Restart of Real Player league from 1871

Pyroman · 01-22-2002, 11:53 AM

I'm not sure if anyone remembers my attempt to use Lahman's database starting in 1871 to come up with an alternate Baseball History before but after many months of thinking and rethinking how to do it, I will be starting it again very soon.

The league will start in 1871 with Eight Teams split between the National League and the American League. There will be two expansions sometime between 1881 and 1900 and from 1901 on, the league Structure will follow that of the real Major League Baseball.

The players will be drafted randomly. All 115 players in 1871 will be drafted to the eight teams, then each year after that, all teams will get up to five rounds of amateur draft picks. Some early years will have less than five due to the shortage of imported rookies.

I will be simming each season more carefully than I have in the past and will be using the ERA Calculator to get the league totals to as close to real life as I can.

I will be posting a website with season results from time to time. I will also try to post recaps here.

Specs · 01-22-2002, 01:51 PM

Can you import as early as 1871?

Pyroman · 01-22-2002, 02:05 PM

[quote]Originally posted by Blind Jerry:
Can you import as early as 1871?

Not really, But I found a way around it. I found a way to only have the Rookies of a particular Season end up in the game. 1871 is the best year to start because all 115 players are considered rookies by the Lahman database.

I created a custom league set in 1870 with fictional players. I quickly simmed and got to the end of the season.

I released all players to the free agent pool and deleted them. I then proceeded to the next season(1871) and clicked on yes when it asked to import real rookies. The 115 players from 1871 were then imported.

I closed the game and deleted the league folder (making sure to keep the player.dat file).

Then I created another Custom League set in 1871 with the eight teams I described originally and players set to none.

Once the league was set up, I copied the first player.dat file I had created into this league folder. Hit "Reset, Release all, Draft" and a new league was ready with the 115 players from 1871.

I did notice a lot of the early players in the database are missing birthdays, and some were missing first names. I maually editted the database for these players before I started.

In one of my earlier attempts, I simmed all the way to 1901 before stopping because I was not happy with a lot of the early stats.

[ January 22, 2002: Message edited by: Pyroman ]

Highlander · 01-22-2002, 02:27 PM

Sounds like a great project!! Hope it works for you. It is VERY frustrating to get a few years into a league only to have something go wrong. Good Luck!

BaseballMan · 01-23-2002, 01:28 PM

i did the same thing as you.
Did you just put in any birthday for the missing ones that had more than just one year played or did you actually find them

landru · 01-23-2002, 01:48 PM

I did roughly the same thing you did for my 1871 start. I did some heavy modification of the Lahman database myself, namely fixing the data integrity problem of the players with names starting with W and coalescing all the career stats into a career average year for their rookie year. Whatever position they had the most innings at was their first position in the master.dat. You avoid a lot of player problems this way.

Also, since there weren't enough pitchers in 1871 to fill out the 4 teams in each league, I reset the pitchers from 1871 to about 1881-2 (I've forgotten what my end date was) to start in 1871 by adjusting their pitcher records, master records (birth date and first year) and batter and fielder records. That way, there was enough pitchers to fill the teams with the acceptable result of that there was no pitcher drafted from 1871 through 1881-2. Also, I multiplied the pitcher's games by 10% while keeping the GS the same to simulate relief games so that even though they all import as starting pitchers, they will actually be used will in relief. I am in 1889 now and I've had two expansion drafts to fill out 8 teams in each league.

Pyroman · 01-23-2002, 06:25 PM

[quote]Originally posted by BaseballMan:
i did the same thing as you.
Did you just put in any birthday for the missing ones that had more than just one year played or did you actually find them

For all players missing birthday's, I just put a 1 for month and day if those were missing, and I set the year to 20 years before their debut. So for a player starting in 1872 with no birthday entered, I entered it at 01/01/1852.

Pyroman · 01-23-2002, 06:33 PM

[quote]Originally posted by landru:
[QB]Also, since there weren't enough pitchers in 1871 to fill out the 4 teams in each league, I reset the pitchers from 1871 to about 1881-2 (I've forgotten what my end date was) to start in 1871 by adjusting their pitcher records, master records (birth date and first year) and batter and fielder records.QB]

I didn't mess with too many other stats other than the birth dates. I kept the small amount of pitchers in the early years. I also manually drafted the pitchers for the first few years to make sure each team was slowly building up their rotation at an equal pace.

I found only having one or two pitchers on a team wasn't a big problem in the early years when I had the schedule from 32 - 60 games. By the late 1870's there were enough pitchers on each team to have 4 man rotations.

I just found that without adjusting the ERA settings, the early years had "good" pitchers with ERA's in the 4 - 5 range, and some hitters consistently hitting over .500 in the season. I have been messing around with the numbers for a while and I am almost happy with what I am geting for 1871. I just need to get the HR's a little closer. They keep coming up either too low or too high.

toanstrom · 01-24-2002, 07:37 PM

[quote]Originally posted by landru:
I did some heavy modification of the Lahman database myself, namely fixing the data integrity problem of the players with names starting with W and coalescing all the career stats into a career average year for their rookie year. Whatever position they had the most innings at was their first position in the master.dat. You avoid a lot of player problems this way.

Would you happen to have a copy of this file?

Pyroman · 03-02-2002, 09:29 AM

Well. I was using OOPT3 and was well on my way through the league. I had simmed up thru 1875 when I ran into a problem where the game would freeze up and stop responding during sims. I wasn't exactly sure what the problem was but it frustrated me and I stopped simming for a while.

Then after experimenting with OOPT4 this morning, I am giving it another go. I especially like the fact that the "switching leagues" problem from v3 is now fixed in v4.

In the same kind of way I figured out how to import 1871 players, I have now found a round about way of adding a second league. So in "Pyroman's 1871 League mock 3", I will be starting out with One Eight team league. A second league will be formed sometime in the 1880's. Other than that, I will pretty much be doing things as I originally intended.

throwstrikes · 03-06-2002, 11:04 AM

What settings did you use to get the era calculator to give you realistic results and how did you get those settings?

Charles

Pyroman · 03-06-2002, 12:04 PM

The final settings I used for 1871 turned out to be 100000,35121,5552,1134,3908,10029,0,9556

I used IatricSB's era calculator located here: http://utilities.boys-of-summer.com

Although it only allows you to start in 1901. I saw that the HTML page it uses to calculate simmed league totals is pretty much the same for OOTP3 and OOTP4. Each year, I manually enter the initial ratings and league totals into the texts files as "1901".

I then back up my league file and keep playing the season, using the ERA calculator to analyse the results and give new settings, and then restoring and replaying the season with those new settings until I am happy with the results. If all stats get to within 10% of the MLB actuals, I usually am satisfied and move on to the next season.

I am currently in 1879 right now. I will probably post a 10 year report after the 1880 season is completed.

throwstrikes · 03-06-2002, 12:13 PM

I have used that for 1901 forward, but I have run seasons as often as 10 times and it still is giving me large variances.

Charles

Pyroman · 03-06-2002, 12:53 PM

Do you click "Verify Results" after each sim and use the new numbers it gives you?

BasicGuy · 03-25-2002, 02:27 PM

Pyroman every season at the end you verify the results and rerun the season with the new numbers?

the reason I ask Is cause this the problem I'm haveing enjoying my 1901 and on league. I want to control a team and what fun is It haveing to sim each year 2 times or more? I want to sim 1 season after another and have my results be within 15% of the real league avg(is that really so much to ask from a game?)

Thats the crossroad I'm at now. Do I just sim seasons cause I want some what releasic numbers or just play and accept what I get.

I do enjoy the game.....but lately its been all fantasy baseball for me.

Brad

Pyroman · 03-26-2002, 02:16 PM

BasicGuy,

That can be a problem. What you could do is sim the seasons without really paying attention to anything but the total numbers until they end up where you are satisfied, then using these new totals, restart on last time and play the way you want.

That is the only way I can thing of adopting this method to a League where you actually want to control one of the teams.

BasicGuy · 03-26-2002, 04:43 PM

thats kinda what I figured.....the only other thing I could think of Is lets someone ran all the way through the history of baseball(or just a era)and got numbers that made the output seem close to realistic. If they gave them out to us and we used the same season(someone sims 1910 gets good numbers and shares his numbers on here).
would this work? what works for one person who simmed 1910 should give close to the same results for another right?

maybe we could work out Numbers for each era that got the game within a certin percent of the real decade(I relize going by decade rather than year isnt going to be as realistic but its a easier goal to set)

I'll give It a try this weekend to see If I cant get a reasonable number for 1901-10
If I can get the numbers 10-20% of the real decade I'd be happy.

Brad

BaseballMan · 03-26-2002, 07:47 PM

would it ber better just to test replay seasons?

Cause lets say i sim a career starting in 1903 up to 1935. Now i may change and test each season league totals till i get satisfactory results but
what if i had edited ty cobbs ratings or traded babe ruth in 1920 but use the rattings he developed in the game and not the one s for 1920.

I hope you see what i am getting at...that unless you have the same park factors and the same same teams as real life the league totals in career leagues will always be incorrect cause you may have eddie plank pitching 4th, i may have him pitching 1st. This would then effect the league totals beacuse maybe the other pitcher you put in the 1 spot gives up 75 more hits than plank.

I hope you understand what i am saying that if your gonna test league totals you should use the ball park factors for that season and the replay
season.
Even then wouldnt your league total only be good for your debut year cause afterwards wouldnt the plaers be rated on what they did.
But whatever you do keep at it caaause i think we have made progress. I know pyroman has tryed to run seasons since 1871 and i think in ootp2 it was nearly impossible cause you had to import indivually. I think if we keep working at it we might figure it out.

BasicGuy · 03-27-2002, 04:39 AM

Baseballman I was under the impression that what ever numbers you use in the league setup screen dont matter on the year you use them on.....what I mean Is lets say I find numbers that work good in 1910 and make the league put up deadball era numbers. If I was to use those same numbers on a league from 1995 It should In turn make that league put up deadball era stats...Maybe I'm wrong on that.

And onto what you said about playing different players then what I might use. I dont see how that could affect this. We're talking league avgs.
If one player you play different does better for you some where one player would play worse to avg It out.

The Stadiums though I agree would more than likely affect the outcome though.

Brad

BaseballMan · 03-27-2002, 10:21 AM

maybe.
It might avg out if the players were both starters. If that player went to the bench it might effect the totals.
But you may be able to find a number cause i know i just used 958 for hr in my 1876-1890 league and homeruns wewre about
pretty low per player but i never checked the whole league total.

01-22-2002, 11:53 AM	#1
Pyroman Minors (Double A) Join Date: Jan 2002 Location: New York Posts: 158	Restart of Real Player league from 1871 I'm not sure if anyone remembers my attempt to use Lahman's database starting in 1871 to come up with an alternate Baseball History before but after many months of thinking and rethinking how to do it, I will be starting it again very soon. The league will start in 1871 with Eight Teams split between the National League and the American League. There will be two expansions sometime between 1881 and 1900 and from 1901 on, the league Structure will follow that of the real Major League Baseball. The players will be drafted randomly. All 115 players in 1871 will be drafted to the eight teams, then each year after that, all teams will get up to five rounds of amateur draft picks. Some early years will have less than five due to the shortage of imported rookies. I will be simming each season more carefully than I have in the past and will be using the ERA Calculator to get the league totals to as close to real life as I can. I will be posting a website with season results from time to time. I will also try to post recaps here.

01-22-2002, 02:27 PM	#4
Highlander All Star Starter Join Date: Dec 2001 Posts: 1,080	Sounds like a great project!! Hope it works for you. It is VERY frustrating to get a few years into a league only to have something go wrong. Good Luck! __________________ "The baseball mania has run its course. It has no future as a professional endeavor." — Cincinnati Gazette editorial, 1879

01-22-2002, 02:05 PM	#3
Pyroman Minors (Double A) Join Date: Jan 2002 Location: New York Posts: 158	[quote]Originally posted by Blind Jerry: Can you import as early as 1871? Not really, But I found a way around it. I found a way to only have the Rookies of a particular Season end up in the game. 1871 is the best year to start because all 115 players are considered rookies by the Lahman database. I created a custom league set in 1870 with fictional players. I quickly simmed and got to the end of the season. I released all players to the free agent pool and deleted them. I then proceeded to the next season(1871) and clicked on yes when it asked to import real rookies. The 115 players from 1871 were then imported. I closed the game and deleted the league folder (making sure to keep the player.dat file). Then I created another Custom League set in 1871 with the eight teams I described originally and players set to none. Once the league was set up, I copied the first player.dat file I had created into this league folder. Hit "Reset, Release all, Draft" and a new league was ready with the 115 players from 1871. I did notice a lot of the early players in the database are missing birthdays, and some were missing first names. I maually editted the database for these players before I started. In one of my earlier attempts, I simmed all the way to 1901 before stopping because I was not happy with a lot of the early stats. [ January 22, 2002: Message edited by: Pyroman ]

01-23-2002, 01:28 PM	#5
BaseballMan Hall Of Famer Join Date: Jan 2002 Posts: 6,658	i did the same thing as you. Did you just put in any birthday for the missing ones that had more than just one year played or did you actually find them

01-23-2002, 01:48 PM	#6
landru Minors (Double A) Join Date: Dec 2001 Posts: 101	I did roughly the same thing you did for my 1871 start. I did some heavy modification of the Lahman database myself, namely fixing the data integrity problem of the players with names starting with W and coalescing all the career stats into a career average year for their rookie year. Whatever position they had the most innings at was their first position in the master.dat. You avoid a lot of player problems this way. Also, since there weren't enough pitchers in 1871 to fill out the 4 teams in each league, I reset the pitchers from 1871 to about 1881-2 (I've forgotten what my end date was) to start in 1871 by adjusting their pitcher records, master records (birth date and first year) and batter and fielder records. That way, there was enough pitchers to fill the teams with the acceptable result of that there was no pitcher drafted from 1871 through 1881-2. Also, I multiplied the pitcher's games by 10% while keeping the GS the same to simulate relief games so that even though they all import as starting pitchers, they will actually be used will in relief. I am in 1889 now and I've had two expansion drafts to fill out 8 teams in each league.

01-23-2002, 06:25 PM	#7
Pyroman Minors (Double A) Join Date: Jan 2002 Location: New York Posts: 158	[quote]Originally posted by BaseballMan: i did the same thing as you. Did you just put in any birthday for the missing ones that had more than just one year played or did you actually find them For all players missing birthday's, I just put a 1 for month and day if those were missing, and I set the year to 20 years before their debut. So for a player starting in 1872 with no birthday entered, I entered it at 01/01/1852.

01-23-2002, 06:33 PM	#8
Pyroman Minors (Double A) Join Date: Jan 2002 Location: New York Posts: 158	[quote]Originally posted by landru: [QB]Also, since there weren't enough pitchers in 1871 to fill out the 4 teams in each league, I reset the pitchers from 1871 to about 1881-2 (I've forgotten what my end date was) to start in 1871 by adjusting their pitcher records, master records (birth date and first year) and batter and fielder records.QB] I didn't mess with too many other stats other than the birth dates. I kept the small amount of pitchers in the early years. I also manually drafted the pitchers for the first few years to make sure each team was slowly building up their rotation at an equal pace. I found only having one or two pitchers on a team wasn't a big problem in the early years when I had the schedule from 32 - 60 games. By the late 1870's there were enough pitchers on each team to have 4 man rotations. I just found that without adjusting the ERA settings, the early years had "good" pitchers with ERA's in the 4 - 5 range, and some hitters consistently hitting over .500 in the season. I have been messing around with the numbers for a while and I am almost happy with what I am geting for 1871. I just need to get the HR's a little closer. They keep coming up either too low or too high.

01-24-2002, 07:37 PM	#9
toanstrom Minors (Triple A) Join Date: Dec 2001 Posts: 277	[quote]Originally posted by landru: I did some heavy modification of the Lahman database myself, namely fixing the data integrity problem of the players with names starting with W and coalescing all the career stats into a career average year for their rookie year. Whatever position they had the most innings at was their first position in the master.dat. You avoid a lot of player problems this way. Would you happen to have a copy of this file?

03-02-2002, 09:29 AM	#10
Pyroman Minors (Double A) Join Date: Jan 2002 Location: New York Posts: 158	Well. I was using OOPT3 and was well on my way through the league. I had simmed up thru 1875 when I ran into a problem where the game would freeze up and stop responding during sims. I wasn't exactly sure what the problem was but it frustrated me and I stopped simming for a while. Then after experimenting with OOPT4 this morning, I am giving it another go. I especially like the fact that the "switching leagues" problem from v3 is now fixed in v4. In the same kind of way I figured out how to import 1871 players, I have now found a round about way of adding a second league. So in "Pyroman's 1871 League mock 3", I will be starting out with One Eight team league. A second league will be formed sometime in the 1880's. Other than that, I will pretty much be doing things as I originally intended.

03-06-2002, 11:04 AM	#11
throwstrikes Bat Boy Join Date: Jan 2002 Location: Chesapeake, Virginia Posts: 16	What settings did you use to get the era calculator to give you realistic results and how did you get those settings? Charles

03-06-2002, 12:04 PM	#12
Pyroman Minors (Double A) Join Date: Jan 2002 Location: New York Posts: 158	The final settings I used for 1871 turned out to be 100000,35121,5552,1134,3908,10029,0,9556 I used IatricSB's era calculator located here: http://utilities.boys-of-summer.com Although it only allows you to start in 1901. I saw that the HTML page it uses to calculate simmed league totals is pretty much the same for OOTP3 and OOTP4. Each year, I manually enter the initial ratings and league totals into the texts files as "1901". I then back up my league file and keep playing the season, using the ERA calculator to analyse the results and give new settings, and then restoring and replaying the season with those new settings until I am happy with the results. If all stats get to within 10% of the MLB actuals, I usually am satisfied and move on to the next season. I am currently in 1879 right now. I will probably post a 10 year report after the 1880 season is completed.

03-06-2002, 12:13 PM	#13
throwstrikes Bat Boy Join Date: Jan 2002 Location: Chesapeake, Virginia Posts: 16	I have used that for 1901 forward, but I have run seasons as often as 10 times and it still is giving me large variances. Charles

03-06-2002, 12:53 PM	#14
Pyroman Minors (Double A) Join Date: Jan 2002 Location: New York Posts: 158	Do you click "Verify Results" after each sim and use the new numbers it gives you?

03-25-2002, 02:27 PM	#15
BasicGuy Major Leagues Join Date: Feb 2002 Location: ND Posts: 478	Pyroman every season at the end you verify the results and rerun the season with the new numbers? the reason I ask Is cause this the problem I'm haveing enjoying my 1901 and on league. I want to control a team and what fun is It haveing to sim each year 2 times or more? I want to sim 1 season after another and have my results be within 15% of the real league avg(is that really so much to ask from a game?) Thats the crossroad I'm at now. Do I just sim seasons cause I want some what releasic numbers or just play and accept what I get. I do enjoy the game.....but lately its been all fantasy baseball for me. Brad

03-26-2002, 02:16 PM	#16
Pyroman Minors (Double A) Join Date: Jan 2002 Location: New York Posts: 158	BasicGuy, That can be a problem. What you could do is sim the seasons without really paying attention to anything but the total numbers until they end up where you are satisfied, then using these new totals, restart on last time and play the way you want. That is the only way I can thing of adopting this method to a League where you actually want to control one of the teams.

03-26-2002, 04:43 PM	#17
BasicGuy Major Leagues Join Date: Feb 2002 Location: ND Posts: 478	thats kinda what I figured.....the only other thing I could think of Is lets someone ran all the way through the history of baseball(or just a era)and got numbers that made the output seem close to realistic. If they gave them out to us and we used the same season(someone sims 1910 gets good numbers and shares his numbers on here). would this work? what works for one person who simmed 1910 should give close to the same results for another right? maybe we could work out Numbers for each era that got the game within a certin percent of the real decade(I relize going by decade rather than year isnt going to be as realistic but its a easier goal to set) I'll give It a try this weekend to see If I cant get a reasonable number for 1901-10 If I can get the numbers 10-20% of the real decade I'd be happy. Brad

03-26-2002, 07:47 PM	#18
BaseballMan Hall Of Famer Join Date: Jan 2002 Posts: 6,658	would it ber better just to test replay seasons? Cause lets say i sim a career starting in 1903 up to 1935. Now i may change and test each season league totals till i get satisfactory results but what if i had edited ty cobbs ratings or traded babe ruth in 1920 but use the rattings he developed in the game and not the one s for 1920. I hope you see what i am getting at...that unless you have the same park factors and the same same teams as real life the league totals in career leagues will always be incorrect cause you may have eddie plank pitching 4th, i may have him pitching 1st. This would then effect the league totals beacuse maybe the other pitcher you put in the 1 spot gives up 75 more hits than plank. I hope you understand what i am saying that if your gonna test league totals you should use the ball park factors for that season and the replay season. Even then wouldnt your league total only be good for your debut year cause afterwards wouldnt the plaers be rated on what they did. But whatever you do keep at it caaause i think we have made progress. I know pyroman has tryed to run seasons since 1871 and i think in ootp2 it was nearly impossible cause you had to import indivually. I think if we keep working at it we might figure it out.

03-27-2002, 04:39 AM	#19
BasicGuy Major Leagues Join Date: Feb 2002 Location: ND Posts: 478	Baseballman I was under the impression that what ever numbers you use in the league setup screen dont matter on the year you use them on.....what I mean Is lets say I find numbers that work good in 1910 and make the league put up deadball era numbers. If I was to use those same numbers on a league from 1995 It should In turn make that league put up deadball era stats...Maybe I'm wrong on that. And onto what you said about playing different players then what I might use. I dont see how that could affect this. We're talking league avgs. If one player you play different does better for you some where one player would play worse to avg It out. The Stadiums though I agree would more than likely affect the outcome though. Brad

03-27-2002, 10:21 AM	#20
BaseballMan Hall Of Famer Join Date: Jan 2002 Posts: 6,658	maybe. It might avg out if the players were both starters. If that player went to the bench it might effect the totals. But you may be able to find a number cause i know i just used 958 for hr in my 1876-1890 league and homeruns wewre about pretty low per player but i never checked the whole league total.