old OOTPer checking on current quality, with specific questions

[obaslg]

Thanks in advance for reading and especially for any answers.

I'm a player from way back - maybe 2000-2007? It was always great, but I did get frustrated when there seemed to be no effort to fix things under the hood. I'm thinking about picking it up again, and I'm hoping some people can tell me where these issues are in the current version.

1) Fictional draftee generation problems. There were major problems in the quality of new players, such that 20 years into the future there would be no superstars. More generally, it seemed to me that the problem was that the game created players on a bell curve of quality, but MLB quality doesn't look like that.

2) Lineup sub problems when simming several days at a time. Long story short, it was just dumb, and therefore frustrating.

3) Baserunning. Most noticeably, scoring from second on a single always seemed extremely low. IIRC, the baserunning play-calling was also limited, e.g. no double steal with two outs.

Thanks!

[koohead]

I think the draft pool is fine. I draft 40 rounds, generating players for 41 rounds. It varies year to year but there is plenty of high end prospects in the first few rounds and with Talent Change Randomness set to 100 (Average) I see my share of mid-late rounders turning into everyday players and even some stars.

I think the day to day lineup subbing is pretty good. I sim all my games and set backup players to start every 7 or 10 games usually and see they get their share of starts and I never see starters exhausted.
Can't speak to in-game subs though...never looked that closely.

[The_Offspring187]

I have been playing off and on since Season Ticket Baseball (OOTP3 I think). When you last played, were feeder leagues a thing? That could help you see better drafts if that's something you never did before.

Sent from my LGLS755 using Tapatalk

[Mat]

MLB players cant be rated on a bell curve?

[obaslg]

Originally Posted by The_Offspring187

I have been playing off and on since Season Ticket Baseball (OOTP3 I think). When you last played, were feeder leagues a thing? That could help you see better drafts if that's something you never did before.

Feeder leagues as in visible minor leagues you can manage? - yes to that.

[obaslg]

Originally Posted by Mat

MLB players cant be rated on a bell curve?

Right - it's just a downward slope (or maybe an inverted curve - don't remember) - more bad players than good players, and more good players than great players. Bill James listed that as one of his top 10 findings. It's old research, but it should still be the same.

It makes perfect sense if you think about it. If the general population is on a bell curve for reflexes, agility, etc., professional players, and MLers in particular, already represent the very far right of that curve. Everyone bad or even good never played seriously or was weeded out before the amateur draft. You have to be spectacular as compared to the general population even to be drafted. If you look at the far right side of a bell curve, it's a curved slope downward, so we'd expect to see that, and we do. (At least Bill James said we do, and that fits with my general impression.)

[NoOne]

yeah, it's not supposed to be a standard distribution. a pyramid is a better description than a bell. very few at the top, many at the bottom

1 - they definitely stay consistent, within reason. we want some fluctuation, of course. with static ltm you will get a range of results that you can count on. that means the talent is remaining consistent to get those results year after year. i had a 100year league in ootp18 that used static LTM/LTs and it did not tailspin, nor did it escalate out of control. (stats and ai settings - unchanging totals and modifiers)

caveat: seed players and RL mlb players are not the same distribution and make-up. so, adjusting ltm while transitioning from such players will be required to ~control results during this time. if fictional, i just pre-date league and zoom out 30 years before erasing history etc, and starting on intended date wiht a nice robust player base built from drafts and normal game processes.

2 - meh... there are custom lineups, 7-day lineups and depth charts... the depth charts still have problems. it still does what it wants. e.g. i have a catcher in ootp18 i put to "never" on backup... it subs him at "97%" still.

it's probably better than it was, and if you go 7 days at a time you can get exactly what you want.... e.g. i think it bumps the subbed catcher down to 9 instead of just placing in lineup at same slot... 95% sure that is 'fixed'

3 - i really don't play out games much anymore. i think if you found a screenshot of it and looked at the available options, you'd recognize what's new. on ootp's website there's media to look at, i'm certain one is an in-game preview pic... or google etc. i bet the defensive positioning options are new for shifts.

[fuzzy_patters]

Originally Posted by obaslg

Feeder leagues as in visible minor leagues you can manage? - yes to that.

Feeder leagues are high school or college leagues that you can create so that the players from those leagues will graduate into your draft pool. They are separate from your minor leagues.

[BIG17EASY]

IMO, there was never a problem with fictional draftee generation. It was just that the first 3-4 draft classes with real-life college and high school prospects were a bit overrated (intentionally) for whatever reason. (Lukas explained the reason a couple years ago, but I forget the exact wording, so I don't want to speak for the devs.)

Once those draft classes passed by, users had a fit because the draft wasn't full of 4- and 5-star prospects, even though that's not realistic. The game still produced (and still produces now) plenty of draftees who turn into superstars. You just have to do a little more work to try to find them in the draft and there's more variance in how early round draftees turn out, which is similar to real life.

[obaslg]

Originally Posted by NoOne

1 - they definitely stay consistent, within reason. we want some fluctuation, of course. with static ltm you will get a range of results that you can count on. that means the talent is remaining consistent to get those results year after year. i had a 100year league in ootp18 that used static LTM/LTs and it did not tailspin, nor did it escalate out of control. (stats and ai settings - unchanging totals and modifiers)

caveat: seed players and RL mlb players are not the same distribution and make-up. so, adjusting ltm while transitioning from such players will be required to ~control results during this time. if fictional, i just pre-date league and zoom out 30 years before erasing history etc, and starting on intended date wiht a nice robust player base built from drafts and normal game processes.

That sounds promising (and thanks for the other detailed responses). Is there a guide somewhere about how to do that, or should I just bug you again once I get started? ( :

[battists]

Originally Posted by BIG17EASY

IMO, there was never a problem with fictional draftee generation. It was just that the first 3-4 draft classes with real-life college and high school prospects were a bit overrated (intentionally) for whatever reason. (Lukas explained the reason a couple years ago, but I forget the exact wording, so I don't want to speak for the devs.)

IIRC the reason was simply that the current minor leaguers / top prospects were hand-rated by the roster team, whereas future draft classes were purely AI generated, and the hand-rated players tended towards overall higher ceilings.

[Lukas Berger]

Originally Posted by BIG17EASY

IMO, there was never a problem with fictional draftee generation. It was just that the first 3-4 draft classes with real-life college and high school prospects were a bit overrated (intentionally) for whatever reason. (Lukas explained the reason a couple years ago, but I forget the exact wording, so I don't want to speak for the devs.)

Once those draft classes passed by, users had a fit because the draft wasn't full of 4- and 5-star prospects, even though that's not realistic. The game still produced (and still produces now) plenty of draftees who turn into superstars. You just have to do a little more work to try to find them in the draft and there's more variance in how early round draftees turn out, which is similar to real life.

This isn't really an issue now though, because we've significantly lowered the hand-rated guys now, so they should be much more in line with the guys the game generates. Maybe not perfectly so, but it's not so far off as it was at one time.

[obaslg]

Originally Posted by battists

IIRC the reason was simply that the current minor leaguers / top prospects were hand-rated by the roster team, whereas future draft classes were purely AI generated, and the hand-rated players tended towards overall higher ceilings.

This is way back, so I couldn't tell you which version, but I remember someone running a bunch of years and reporting that the top generated players ended up well short of star quality - something like .350/.450 - so the issue was definitely with the randomly generated players. But it sounds like it's better now.

[BIG17EASY]

Originally Posted by Lukas Berger

This isn't really an issue now though, because we've significantly lowered the hand-rated guys now, so they should be much more in line with the guys the game generates. Maybe not perfectly so, but it's not so far off as it was at one time.

Good to hear. I haven't played the MLB quickstart in 19 yet that uses the hand-rated draft classes, so I probably spoke out of turn in the above post.

[BIG17EASY]

Originally Posted by obaslg

This is way back, so I couldn't tell you which version, but I remember someone running a bunch of years and reporting that the top generated players ended up well short of star quality - something like .350/.450 - so the issue was definitely with the randomly generated players. But it sounds like it's better now.

I remember the thread you're talking about. But I never experienced that in my long-running world that started with the quickstart in OOTP15. Maybe because I always made some minor settings adjustments and I think the person who ran that test was using the standard settings in the quickstart.