Realism question

[jpblitz]

Has anyone found a way to make a historical league represent realistic stats without having to do a replay?

Every time I run through a league the individual careers are too long on most players, the offensive power is too great from era to era, and the pitching stats are weakened as a result. Ending up with 100+ players with 500+ homeruns after a hundred years and about 400 with 3000 hits dampens a bit of the fun.

However, when I run a replay with real transactions the stats come out close to real life.

Any suggestions would be welcomed.

Thank you.

[thehef]

I suggest providing more details as to how you set up your non-replay sims, and what you're after.

One thing you mentioned: "...the individual careers are too long on most players." You can fix this by checking the "Retire According to History" box on the Game Settings > League Settings > Historical panel... The other things you mentioned (that you do not like) are very-likley fixable, but it all depends upon the details...

Note that there are so many difference settings & toggles & so forth that OOTP is capable of:

1) a pretty solid historical replay
2) something that's so divergent from history that it becomes almost fictional
3) everything in-between 1 and 2

IOW, you can start off by selecting Replay on the Historical Wizard panel #1, but then tweak a bunch of other settings to where things will be different from a strict-ish replay. Which, again, means that what you are after is likely doable.... just need more deets.

[jpblitz]

Thanks for the reply!

My non-replay historical simulations are very simple setups. I don't tweak anything and have the recalculations mimic real life automatically year over year and era over era. Essentially, I just start up a league, have a draft, and run with it. Typically, I also have the OOTP dev engine determine player development.

It appears I may have to dig deeper into the wizard and settings.

I've been running a historical replay with real transactions and it's been outstanding. What I am looking for is a similar experience without 90% of players having 20 year careers, still in their prime from 38-41 years old.

I am also interested in running the same thing but with real players throughout history randomly appearing in different eras, so having them retire according to history would complicate that.

[Charlie Hough]

You shouldn't be seeing that many players have such long careers, so I would definitely look at your settings closely.

In addition to having players retire according to history, you can adjust some of the development engine's settings for age factors. You can have the development engine cause older players to decline more rapidly and significantly. But if you're using recalculation, that will override the game's development engine anyway. If you're not already doing so, maybe you should try turning off recalculation and just let the game engine determine players' ratings over time.

If you decide to keep using recalculation, keep in mind that, if you're playing games from the early decades of baseball, it was more common for players in their late 30s and early 40s to keep playing at a respectable or sometimes star level. So, if their real-life stats are good during those years and you're using recalculation, they're going to keep producing good numbers.

[Brad K]

Originally Posted by jpblitz

Thanks for the reply!

Typically, I also have the OOTP dev engine determine player development.

That's not what you do when you want close to historical results.

[thehef]

Originally Posted by Brad K

That's not what you do when you want close to historical results.

I don't think he's necessarily looking for "close to historical results" on a player-by-player basis. Rather, it sounds like he wants a similar number of 500+ homer hitters & 3000+ hits guys; similar league totals by era for offense & pitching numbers. And guys retiring similar to real life.

Those are things that can be done in OOTP, but there may be something screwy with some of his settings. Hard to say at this point.

[luckymann]

I have definitely noticed that tinkering with the age / dev speed settings can really extend the general length of players' careers, although it sounds to me you haven't touched them so not sure that's the answer.

But for anyone else reading this take it as a warning when bumping these up - go gently or you'll see a bunch of old fogies running around your sim!

G

[Brad K]

Originally Posted by thehef

I don't think he's necessarily looking for "close to historical results" on a player-by-player basis. Rather, it sounds like he wants a similar number of 500+ homer hitters & 3000+ hits guys; similar league totals by era for offense & pitching numbers. And guys retiring similar to real life.

Your view requires that the development engine creates a talent distribution that matches the historical distribution. I doubt this is the case.

[David Watts]

One thing you can do that might lower the number of 3000 hit guys is to use high fatigue. I would also suggest moving the injury settings to at least normal or modern realistic once teams have enough players to do so. Just my opinion, but I think OOTP allows way too many players to become what in real life would be considered ironmen. I also think the injury setting might be even more important if one is playing without recalc, because injuries can help trigger a player's decline.

[jpblitz]

Originally Posted by Charlie Hough

You shouldn't be seeing that many players have such long careers, so I would definitely look at your settings closely.

...But if you're using recalculation, that will override the game's development engine anyway. If you're not already doing so, maybe you should try turning off recalculation and just let the game engine determine players' ratings over time.

.

Good to know. It sounds like I had contradicting systems set up against each other. After simming 100 years it didn't matter which era I was in, players were lasting into their early 40s with great ratings. Hence, the ridiculously inflated numbers I got in the game. I guess I also misunderstood how the annual recalculation worked in practice.

Will play around with the settings to see if I can come up with something more realistic over a 20 year period. Thanks for your input.

[jpblitz]

Originally Posted by thehef

I don't think he's necessarily looking for "close to historical results" on a player-by-player basis. Rather, it sounds like he wants a similar number of 500+ homer hitters & 3000+ hits guys; similar league totals by era for offense & pitching numbers. And guys retiring similar to real life.

Those are things that can be done in OOTP, but there may be something screwy with some of his settings. Hard to say at this point.

Correct. I'm looking for numbers commensurate with reality and not historical reality. I can get that through a replay season, which I'm doing.

What I'm trying to do is get away from a replay season but still have a semblance of non-fictional numbers.

Are there any specific metrics and metric ranges you guys tweak to ensure you get results more to your liking?

[jpblitz]

Originally Posted by David Watts

Just my opinion, but I think OOTP allows way too many players to become what in real life would be considered ironmen. I also think the injury setting might be even more important if one is playing without recalc, because injuries can help trigger a player's decline.

This was part of my experience as well. In the beginning I would be happy players weren't getting injured. Then when looking at their career stats the theme of everyone with 3000 hits and 500 homeruns was a retirement age of 43-44 and a couple of minor injuries. These guys filled the the entire record books. Had something like 23 with 700+ homeruns and 6 with 800+

[Brad K]

Originally Posted by jpblitz

Correct. I'm looking for numbers commensurate with reality and not historical reality. I can get that through a replay season, which I'm doing.

What I'm trying to do is get away from a replay season but still have a semblance of non-fictional numbers.

Are there any specific metrics and metric ranges you guys tweak to ensure you get results more to your liking?

You want the same number of players to hit 500 home runs as historical but you want them to be different than the players who did it historically. So it looks like development only is the way to do that except your results show that it isn't. That's because you can't count on development giving the same distribution of talent as historical. Actually you can't count on it giving the same total talent as historical, let alone a similar distribution.

Think about why the league totals modifiers exist and why the game runs some quick sims of the season before it calculates the final modifiers to be used. It's because even with recalc on the game can't calculate talent ratings that will by themselves produce a season that is overall close to historical. It needs the modifiers to slam the output into compliance.

Since the game is incapable of producing an accurate season overall on talent with recalc there should be no expectation development would be able to produce an accurate season and by extension the more difficult mission of duplicating historical distribution of talent.

[David Watts]

Originally Posted by jpblitz

This was part of my experience as well. In the beginning I would be happy players weren't getting injured. Then when looking at their career stats the theme of everyone with 3000 hits and 500 homeruns was a retirement age of 43-44 and a couple of minor injuries. These guys filled the the entire record books. Had something like 23 with 700+ homeruns and 6 with 800+

For the first time in a very long time I'm playing without recalc. I started in 1954 and I'm using historical minors as well. I'm up to August 13 1955. Having a blast. I will keep an eye on the stuff you noted in your opening post.

[Charlie Hough]

I suspect the issue is with injury and fatigue settings, as was suggested earlier. It could also be a problem with not setting up the game to keep evolving and import the right modifiers and settings for every historical season.

If you have a large number of guys playing until they're 43 with only a couple of minor injuries, then clearly you must be using some lower injury settings. Fatigue settings might also be low. Both of these will compound the statistical issues, as players who don't get hurt or fatigued much will play a LOT of games, allowing them to produce more of everything over the span of their careers.

[David Watts]

Originally Posted by Charlie Hough

I suspect the issue is with injury and fatigue settings, as was suggested earlier. It could also be a problem with not setting up the game to keep evolving and import the right modifiers and settings for every historical season.

If you have a large number of guys playing until they're 43 with only a couple of minor injuries, then clearly you must be using some lower injury settings. Fatigue settings might also be low. Both of these will compound the statistical issues, as players who don't get hurt or fatigued much will play a LOT of games, allowing them to produce more of everything over the span of their careers.

I ran a random debut for 30+ seasons using 24. 3 year recalc with development on. I stared league with low long term injuries and High(modern day) short term. Normal fatigue. By the time I stopped play, I had what I considered to be too many 3000 hit guys. Tony Gwynn actually reached 4000 hits. I don't think it was a matter of guys playing too long, instead it was like you said too many games per season. Gwynn played pretty much injury free his entire career.

[jpblitz]

Originally Posted by David Watts

I ran a random debut for 30+ seasons using 24. 3 year recalc with development on. I stared league with low long term injuries and High(modern day) short term. Normal fatigue. By the time I stopped play, I had what I considered to be too many 3000 hit guys. Tony Gwynn actually reached 4000 hits. I don't think it was a matter of guys playing too long, instead it was like you said too many games per season. Gwynn played pretty much injury free his entire career.

You guys might have figured it out. I had everything set to automate recalcs and modifiers to advance with each progressive year.

The one noticeable thing is almost every player in question has 20+ years with few injuries, presumably keeping their talent ratings high deep into their careers. I'm going to restart and make some changes to the injury settings and probably even the aging modifier just to be safe.

[thehef]

Originally Posted by jpblitz

You guys might have figured it out. I had everything set to automate recalcs and modifiers to advance with each progressive year.

The one noticeable thing is almost every player in question has 20+ years with few injuries, presumably keeping their talent ratings high deep into their careers. I'm going to restart and make some changes to the injury settings and probably even the aging modifier just to be safe.

Cool. Be sure and report back here with your findings

[David Watts]

Just for fun, I created a historical league starting in 1901. Using the OOTP development engine only, so no recalc. As soon as teams had enough players I set injuries to High(modern day). I have fatigue set to high. 14 man rosters and I have it set so teams have 5 starters with 1 pitcher in the bullpen. Starters can be used in relief.

I'm up to 1927. I'm really impressed with what I'm seeing overall. In fact, the combo of high fatigue and the modern day injury setting may be too much when recalc is off. Cobb may end up playing less games overall than he did in real life. He just passed Cap Anson to become the all time hits leader, but unless he gets blessed with the youth molecule, I figure he closes out his career with 3500 to 3600 hits.

I turned off two way players after the 1918 season, so Ruth would have a better chance of being Ruth. He crushed 65 home runs in the 1921 season and has a 61 homer season as well. Like Cobb, he's spent significant time on the IL

[Charlie Hough]

Originally Posted by David Watts

In fact, the combo of high fatigue and the modern day injury setting may be too much when recalc is off.

Yes, high injury rates, especially for long-term injuries, seem a bit extreme in my experience. In a recent historical game in the 1930s, I tried using a modern MLB rate for both short-term and long-term injuries, and the effect was instantaneous. Quite a few players ended up with pretty serious injuries, and I have ended up settling on a high rate for short-term injuries and a normal rate for long-term injuries. Unfortunately, the normal rate for long-term is a bit too infrequent, but it's better than what I was seeing in the alternative. Using the normal setting for long-term injuries will probably result in more players being able to reach or exceed career milestones, but it still results in some crucial injuries to star players.

For example, in my 1930s game, even with the normal setting for long-term injuries, teams would lose a key player for several weeks or several months a few times per season. In one case, Lou Gehrig missed the end of the season and then the World Series due to an injury that was supposed to last for just three weeks, but after his injury time expired, OOTP said he was still out. It no longer listed an estimated length of time that he would miss, but his profile said he might return to the injured list. So it was a random lingering injury where he was out indefinitely. That played a role in the Yankees losing the World Series, and Chick Hafey, whom they had acquired a couple of years earlier, also ended up getting hurt in the field at the start of game three, and he missed the rest of the series.