Seeing some good things in OOTP6 FA Acquistion AI.
 

This is from my work-in-progress free agency report in my latest season of my OOTP 6.03 Dynasty:





A few things that stand out...

• "Murphy, my former closer, signed with Anaheim on Day 8 for $9.5M, leaving Albacete as the only decent closer out there, and it incited a bidding war."

• I've seen this more than once. When there's one starter-quality player and a bunch of scrubs left in free agency, *somebody* overpays for him. I'm finding that if I don't build through prospects, I have to overpay to fill slots via free agency.
  
  
  
  
• Chad Daulton's salary. I used to be able to sign a guy like him a lot cheaper than $3.25M. That's what it took to land him against the AI, though.
  
  
  
  
• "I felt like I had to spend to get a quality closer, and therefore I couldn't afford a starting CF like I wanted." That pretty much is self-explanatory. The aggressive AI forced me to make a decision.
  
  
  
  
• "Good to see guys like this being recognized by the AI for their value. He only has 163 major league at-bats over the last two years, but still commanded a decent salary on the open market. (He was only asking for around $450K.)"--Again, self-explanatory.

I'm pretty impressed right now.

I had to pay 16 million for 4 years for a closer. FA is tough

I'm also finding FA to be much improved. Last night, I went after a Star SP, and couldn't get him because the Yankees offered too much money. So, I went after the next best starter, but with the best SP gone, a bidding war suddenly ensued for the 2nd best SP. I had made a bid for him early in the process, but he decided to wait for what the market would bear. As it was he made a smart decision, I had to pay 8 million a year for a fairly average starter, when his initial asking price was only 3 million. In this case, it was a necessity as I didn't get enough decent starting pitchers in the initial fantasy draft.

I've been much more impressed with the trade AI as well, using low frequency, hard trading, and favor prospects. It's suitably difficult to trade my way to a contender now. I just hope my minor leaguers develop before I'm out of a job ;)

This, to me, is the coolest thing that I am seeing with the AI. Further, it sounds like we're seeing it enough to be sure that it isn't just coincidence, but a definite pattern. :)

Are you seeing bidding wars over players that your team isn't pursuing?

(I seem to recall some people claiming that the AI teams were just manufacturing bids for the top players that the human team was going after... but I haven't even played the game yet to have my own opinion)

I'll need to check that. I know I've seen some FA's getting signed for more than what I think they're worth, but I'll look more carefully in the next FA period.

Just a quick look at the signings log gives some anecdotal evidence that there are AI bidding wars as well. There was only one 5-star SP in my most recent free agency stage. I didn't bid on him at all, but this 39-year-old went for almost $19M for one year. He was good, but not THAT good. ;) Here are his ratings:

Pitching Stuff Movement Control Velocity Overall 52 86 73 88-91 Mph Versus LHB 51 86 66 Versus RHB 52 86 79 Talent 56 94 66

...and his last three seasons... 2020 36 36 14 5 0 3.15 208.2 199 82 73 40 132 0 0 1.15 SD 2021 32 32 12 7 0 3.08 169.2 175 69 58 46 140 0 0 1.30 SD 2022 35 34 18 6 0 2.92 200.2 181 70 65 50 154 0 0 1.15 SD

patch 6.02a in the readme it says "Improved free-agent signing AI. It's now more aggressive when having enough money available."


i glad to see that working out well..

I've seen instances of AI vs. AI bidding wars. I'm always curious where my free agents go and how much they sign for. Just Recently, I had a 3rd baseman that started free agency by requesting $4.2 million per year. Too much for me at the time as I had a young 3rd baseman about a year away from playing every day. Anyway, my free agent 3rd baseman ended up getting over $6 million per year from New York.

I'm looking to pick up OOTP, but as a single player game only. I haven't played since OOTP3. Is this game worth the pickup as is right now as a single player game?

The AI still has a little ways to go imo.

I trade a 5 star OF and a 5 star 1B to Las Vegas, they waive their 3 star 1B, who i pickup and then 3 days later they waive the 5 star 1B I traded too them. 3 weeks later, the 5 star 1B is still sitting on the FA list. blech.

Are you using the suggested evaluation pattern (of players) from other threads? I'm using the compromise of 50/10/30/10 and I'm not seeing this insane waving of players that you describe.

Very nice work. I like your 'what's wrong with the Ai???' thread at ootpd.

I'm using the 50/10/30/10 compromise as well, and seeing good results. The best performance I saw last season by a waived player was a guy that I waived.

Same here, the waiver wire in my league is annoyingly devoid of any real talent.

If you don't mind my asking, which numbers correspond to which categories?

eg ratings, 2004 stats, 2003 stats etc....

50-->Ratings
10-->Current Year Stats
30-->Last Year Stats
10-->2 Years Ago Stats

thanx sky

very cool, ill give it a shot. thanks guys!

Great, that means you're not using the numbers posted above then, which is a relief to me. :)

I honestly think 6.03 is a gem. Thanks goes out to Markus and the beta team! I'm still seeing great things from this game as far as solo play goes. I knew taking over the Brewers was going to be rough, but damn. :D


Todd

just picked up the game, i got the:

50-->Ratings
10-->Current Year Stats
30-->Last Year Stats
10-->2 Years Ago Stats

Some more Q's for a competive game. What should I set the AI to favor in trades? Also what level should it be set to. Right now I set it to Hard and Favor Prospects. Does very hard make it unrealistically hard?

Hard and Favor Prospects is what I use, and it seems about right. I also set it to Low number of trades.

FWIW, my next little mini-campaign will be to get Markus to make 50/10/30/10, Hard, Favor Prospects & $25M-ish cash to be the default settings of the game. The thing needs to work "out of the box." Let people make adjustments if they want to do something UN-realistic.

Unrealistic? You mean like simulating the way baseball was in the 1910s or the 1960s? :)

Markus has clearly made a commitment to making the default settings work for modern baseball. Deal with it. Your comments are superfluous with regard to the default settings.

That's fine, modern baseball should be the default. But you strongly implied that not playing modern baseball is UN-realistic. Out of the ordinary, yes, but not out of the realm of realism.

No, I didn't imply that. You must have read that into it. I wasn't even thinking about not playing modern baseball.

since i do not want to start my own thread i will post my complaints here:

The drop down menus are nearly impossible to read. Anyway to change the color structure? I tried the skins that come with the game and none of them allowed me to read them any better.

When you try to offer trades and click on a position it writes over top of the whole roster making it impossible to read, also the in game commentary is nearly impossible to read b/c it just writes over top of things that are already there.

I'm seeing great results with the 50/10/30/10, Hard, Favor Prospects, Low Frequence settings. As far as the cash, I think I left it on what you had in your league file that I downloaded which was 5 million dollars? What seems to be the general consensus on max cash setting?

There's an issue on some machines. I think you have to disable Menu Animation in the "Game Setup" screen and it corrects this problem.

tried that and it still just writes overtop of what is already there. Also any way to change the color of the drop down menu from black to something else?

Did you disable Direct3d? Ive read numerous times that this cures your exact problem. Start->programs->Out of the Park Developments->OOTP6->Game Configuration

disable direct3d here.

thank you VERY much

trade= low, favor prospect, diffcult on hard...
and 25 mil cash max gives great result....very Realistic



i sure hope u can convince Markus to make it the default for ootp 6

Ok, I need to make some quick changes to my max cash ;)

Do you all think, that having it at 5 million for 2 seasons ruined this career, or will things adjust reasonably with time by switching it now at the end of my 2nd season?

what about 50mil max cash? - that's what I use

also, do you have your lineups favoring speed? I think that has helped as well, at least imo.

Also,

Troll, use the ESPN skin - I've found that one much easier on the eyes.


You can get it at baseballsimcentral, I think.


Todd

skydog been going with 25 mil cash max..and he said it looking good for patch 6.03

i like favor speed....but favor speed or favor OBP i think are both good options to choose in the edit era...pick one and stick with it...


skydog for steals do you recommend normal? you getting good stats with normal?

thanks

Does it really matter now that you know it works what the default is?

1. I personally don't want to have to change the settings to make the game work every time I start a new career.
2. I would think for Markus and OOTPDev's sake, they'd want the game to work optimally "out of the box", so that those who don't read the boards to tweak their settings will get solid results.

Favoring Speed has been working great, imo. I do however, like 50mil cash max. When my season is over I'll post what I think about it. Being Milwaukee, $25mil would take some of my cash and give it to other teams. :)

Todd

I just saw a $33M per year FA signing with it set at $25M. :eek:

Yow! Who is dat dude?


Todd

http://benelou.com/ootp6/p3067.html

Perhaps this has been brought up before, but I just saw something that chapped my tender buttocks.

Last offseason I was bidding on a FA starter (I desperately needed one), fairly middle-of-the-road type but the best starter available. The AI kept bidding above me, to the point where this pitcher who should be earning maybe $5-7 million was offered $17million (3 yrs). Once the salary offers got into that nutty range I bailed on the bidding process, but then the FA period was over I went to check on how much he wound up getting: 4 yrs at $5.5 million!

That, my friends, is crap.

Ouch. I've never noticed that during my various bidding wars. That would greatly upset me.

Has anyone else seen this?

Ben,

Regarding the tweaking that needs to be done in order for OOTP to "play right", specifically the way the AI addresses it's decisions according to ratings vs. past year's statistical database. When I originally talked to Markus about this, he did put in a piece that automatically made OOTP treat small sample sets with less value. Therefore, despite giving 50% of the relationship to this years stats, if a guy has only 100 atbats, it would therefore minimize this 50% in relation to the entire equation. Immediately upon release, I saw a lot of people recommend tweaking those values based on the month of play - though I never saw value in this due to that fact. Mind you, perhaps Markus tweaked this number, or it's not working properly, but the need to change the AI based on the current month is not goal of this process.

I'd be interested in hearing from people who specifically saw issues (granted this would be a tought nut to crack) because they did not tweak these numbers based on the general practices of the userbase.

-Craig

Craig:

If you want to look around my dynasty league at www.benlou.com/ootp6 you can find a pretty significant difference in waiver pickups/cuts before 2020 (when I quicksimmed and used the default settings) as opposed to 2021-2024. I'm heading out the door right now, or I'd post some examples myself. If you don't have time to look over it, I'll do it later tonight. One guy comes to mind immediately, my best player, James Culbreath (http://benelou.com/ootp6/p2598.html). Notice that he was waived in 2020. I'm not seeing these sort of moves with *nearly* as much frequency since I changed the settings. The reason I (begrudgingly) changed was that I was still seeing this stuff after 6.03. I'm out, but I'm pretty sure the evidence is there in that league.

Dola:

I did take a quick look. Check out the National League (L2) batting average leaders for the current season (2024). Seems like nearly half of them were released or waived prior to 2020, and usually it was early in the season.

Your opinion of optimally is different then mine, which is different then someone else's on down the line. If you know how it works well for you, is it a big deal to change it to hard/favor prospects? It doesn't even take 10 seconds.