compute that most trading occurs in July, but I still feel like there should be a little more early activity based on spring training position battles and early season ending injuries.

Six before July 1, and fifty were made in July. It was heavier towards the end of the month, but don't have the game open to find exact numbers.

I think the AI does a lot of it's roster sorting before ST arrives leaving fewer deals "needing" to be made then. FWIW I have done no research into the timing of offseason/ST trades in real life to have an opinion on how OOTP compares . Could be OOTP is sorting rosters before real life teams do? Could be it's close to real life? Could be it should be doing it later in the cycle? IDK.

Could the early season "trades due to injuries" be done better? Like any type of trading in OOTP I'm sure it could.