Spritze or Other Database Gurus: What Is the Best Way to Edit Pitchers' Ratings?

Eugene Church · 04-26-2018, 05:43 PM

What is the best way to edit a pitcher's ratings to get his stats more realistic?

I import the players from the Baseball-Reference Database, but the stats are not complete... a lot of them are missing... I get stats from other sources to come as close to the real stats as possible.

I am replaying the Southern Association from the 40s and 50s... I have played the 1947 season and the 1948 season... now getting ready for the 1949 season... I use the newspaper archives to get the real box scores so my historical lineups are correct and the pitching matchups are correct on a game-by-game basis.

I use the Player Editor to edit the ratings for each year for each player... it works well for the batters, but not well for the pitchers. (see screenshot)

I enter the IP, HRA, BB, K, HB, WP in the Ratings Editor, but the ERA is not correct... it is usually much higher than it should be... I'm sure pitch repertoire, movement, stuff and control ratings affect it.

Why doesn't the Ratings Editor allow you to pitch HA in? That might be part of the problem.

I just play around with the HRA, BB and K until I get something close to the actual ERA for that season.... but usually the BB and K are not close to real life.

I inserted John Hall's real 1949 stats below in the ratings editor.. his real ERA was 3.08... not 4.60 as you see on the Screenshot below.

Much appreciate your help.

Screenshot of Player Profile Screen

Eugene Church · 04-26-2018, 05:51 PM

Deleted.

Eugene Church · 04-26-2018, 05:53 PM

Screenshot
John Hall's pitch repertoire

Eugene Church · 04-26-2018, 06:00 PM

Screenshot
John Hall's Stuff, Movement, Control Ratings

I put his real life stats in again into the Ratings Editor and the same entries produced different results.... the ERA is now 4.91.

Why?

Spritze · 04-27-2018, 11:43 AM

The resulting stats are produced for a "neutral modern major league environment"
NOT
a "non-neutral 1949 minor Southern league environment"

One is mammals the other is cephalopods. Not the same thing at all.

Eugene Church · 04-27-2018, 03:08 PM

Quote:

Originally Posted by Spritze

The resulting stats are produced for a "neutral modern major league environment"
NOT
a "non-neutral 1949 minor Southern league environment"

One is mammals the other is cephalopods. Not the same thing at all.

Sorry, I'm not the sharpest tool in the woodshed... I know nothing about "neutral modern major league environment"... I just thought this was the place to edit a pitcher's ratings.

Spritze, thank you for your reply... much appreciate it.

Are the ratings imported from the minor-league historical database fairly accurate?

I know from experience that the Baseball-Reference Database is missing a lot of stats for a lot of minor leagues... no league standings... no team batting, team pitching or team fielding stats... individual player stats are just BA, Hits, doubles, triples, home runs... very little else.

To accurately replay a season, accurate stats and ratings are needed.

Am I trying to do something that can't be done in OOTPB?

Is it possible to change the imported ratings for pitchers when they need adjustment?

If so, where in the game do I change them?

As you know, I have the Southern Association stats for batting, pitching and fielding... they are pretty accurate, but have some mistakes in them.

I can manually put them in except for the pitchers. I apparently am using the Ratings Editor when I shouldn't.

Is there another way to change the pitcher's ratings to get them to perform somewhat close to what they did in real life?

Spritze · 04-27-2018, 05:54 PM

Quote:

Originally Posted by Eugene Church

Sorry, I'm not the sharpest tool in the woodshed... I know nothing about "neutral modern major league environment"...These words are in your screenshots I just thought this was the place to edit a pitcher's ratings.

Spritze, thank you for your reply... much appreciate it.

Are the ratings imported from the minor-league historical database fairly accurate? Yes

I know from experience that the Baseball-Reference Database is missing a lot of stats for a lot of minor leagues... no league standings... no team batting, team pitching or team fielding stats... individual player stats are just BA, Hits, doubles, triples, home runs... very little else. Many of the OOTP stats come from the Spink Baseball Guides which are more complete.

To accurately replay a season, accurate stats and ratings are needed.

Am I trying to do something that can't be done in OOTPB? I do not know why you are needing to change the built in ratings so I cannot answer this

Is it possible to change the imported ratings for pitchers when they need adjustment? Not in the way you wish to

If so, where in the game do I change them?

As you know, I have the Southern Association stats for batting, pitching and fielding... they are pretty accurate, but have some mistakes in them.

I can manually put them in except for the pitchers. I apparently am using the Ratings Editor when I shouldn't.

Is there another way to change the pitcher's ratings to get them to perform somewhat close to what they did in real life?

They will do so within the game parameters you have chosen by default. It rests in your hands.

NoOne · 04-27-2018, 06:00 PM

basically it's converting relative to modern day values... so, it won't match up to historical environment..

e.g. if the editor consistently under reports walks, you can expect it for every player's 'estimates' in the editor for that league.

easier to explain with batters, but same concepts:

i don't use modern values for HR.. i use fewer. so, when i look in an editor it over-predicts homeruns by about ~10-15%... probably disproportionately affects the high end guys more than the middle of the road guys, too. i.e. and not just due to "10%" multiplied by a larger estimated #.

for your situation to get a rough estimate without any observation:

i would compare the league totals (relative to #of teams and # of games played - not just the "totals" on their own) between the league totals you see for MLB quickstart and your historic league...

e.g. if you see 25% fewer strikeouts, you can expect fewer strikeouts than what the editor predicts... will it be 25%? probably not, but a good ballpark to start... with more experience and observation you can have better guess as to the real difference.

no guarantee the modern Stats and AI total values from a modern league are what the editor is using, either... but a good starting point, nonetheless.

use multiple players and compare eresults to prediction in editor... figure out a rough Conversion % per stat...

if you want to change ratings based on stats input into editor, factor that %-difference between prediction and results into what values you use.

if you find that the editor is 25% over-predicted for K's, make sure to reduce by 20% or in other words use 80% of that value instead (use recipricol of difference - in this case 4/5 is reciprocal of 5/4 - 5/4ths is a 25% over-prediction)

Eugene Church · 04-27-2018, 06:30 PM

I'll just import the players and use the ratings they import with... and just hope they will give a fairly realistic season for each player... knowing that there will always be some variation in OOTPB.

It's darn good at replay a season... but not perfect... has a few aberrations.

I can live with some variation from real life stats.

Eugene Church · 04-27-2018, 06:30 PM

Thanks to Spitze and NoOne for their replies.

Much appreciate them.

NoOne · 04-27-2018, 06:46 PM

if you want to affect statistical output, you are better off adjusting the modifiers for that particular stat.

using "walks" as an example:

too many league wide? or... too many for an individual player? through you observations and personal preferences as to what is realistic/correct relative to your league's context.

one of those being accurate may not lend itself to the toher being accurate.. .which do you care about more?

if you see a value that is way to high for an individual, go ahead and tone down the Walk League Total Modifier (LTM) in Stats and AI settings. -- don't change the totals! use the modifiers.

if you do change walks, it will ikely affect other things related to # of walks per PA etc. keep eye out for additional adjustments teh next few years after any LTM change.

ie you increase walks, which reduces # of ab, so that likely increases BABIP or reduces hits... think it's the babip relative to how LT/LTM work, but proof is in the pudding.

-----------------
you could adjust ratings if you wanted to -- much more complicated and much less predictable results... start by export players to be used in a spreadsheet, apply a factor to the control rating column (one formula at top, then fill down in an empty column), replace that column, delete 'temp' created column for proper formatting, and re-import into game... can quickly add 5% to each players control rating or even using "IF" qualifications make it tiered... if between 0-50 only had 5%, between 51-150 add 10%, 151-200 add 20% etc... you could reshape distribution curve if you wanted to.

pretend the column being altered is "A"

in some empty column type this in the same row# as the top row of the data --

"=a1*1.05"

then "fill" down all the rows that exist in original A column... copy and 'paste special' - "numbers only" option over the original A column. it's now all 5% higher. delete the column you just created... save and re-import. mass adjust every single control rating of every single player in the game.

A DB program like Access/hiedisql can do this even more easily with a sql statement...

Eugene Church · 04-27-2018, 09:17 PM

A huge thank you for taking the time to help me.

I have played two seasons manipulating the batting editor for players and trying to get the pitcher's ERA somewhere close to the real life stats, but it was a lot of manually labor.

Actually the individual stats were really close to real life about 80% of the time... especially the batters.

I think I will just import the players from the 1949 Southern Association season individually and see what happens... if they are reasonably close to real stats, I will continue to play it... if not, just scrap the league.

Thanks again for your help... and thanks for always helping people on the Forum.

You are a very valued member of our OOTPB community.

Spritze is special, too.

He is always ready to answer questions... just like you.

04-26-2018, 05:43 PM	#1
Eugene Church Hall Of Famer Join Date: Aug 2002 Posts: 35,801 Blog Entries: 1	Spritze or Other Database Gurus: What Is the Best Way to Edit Pitchers' Ratings? What is the best way to edit a pitcher's ratings to get his stats more realistic? I import the players from the Baseball-Reference Database, but the stats are not complete... a lot of them are missing... I get stats from other sources to come as close to the real stats as possible. I am replaying the Southern Association from the 40s and 50s... I have played the 1947 season and the 1948 season... now getting ready for the 1949 season... I use the newspaper archives to get the real box scores so my historical lineups are correct and the pitching matchups are correct on a game-by-game basis. I use the Player Editor to edit the ratings for each year for each player... it works well for the batters, but not well for the pitchers. (see screenshot) I enter the IP, HRA, BB, K, HB, WP in the Ratings Editor, but the ERA is not correct... it is usually much higher than it should be... I'm sure pitch repertoire, movement, stuff and control ratings affect it. Why doesn't the Ratings Editor allow you to pitch HA in? That might be part of the problem. I just play around with the HRA, BB and K until I get something close to the actual ERA for that season.... but usually the BB and K are not close to real life. I inserted John Hall's real 1949 stats below in the ratings editor.. his real ERA was 3.08... not 4.60 as you see on the Screenshot below. Much appreciate your help. Screenshot of Player Profile Screen Attached Images Last edited by Eugene Church; 04-27-2018 at 12:17 AM.

04-26-2018, 05:51 PM	#2
Eugene Church Hall Of Famer Join Date: Aug 2002 Posts: 35,801 Blog Entries: 1	Deleted. Last edited by Eugene Church; 04-26-2018 at 06:51 PM.

04-26-2018, 05:53 PM	#3
Eugene Church Hall Of Famer Join Date: Aug 2002 Posts: 35,801 Blog Entries: 1	Screenshot John Hall's pitch repertoire Attached Images Last edited by Eugene Church; 04-26-2018 at 06:46 PM.

04-26-2018, 06:00 PM	#4
Eugene Church Hall Of Famer Join Date: Aug 2002 Posts: 35,801 Blog Entries: 1	Screenshot John Hall's Stuff, Movement, Control Ratings I put his real life stats in again into the Ratings Editor and the same entries produced different results.... the ERA is now 4.91. Why? Attached Images Last edited by Eugene Church; 04-26-2018 at 07:09 PM.

04-27-2018, 11:43 AM	#5
Spritze OOTP Historical Czar Join Date: Dec 2001 Location: Bothell Wa Posts: 7,253	The resulting stats are produced for a "neutral modern major league environment" NOT a "non-neutral 1949 minor Southern league environment" One is mammals the other is cephalopods. Not the same thing at all. __________________ It's madness, madness, I tell you! For the love of God, don't do it!

04-27-2018, 06:00 PM	#8
NoOne Hall Of Famer Join Date: Apr 2015 Posts: 7,228	basically it's converting relative to modern day values... so, it won't match up to historical environment.. e.g. if the editor consistently under reports walks, you can expect it for every player's 'estimates' in the editor for that league. easier to explain with batters, but same concepts: i don't use modern values for HR.. i use fewer. so, when i look in an editor it over-predicts homeruns by about ~10-15%... probably disproportionately affects the high end guys more than the middle of the road guys, too. i.e. and not just due to "10%" multiplied by a larger estimated #. for your situation to get a rough estimate without any observation: i would compare the league totals (relative to #of teams and # of games played - not just the "totals" on their own) between the league totals you see for MLB quickstart and your historic league... e.g. if you see 25% fewer strikeouts, you can expect fewer strikeouts than what the editor predicts... will it be 25%? probably not, but a good ballpark to start... with more experience and observation you can have better guess as to the real difference. no guarantee the modern Stats and AI total values from a modern league are what the editor is using, either... but a good starting point, nonetheless. use multiple players and compare eresults to prediction in editor... figure out a rough Conversion % per stat... if you want to change ratings based on stats input into editor, factor that %-difference between prediction and results into what values you use. if you find that the editor is 25% over-predicted for K's, make sure to reduce by 20% or in other words use 80% of that value instead (use recipricol of difference - in this case 4/5 is reciprocal of 5/4 - 5/4ths is a 25% over-prediction)

04-27-2018, 06:30 PM	#9
Eugene Church Hall Of Famer Join Date: Aug 2002 Posts: 35,801 Blog Entries: 1	I'll just import the players and use the ratings they import with... and just hope they will give a fairly realistic season for each player... knowing that there will always be some variation in OOTPB. It's darn good at replay a season... but not perfect... has a few aberrations. I can live with some variation from real life stats. Last edited by Eugene Church; 04-27-2018 at 06:36 PM.

04-27-2018, 06:46 PM	#11
NoOne Hall Of Famer Join Date: Apr 2015 Posts: 7,228	if you want to affect statistical output, you are better off adjusting the modifiers for that particular stat. using "walks" as an example: too many league wide? or... too many for an individual player? through you observations and personal preferences as to what is realistic/correct relative to your league's context. one of those being accurate may not lend itself to the toher being accurate.. .which do you care about more? if you see a value that is way to high for an individual, go ahead and tone down the Walk League Total Modifier (LTM) in Stats and AI settings. -- don't change the totals! use the modifiers. if you do change walks, it will ikely affect other things related to # of walks per PA etc. keep eye out for additional adjustments teh next few years after any LTM change. ie you increase walks, which reduces # of ab, so that likely increases BABIP or reduces hits... think it's the babip relative to how LT/LTM work, but proof is in the pudding. ----------------- you could adjust ratings if you wanted to -- much more complicated and much less predictable results... start by export players to be used in a spreadsheet, apply a factor to the control rating column (one formula at top, then fill down in an empty column), replace that column, delete 'temp' created column for proper formatting, and re-import into game... can quickly add 5% to each players control rating or even using "IF" qualifications make it tiered... if between 0-50 only had 5%, between 51-150 add 10%, 151-200 add 20% etc... you could reshape distribution curve if you wanted to. pretend the column being altered is "A" in some empty column type this in the same row# as the top row of the data -- "=a11.05" then "fill" down all the rows that exist in original A column... copy and 'paste special' - "numbers only" option over the original A column. it's now all 5% higher. delete the column you just created... save and re-import. mass adjust every single control rating of every single player in the game. A DB program like Access/hiedisql can do this even more easily with a sql statement... Last edited by NoOne; 04-27-2018 at 06:54 PM.*

04-27-2018, 09:17 PM	#12
Eugene Church Hall Of Famer Join Date: Aug 2002 Posts: 35,801 Blog Entries: 1	A huge thank you for taking the time to help me. I have played two seasons manipulating the batting editor for players and trying to get the pitcher's ERA somewhere close to the real life stats, but it was a lot of manually labor. Actually the individual stats were really close to real life about 80% of the time... especially the batters. I think I will just import the players from the 1949 Southern Association season individually and see what happens... if they are reasonably close to real stats, I will continue to play it... if not, just scrap the league. Thanks again for your help... and thanks for always helping people on the Forum. You are a very valued member of our OOTPB community. Spritze is special, too. He is always ready to answer questions... just like you.