Fictional/Historical Tests Thread

[Hrycaj]

I came to the realization that I needed a place to store all my thoughts for the various tests I do with this game. I'm fortunate enough to be on the beta team so I had several months to test things but I have to say, I am still trying to get things to work the way I want. Understandably, when builds get changed so does the way the game acts. Which is totally fine. However, now that the game is out I'm really trying to dial in the game to my style of play. I'm putting my notes here simply to make connections with other players that want to play a similar playstyle or if people had tips for me because I am far from an OOTP expert.

Playstyle Goal: I want to play a total fictional world where I get a similar historical statistical output to real life. I do not want to take a historical game and change settings to do it either. I want to do it through the fictional option.

* I do not need for things to be exact, but I really believe that with the right settings you can play a fictional universe that acts very close to how things played out statistically in real life. I also do not want to edit players. I want to let the sim engine take care of things and I just want to be an observer.

I have read numerous threads about light historical leaderboards. To me the problem comes down to a couple of things. For hitter it is Plate Appearances and for pitchers Innings pitched. If you can get players to achieve historical thresholds in both the counting stats will line up. Of course, that is the hard part. To give you some idea from the hitters standpoint this is the historical leader board for hitters.


Rank Player PA
1 Pete Rose 15,890
2 Carl Yastrzemski 13,992
3 Hank Aaron 13,941
4 Rickey Henderson 13,346
5 Ty Cobb 13,103
6 Albert Pujols 13,041
7 Cal Ripken Jr. 12,883
8 Eddie Murray 12,817
9 Stan Musial 12,721
10 Barry Bonds 12,606
11 Derek Jeter 12,602
12 Willie Mays 12,545
13 Craig Biggio 12,504
14 Dave Winfield 12,358
15 Robin Yount 12,249
16 Alex Rodriguez 12,207
17 Paul Molitor 12,167
18 Adrián Beltré 12,130
19 Eddie Collins 12,087
20 Rafael Palmeiro 12,046
21 Tris Speaker 12,020
22 Omar Vizquel 12,013
23 Miguel Cabrera 11,796
24 Brooks Robinson 11,782
25 Honus Wagner 11,756
26 Frank Robinson 11,744
27 George Brett 11,625
28 Al Kaline 11,597
29 Reggie Jackson 11,418
30 Mel Ott 11,347
31 Cap Anson 11,331
32 Joe Morgan 11,329
33 Ken Griffey Jr. 11,304
34 Rabbit Maranville 11,260
35 Lou Brock 11,240
36 Luis Aparicio 11,231
37 Rusty Staub 11,229
38 Harold Baines 11,092
39 Carlos Beltrán 11,031
40 Gary Sheffield 10,947
41 Johnny Damon 10,917
42 Tony Pérez 10,861
43 Ozzie Smith 10,778
44 Max Carey 10,772
45 Andre Dawson 10,769
46 Paul Waner 10,767
47 Wade Boggs 10,740
48 Darrell Evans 10,737
49 Ichiro Suzuki 10,734
50 Babe Ruth 10,626
Rank Player (2024 PAs) PA
51 Sam Crawford 10,625
52 Chipper Jones 10,614
53 Dwight Evans 10,569
54 Rod Carew 10,550
55 Luis Gonzalez 10,531
56 Billy Williams 10,519
57 Jake Beckley 10,518
58 Nap Lajoie 10,468
59 Steve Finley 10,460
60 Bill Dahlen 10,429
61 Vada Pinson 10,405
62 Roberto Alomar 10,400
63 Ernie Banks 10,396
64 Tim Raines 10,359
65 Nellie Fox 10,351
66 Jim Thome 10,313
67 Harry Hooper 10,277
68 Iván Rodríguez 10,270
69 Sam Rice 10,260
70 Luke Appling 10,254
71 Charlie Gehringer 10,245
72 Jimmy Rollins 10,240
73 Tony Gwynn 10,232
74 Graig Nettles 10,228
75 Roberto Clemente 10,212
76 George Davis 10,186
77 Dave Parker 10,184
78 Fred McGriff 10,174
79 Frankie Frisch 10,101
Eddie Mathews 10,101
81 David Ortiz 10,091
82 Bobby Abreu 10,081
83 Frank Thomas 10,075
84 Mike Schmidt 10,062
85 Bill Buckner 10,037
86 Buddy Bell 10,009
87 Zack Wheat 10,007
88 Chili Davis 9,997
89 Lou Whitaker 9,967
90 Doc Cramer 9,934
91 Mickey Mantle 9,910
92 Sammy Sosa 9,896
93 Fred Clarke 9,860
94 Carlton Fisk 9,853
95 Mickey Vernon 9,839
96 Harmon Killebrew 9,833
97 Goose Goslin 9,830
98 Willie Davis 9,822
99 Gary Gaetti 9,817
100 Ted Williams 9,792

Pete Rose is a bit of an outlier. He played 24 seasons so I'm not necessarily looking to target that sort of career. I would be happy with players landing in the 11 to 14 thousand range. To do so you will need a player to give you a at absolute minimum 17 strong seasons of play. That would get you to 10k in plate appearances. What you really wany is players to give you 20 to 22 year careers. Right now the development engine in a fictional league is too slow to develop players. They are ready to hit the ground running closer to 24 and 25 when wee need them to be ready at 21 or 22. A few 19 and 20 year old outliers would be great. The reason you can make this happen in historical is because of re-calc. We do not have that in fictional. Here is a look at pitchers innings:

Rank Player IP
1 Cy Young * 7,356
2 Pud Galvin * 6,003.1
3 Walter Johnson * 5,914.1
4 Phil Niekro * 5,404
5 Nolan Ryan * 5,386
6 Gaylord Perry * 5,350
7 Don Sutton * 5,282.1
8 Warren Spahn * 5,243.2
9 Steve Carlton * 5,217.2
10 Grover Cleveland Alexander * 5,190
11 Kid Nichols * 5,067.1
12 Tim Keefe * 5,049.2
13 Greg Maddux * 5,008.1
14 Bert Blyleven * 4,970
15 Bobby Mathews 4,956
16 Roger Clemens 4,916.2
17 Mickey Welch * 4,802
18 Christy Mathewson * 4,788.2
19 Tom Seaver * 4,783
20 Tommy John 4,710.1
21 Robin Roberts * 4,688.2
22 Early Wynn * 4,564
23 John Clarkson * 4,536.1
24 Tony Mullane 4,531.1
25 Jim Kaat * 4,530.1
26 Charles Radbourn * 4,527.1
27 Ferguson Jenkins * 4,500.2
28 Eddie Plank * 4,495.2
29 Eppa Rixey * 4,494.2
30 Tom Glavine * 4,413.1
31 Jack Powell 4,389
32 Red Ruffing * 4,344
33 Gus Weyhing 4,337
34 Jim McCormick 4,275.2
35 Frank Tanana 4,188.1
36 Burleigh Grimes * 4,180
37 Ted Lyons * 4,161
38 Randy Johnson * 4,135.1
39 Red Faber * 4,086.2
40 Jamie Moyer 4,074
41 Dennis Martínez 3,999.2
42 Vic Willis * 3,996
43 Jim Palmer * 3,948
44 Lefty Grove * 3,940.2
45 Jack Quinn 3,920.1
46 Bob Gibson * 3,884.1
47 Sad Sam Jones 3,883
48 Jerry Koosman 3,839.1
49 Bob Feller * 3,827
50 Jack Morris * 3,824

Pitchers are going to be a bit harder to dial in I think. I'll be honest, I have already been able to get the hitters close to where I want but the problem I have found is that pitchers last too long career wise. When I mess with pitcher settings it has an effect on the hitters. So that is in essence the problem. Dialing something in to get a believable mix. It won't be perfect, and I'm not interested in perfect. Just semi-realistic.

So all of my changes come from the "Player Development" settings located in the players and team tab under game settings. I do not mess with anything on the stats and ai page other than telling it that I want auto-calc enabled. Pre-calc does not seem to get me anywhere. Also in the options section I have the following checked:
automatically adjust league strategy when advancing to next season. Import adjusted financials settings after each season and auto import historical player creation modifiers. I also always play my all my leagues with injuries set to high realistic modern day regardless of time period. If anyone is actually interested in this work as has more specific questions I will be more than happy to share info.

[luckymann]

Eagerly awaiting what you have to say.

G

[Hrycaj]

While I call this the first test it is not actually even close. I spent some time running tests during beta that often times went nowhere.

Goal: To see if I can get players to last longer to have careers where they approach 14k in plate appearances. For pitchers I want to simply document innings pitched and adjust from there.

Setup: 16 team league starting in 1895 with a full set of minor leagues. No draft. After initial roster construction, all new players enter the league as 18 year old's. Teams sign them as free agents. 156 game seasons.

Player Development Settings:

Batter aging speed = .200
Batter development speed = 3.500
Pitcher aging speed = .200
Pitcher development speed = 3.500
Development target age = Much Younger
Aging Target Age= Much Older

You would think that ramping up development speed to 3.5 times the default level while slowing aging down significantly would break the game. It doesn't. Also, I know my league setup is not historically accurate but I need to dial in stats first before I attempt to add in minors, a draft and historical expansion.

MLB Historical Numbers for PA's

+15k = 1
+14k = 0
+13k = 5
+12k = 16
+11K= 17
+10K = 47

Test Results for PA's

+15k = 1 (15,111)
+14k = 3
+13k = 3
+12k = 13
+11k = 16
+10k = 23

So this was very encouraging! I would be happy with these results for hitters. I have to also keep in mind I'm dealing with a small sample size as this was just one test. I would have to run this setup many times over to see if the results can be duplicated.

Here was what the historical leaderboard looked like:

11 times .400 or more was hit in a season.
27 players were a part of the 3000 hit club. (3819)
2 members of the 600 home run club (692)
1 member of the +2000 RBI club. (5 in real life)

For me, this type of output is something I would be happy with. In my opinion it is a semi-plausible. There are of course a few things I would have to live with. One being the length of player careers. The career leader in hits played at the top level from age 20 to age 49. That is a 30 year career. Nobody has ever done that. (27 is the most). The player with over 15k PA played 31 years. So that is 2 examples right there without looking too deep at everyone. So I have to ask myself is that something I want to overlook. What is more important to me? Service time of stats?

Whether I can live with that or not, what I can't get behind was the pitching results from this test run. In real life only one pitcher threw over 7000 career innings and that was Cy Young. Pug Galvin barely eclipsed 6000 career innings. 11 pitchers (13 if you add Young and Galvin) threw 5000 career innings. In my test the career leader threw 7984.0, there was also another guy that barely eclipsed 7000 innings pitched. 10 players were a part of the 6000 IP club and 19 in the 5000 IP club. So while not horrible I'm wondering if I dial back pitching a bit somewhere I can get these arms more in line.

In conclusion this was a pretty good test. I'm going to do another where I leave the hitters the way I had them for this test and dial back the arms a bit. (I don't know where yet). I'm also going to drop the aging target from much older to older to see if that does anything significant. I'm also going to try and follow history a bit closer. I won't have minors enabled until 1920 (I'll still start in 1895) and then add a level every 10 seasons. I will also add in expansion in the 60's and divisional play and the draft in 1965. I fully realize that I may be changing so many factors that it will be tough to single out any one change but I feel like I'm close to there with the hitters and the pitchers are not far off and I just want to see what it looks like anyway.

[Hrycaj]

Originally Posted by luckymann

Eagerly awaiting what you have to say.

G

Any feedback would be appreciated!

[Scoman]

I have found through testing in historical from 1901 to the present using default setting with these changes produces realistic career ABs for players.

Default historical settings with player fatigue set at very high,5 year recalc and adjust/weaken settings set at 1000/250 and 125/50.


I ran a test 1901-2023 with very good results.

IRL there have been 30 players total in MLB history with over 10000 ABs for their career,these setting produced 35 players over 10000,so pretty close.

[Hrycaj]

Originally Posted by Scoman

I have found through testing in historical from 1901 to the present using default setting with these changes produces realistic career ABs for players.

Default historical settings with player fatigue set at very high,5 year recalc and adjust/weaken settings set at 1000/250 and 125/50.


I ran a test 1901-2023 with very good results.

IRL there have been 30 players total in MLB history with over 10000 ABs for their career,these setting produced 35 players over 10000,so pretty close.

In historical with recalc on you can get great results. What I'm trying to do is make that happen in a fictional world.

[Tiger Fan]

Nice stuff. Appreciate you sharing it. A quick question and maybe I missed it but what were you using for league totals modifiers? Were they stagnant at a particular year setting or did you have them evolving by using the historical adjustments each year. And same question for your player creation modifiers each year. Thanks.

[Hrycaj]

Originally Posted by Tiger Fan

Nice stuff. Appreciate you sharing it. A quick question and maybe I missed it but what were you using for league totals modifiers? Were they stagnant at a particular year setting or did you have them evolving by using the historical adjustments each year. And same question for your player creation modifiers each year. Thanks.

Copied from above. (It was easily missed!)

So all of my changes come from the "Player Development" settings located in the players and team tab under game settings. I do not mess with anything on the stats and ai page other than telling it that I want auto-calc enabled. Pre-calc does not seem to get me anywhere. Also in the options section I have the following checked:
automatically adjust league strategy when advancing to next season. Import adjusted financials settings after each season and auto import historical player creation modifiers.

[Hrycaj]

So this really was not the second test but since it is the second one I will post about I will call it as such. The real second test was a sim that went from 1895-2024 and started as a 16 team league till 1920 where I added a AAA league. I added the rest of the minors in 10 season increments, expanded in 1961, added the draft in 1965 and went to divisional play in 1969. I left the league at 12 teams were league for the rest of the sim at that point. The results were essentially the same where my batters were hanging around for a long time (around 30 years) giving me plenty of plate appearances and the innings pitched were still too high even though I adjusted some setting on the pitching end. This sim crashed a lot (like four or five times) I have been having crashing issues with this version for some reason. I'm not posting much about this run because as soon as I finished I caught notice that a new build was dropped in beta so I knew I would be back at square #1. There was one large takeaway though when I compared the two sims.

The players in these fictional worlds are not sustaining dominance. They are compilers. Here is the ultimate example. The player that led the sim in hits tallied 4199 with over 16,000 plate appearances. In his 30 year career he never once led the league in hitting and his career average for hits in a season was 168. He never had over 200 hits in any one season. In comparison Pete Rose led the league in hitting 7 times and had over 200 hits in a season 10 times. Rose also averaged 194 hits a season in his 24 year career. Ty Cobb led the league in hitting 8 times, hit over 200 a season 9 times and averaged 224 hits a season over his 24 year career. So this will be something I need to look closer at. For now though, on with the tests.

The second test I'm posting about went from 1895-1941 where I pulled the plug on it. I could already tell that it wasn't going to work. Here were my settings:

Batter aging speed = .200
Batter development speed = 3.500
Pitcher aging speed = .800
Pitcher development speed = 3.500
Development target age = Much Younger
Aging Target Age= Default

As you can see I left the batters stuff alone and adjusted the pitcher aging speed and went with default as my aging target.

Pitchers were a bit better in that 8 players were in the 4000 IP pitch club which is a hair low as it looks like nobody was going to make it to 5k IP for the career. Although, I did cut the sim early so you never really know.

I stopped the sim because of the hitters. At this point in the sim I should have had plenty of players over 10k PA's. I didn't have a single one. Our career hit leader had just over 2700 hits. I wasn't seeing the 30 year careers any longer but I also was not seeing guys enter into the league at 20. Most were starting at 23. Some were at 22. This could be because of another change I made based on a comment from Tiger Fan in another discussion. In my last sim I had the newly created players capped at 18. So every time the game generated new players 18 was their max age. In this one he suggested that could be the reason for an issue on players not having dominate careers. So I moved it to 22 for this run and that is why I think I'm seeing the guys at the top of the leaderboards starting their MLB careers at 23.

With all that in mind I think I'm scrapping this run and will test these settings again but with the created player age max going back to 18. The reason why I'm just moving that setting and nothing else is because I did start to see careers lengths come down where players were sometimes playing to their early 40's. So my hope is if I can get some outliers to be MLB ready at 20 or 21 I will see good results. That is all for now.

[Hrycaj]

So, I haven’t posted in a while but I have been testing. Everything I try gives me essentially the same issue. The game is creating compilers and not stars. Players are hanging around for 30 some years to get the statistical outputs they are getting. Talking with Tiger Fan we are trying a different direction. The new idea is to mess with the era modifier file and pick certain years to bump up the modifiers to create “outliers”. Still in the initial phases with this idea but I will post some findings when we have them.

[Syd Thrift]

The other metric for pitchers IMO is IP/season. I'm almost positive you're going to see this trend low for pitchers particularly between around 1950-1980 and while they might make it up by having longer careers, that represents a flatter career path that is ultimately not as valuable to a team's success than a shorter one with smaller tails at both ends.

This seems to be reeeeeeeeeally off for relief pitchers on a year to year basis although again you might well see a Wayne Granger finish his career with around 450 appearances but you're not likely to see 70 games and 100 IP, let alone the 90/144 he did one year (which, you can definitely get the former in fictional leagues so I have argued and will continue to argue that this is not a game engine issue so much as it is a "the way relievers are generated for historical replays" issue).

[Syd Thrift]

dola,

The issue with making your league that flat by boosting development and knocking down aging to that degree is that you're going to create a league with altogether too much talent compared to real life. So every year a team has not just, say, 2 MLB quality first basemen to choose from but 6, and since the game doesn't give a hoot about the name or even the past performance of the incumbent, even a small regression by a starter is going to make him a backup now. And on top of that, you're going to have so many guys with similar numbers that it makes more and more sense to platoon stars in every season except for their absolute primes and even if you don't, to give them set rest patterns to ensure they never play tired.

I guess what I'd probably run is leave the 3 year calc on and see what aging patterns *that* gives you - probably using 1s for both aging and development. If you're testing the game engine itself I'd probably just go for full fictional players without worrying as much about what "Pete Rose" does but YMMV.

FWIW I currently have my league set to I think 1.25 development and 0.75 aging and try to balance out leaguewide talent with a 150 TCR, which is probably not a thing you necessarily want to impose on a "historical" league. I still find that the game does a relatively bad job at cutting 30+ year olds who aren't significantly better than 25 year olds in the system; I think it really just looks at performance level and pays very little attention to age in those cases.

[Tiger Fan]

We are actually using fictional players in a historical environment with a goal - at least for me- is to have a completely fictional player base but still statistical output that believably mirrors real life as we progress through the 20th century. I want a 500 HR guy or 2 by 1940. I want a few 3000 hit club members by that point and maybe a legend approaching 4000 for his career. But I also want league totals and results to be believable. Basically I want to play historical but without real players so the challenge of being a GM is still there. I want a Babe Ruth or Ted Williams but when playing as a GM in a solo league I don’t want to know for certain who those players will be when drafting or trading. An ideal situation for me, especially since I exclusively play stats only, would be a way to use the historical player database but have random names and no real life history tab for all players.

I agree with you that boosting development speed and slowing the aging process does not work well because, while you can get the 4000 hit guy, he plays until age 49 to do so. And as you mentioned it creates a glut of talent pulling everyone to the middle level.

My latest tests with some tweaks to TCR, default injury settings and some player creation settings modifiers are looking somewhat promising without the need to make major adjustments to aging or development speed. But I have to look closer and see if there are unintended side effects.

[Hrycaj]

Originally Posted by Tiger Fan

We are actually using fictional players in a historical environment with a goal - at least for me- is to have a completely fictional player base but still statistical output that believably mirrors real life as we progress through the 20th century. I want a 500 HR guy or 2 by 1940. I want a few 3000 hit club members by that point and maybe a legend approaching 4000 for his career. But I also want league totals and results to be believable. Basically I want to play historical but without real players so the challenge of being a GM is still there. I want a Babe Ruth or Ted Williams but when playing as a GM in a solo league I don’t want to know for certain who those players will be when drafting or trading. An ideal situation for me, especially since I exclusively play stats only, would be a way to use the historical player database but have random names and no real life history tab for all players.

I agree with you that boosting development speed and slowing the aging process does not work well because, while you can get the 4000 hit guy, he plays until age 49 to do so. And as you mentioned it creates a glut of talent pulling everyone to the middle level.

My latest tests with some tweaks to TCR, default injury settings and some player creation settings modifiers are looking somewhat promising without the need to make major adjustments to aging or development speed. But I have to look closer and see if there are unintended side effects.

Right. I think people may be missing our intent here. We want to do things in a fictional universe. In that setup you do not have a re-call option.

[Syd Thrift]

So... the thing with big HR guys in pre-1950s environments is kind of similar to having Ty Cobbs or Nap Lajoies in a fictional deadball era but with the added problem of there not being a lot of HRs to go around. In 1920 the Babe hit I think it was 11% of all of the homeruns hit in the American League and I just don't think you'll ever find a way to run that. HR rates of course went way up after 1920 but even by 1930 they just weren't distributed anywhere near as evenly as they are today. Using historical parks helps with that a little bit - you'll get a boost from the Baker Bowl and from lefties at Yankee Stadium, for example, while guys who play at Griffith Stadium will still put up deadball era type numbers - but IME it doesn't really help enough to get what I think you're looking for.

The real issue IMO is that the average distance in pure talent from the stars of 1900, or 1920, or 1940, compared to the average player (or replacement player if you want to look at it that way) is just plain large than it is today. The color barrier was part of that, probably the slow increase in having a career in baseball being a desired thing as well, but whatever the reasons it's just really, really hard to see a guy hit like .424 in any league because guys who lead the league in hitting simply don't hit 150 points or more above the league average anymore. One way to express that would be to have a lower level of overall talent leaguewide but the top guys as good as they are now. I honestly don't see a way to do that. PCMs adjust everyone's power up or down 10% or whatever; what you want is the game to flatten out the bell curve itself with the best players better relative to the league compared to now.

The best way I can think of doing that is to export players yearly (and on league creation), run an algorithm over the ratings that makes some calculations based on the actual standard deviations against the means of ratings vs what you'd want them to be, and then import them and try it out. I did something similar when trying to create a more realistic 1970s-style league in Draft Day Sports: Basketball and, if I'm being honest, this issue and the amount of work I'd have to do to "fix" it on my own is why my own fictional/historical league is based in the early 70s and not the 1910s.

Tl;dr: I don't think you're going to be able to do this without editing the ratings, and from a development standpoint it might (hey, might not for all I know but I suspect it would) require a lot of work just to make this happen, let alone get it right.

[Tiger Fan]

Originally Posted by Syd Thrift

So... the thing with big HR guys in pre-1950s environments is kind of similar to having Ty Cobbs or Nap Lajoies in a fictional deadball era but with the added problem of there not being a lot of HRs to go around. In 1920 the Babe hit I think it was 11% of all of the homeruns hit in the American League and I just don't think you'll ever find a way to run that. HR rates of course went way up after 1920 but even by 1930 they just weren't distributed anywhere near as evenly as they are today. Using historical parks helps with that a little bit - you'll get a boost from the Baker Bowl and from lefties at Yankee Stadium, for example, while guys who play at Griffith Stadium will still put up deadball era type numbers - but IME it doesn't really help enough to get what I think you're looking for.

I was thinking we had to edit players to get the Babe Ruth type stats but not in my recent tests. Here is an example from my most recent sim. The homerun leader hit 75 homers. One other player hit over 40 and 2 others over 30.

1928 HOME RUN LEADERS IN EACH LEAGUE

The league totals for HR were slightly higher than ideal in 1928 at 7.4% more than real life but I am working on that.

HISTORICAL LEAGUE TOTALS COMPARIONS

I made no edits to any individual player in this test. In fact I simply started this league up with my preset settings in 1871 and did nothing but let it sim on its own overnight and stop at 1950.
I am working with the player creation modifiers, injury and suspension settings and the TCR settings with a goal of having outlier type players but still realistic results from fictional players in a historical league.

I do not want to edit individual players because if I am playing a solo GM I do not want to know who the outliers are so this is looking to be a suitable work around.

CAREER LEADERS 1871-1950 SIM

SINGLE SEASON HITS AND HR LEADERS 1871-1950 SIM

I have still have work to do including:

• The outliers might be just a little too good right now and I am seeing a few too many 19-20 years step in a dominate.
• Triples seem out of whack among leaders which may be a cause of bumping up HRpower and gap power in some player creation classes.
• The league is set for no draft with rookie free agents coming in at age 18 and 5 levels of minors. Right now, I am having a little better success when I bring rookies in from a draft instead of as free agents.

Bottom-line is my belief was the same as yours that power hitting outliers like Ruth were impossible to create without manually editing a player but I am finding, with some work and testing, that is simply not the case.

Not there with a league I would be comfortable investing time in playing just yet, but getting closer every day.

[Hrycaj]

Originally Posted by Syd Thrift

So... the thing with big HR guys in pre-1950s environments is kind of similar to having Ty Cobbs or Nap Lajoies in a fictional deadball era but with the added problem of there not being a lot of HRs to go around. In 1920 the Babe hit I think it was 11% of all of the homeruns hit in the American League and I just don't think you'll ever find a way to run that. HR rates of course went way up after 1920 but even by 1930 they just weren't distributed anywhere near as evenly as they are today. Using historical parks helps with that a little bit - you'll get a boost from the Baker Bowl and from lefties at Yankee Stadium, for example, while guys who play at Griffith Stadium will still put up deadball era type numbers - but IME it doesn't really help enough to get what I think you're looking for.

The real issue IMO is that the average distance in pure talent from the stars of 1900, or 1920, or 1940, compared to the average player (or replacement player if you want to look at it that way) is just plain large than it is today. The color barrier was part of that, probably the slow increase in having a career in baseball being a desired thing as well, but whatever the reasons it's just really, really hard to see a guy hit like .424 in any league because guys who lead the league in hitting simply don't hit 150 points or more above the league average anymore. One way to express that would be to have a lower level of overall talent leaguewide but the top guys as good as they are now. I honestly don't see a way to do that. PCMs adjust everyone's power up or down 10% or whatever; what you want is the game to flatten out the bell curve itself with the best players better relative to the league compared to now.

The best way I can think of doing that is to export players yearly (and on league creation), run an algorithm over the ratings that makes some calculations based on the actual standard deviations against the means of ratings vs what you'd want them to be, and then import them and try it out. I did something similar when trying to create a more realistic 1970s-style league in Draft Day Sports: Basketball and, if I'm being honest, this issue and the amount of work I'd have to do to "fix" it on my own is why my own fictional/historical league is based in the early 70s and not the 1910s.

Tl;dr: I don't think you're going to be able to do this without editing the ratings, and from a development standpoint it might (hey, might not for all I know but I suspect it would) require a lot of work just to make this happen, let alone get it right.

I agree with a lot of this. I learned quickly that flattening out the curve was not working. However, this modifier stuff Tiger Fan and I have been messing with has been promising. I like what you were saying about the league talent. I also believe that the game has a hard time providing an environment where the stars are so much better than the average players. I'm hoping though, with the 550 max ratings now we can maybe do some of these things now. I did take a test league and made each team have scrub players and then created 2 studs. They destroyed the league. So obviously with heavy editing anything is possible. We just want to avoid the edit part. I think this game is actually capable of doing a bit more than what a lot of people give it credit for. In a few days we will probably release a set of things to change and let people judge it for themselves.

[Syd Thrift]

That's really cool and also really promising! I also have a 70 HR guy in my league who's actually more of a 50 HR guy who plays in good HR parks but again this is the 70s, not the 20s and 30s. So it's nice to see that you do get outliers that big. Just getting the leaguewide rates down to their historical numbers probably also shoots that 75 HR year down to 70, which is still too high but arguably still within the bounds of reason (I might not have thought that way prior to 1998 but still).

[Thundercrack]

great thread. looking forward to the results

[Tiger Fan]

I have done about 100 tests and I can almost get what I want but not quite.

I can get the outliers with big single season HR totals and monster career numbers in the 1920-40 era but I am frustrated I cannot get things exactly right.

TEST SETUP
To preface I set up all my test leagues the same way for consistency. I import a historical major league (either 1871 or 1901 as my start usually) and then delete all real players and uncheck the import historical rookies while enabling the checkbox to allow fictional players. I have tried tests both with a draft (using the default which I believe is 20 rounds) and without a draft by having fictional rookie free agents created. This post will focus on tests with the draft enabled.

I also set injury frequency and Suspension frequency to VERY LOW. Note if you start before 1901 you need to pause and change it in 1901 because it automatically flips to LOW that year in historical play. My thinking would be once I found the settings for pre-1950 I liked I would change it to LOW in 1950 and to NORMAL likely somewhere around 1975.

Also note if you are playing with fictional minors you have to at leage startup go to the STATS&AI tab in league settings for each of your minor leagues and manually check "automatically adjust League strategy when advancing to the next season" as it is unchecked by default so if you start in the 1870s your minor leagues will only ever use a 1-man pitching rotation if you fail to change it. I also turn off "enable ghost players" and in year one of my league create 50 scrub level veteran free agents just to ensure I don't run out of players in that first season.

I am also using an era_modifiers.csv and an era_ballparks.csv that I have modified slightly after a number of tests. They are attached at the bottom of this message.

OBJECTIVE
I am trying to get as close as possible to believable, era-appropriate results in both single-season and career totals to real life as possible but with fictional players. I don't need a guy hitting 60 homers in the 1920s but I need someone at least approaching 50. I want a few players around 500 career homers by 1950, at least a dozen with 3000 hits and maybe somebody challenging or even surapssing 4000. I want somewhat realistic pitcher innings pitched and strikeout numbers. Basically I want historical baseball with fictional players so being a solo GM is much more challenging than simply trying to sign Ruth or Gehrig.

RESULTS
I have done probably close to 100 tests using the date range of 1871-1950 with about half starting in in 1871 and the other half in 1901. I have monkeyed with a lot of settings and using both a draft and no draft. I can get close to what I want, frustratingly close, but regardless of the options I try there seems to be one or 2 major sticking points that make it a no go for me.

Here is the single season and career HR leaderboard for my most recent test. It was an 1871 start and is in 1947 now using a draft set to the default 20 rounds and generate players for 27. (5 levels of minors, 16 MLB teams like all my tests)

The single season leaderboard is almost perfect. Elliott Floyd might have got a little carried away with 75 homers in 1921 but I can really live with those results. The career leaderboard is a little disappointing as obviously Floyd flamed out quickly and I would like to see a player or two at least approaching 500 career HR by this point. I will also add that my career hits leaderboard (not pictured) was a little low with the leader at 3,138 and only 6 guys over 3,000. Not ideal, but I can still accept it as it is better than many tests.

Before I get to the strikeout column above I should add that my historical sim accuracy numbers were nearly perfect with only a very few exceptions showing in the red. Here is the numbers from 1921-47 as that is all that fits in a single screenshot.

But the sim does not work for me primarily because of the crazy batter strikeout numbers. And they are actually a little better in this test than a few others I ran with the same settings. Look at the totals in the first image with the homerun numbers. Elliott Floyd, my single season HR king, fanned 260 times in 641 plate appearances in his record setting 75 HR season. In other tests I have seen guys with half of their plate appearances being K's if they are big home run hitters.

My league totals for K's/AB for 1921 were almost perfect according to the Historical Simulation Accuracy report. My league had 1.4% more than real-life (1922 spiked for some reason) in 1921 and a look at the single season leaderboard for the 1921 NL shows Floyd was the culprit. Here they are sorted by Home Run leaders.

And that is where it breaks down for me. So Floyd was a Babe Ruth that year, a total outlier (the AL HR King had 15 that season) with 60 homers more than anyone else in baseball. But he also struck out 182 times more than anyone else. His eye rating (I use 1-100 with 100% scouting accuracy so I can check basic ratings in their history for testing) was 55. It was higher than each of the next 3 players on the HR leaderboard for that season. They were 47,39,47. His contact rating is the culprit as he was at 35 while those next 3 highest in homers ranged from 71-98.

I would not worry about it if it was just a unique player but in every test I run guys with monster HR totals in the 1920s or 30s also fan an incredibly unrealistic number in compared to the league average.

I am only guessing but I assume it has something to do with creating players to match modern day parameters so a high HR hitter is also going to be a very high strikeout guy.

I should note that in many of these tests with the draft turned on I never see a player in this era belt more than 35 homers. So it really feels like hit and miss but invariably if I get a guy who hits more than 40 HR, he has a very high K total and also a very short career. Here is the career totals of Elliott Floyd

What I love is he was a 5th round draft pick. I know some users do not like it but count me as a huge fan of finally seeing more realistic draft boom/busts in this build. In my testing here I have seen so many later round picks blossom and first rounders fail to live up to expectations (granted I am using 200 TCR) so that it feels like a baseball draft with greater unpredictability. The drafting in previous OOTP versions was disappointing in my opinion because even with ramped up TCR numbers you still had a far higher percentage of first round picks becoming solid big leaguers than you ever see in real life.

SUMMARY
If you are still with me- and I apologize if this is so long- I am almost completely satisfied with the draft option and my settings. However two things kill it for me
1- The crazy number of strikeouts that my outlier power hitters accumulate.
2- The fact that the career leaderboards are not there. With a draft even when I bump up batter development speed (see screenshot below for my settings with this test) guys for the most part do not debut early enough to have long careers where they can compile big counting stat numbers. The HR guys make sense I suppose as the strikeouts end up dooming them quickly as I imagine that is why Cincinnati cut Elliott Floyd's playing time and ultimately dealt him to Detroit.

You might suggest bump up batter development even higher but when I do that it hurts the career total accumulators even more as guys then get forced out too quickly by new young players coming in. I just can't find the balance that works.

Here are the settings I used on the PLAYERS & TEAM screen. All other settings were the same every test as listed above.

Notes on my modified text files.

era_modifiers.txt file. I think the changes to it are the key to getting those outlier power hitters as I never had them before the adjustments started. I bumped up the power numbers and made a couple of other minor adjustments in various years between 1915-1940 with the idea it would give me a few elite power hitters. I have tinkered with it a fair bit and found this file works well for no draft. It is attached to this message. To use place in your database folder and rename it to era_modifiers.txt but backup your original first

era_ballparks.txt file I edited it to make no stadium give a HR rating over 1.300 or under 0.800. I did this because in tests my HR king always came from St Louis -usually the Browns- because the modifiers for Sportsman's Park in that era feel far too high.

(EDIT) the era ballparks file is too large to attach. It is easy enough to edit. I just sorted by HR overall, HRvsL and HRvsR and anything over 1.300 I lowered to 1.300 and anything below .800 I bumped up to .800.


I will post the results of my second testing option- which is import new fictional rookies as free agents instead of a draft. It has a lot of positives but also has me stuck because of its own major issue.