Historical database issues - Page 6

dewelar · 04-10-2018, 12:44 AM

Quote:

Originally Posted by Spritze

This is unfixable.

Well, phooey

. I guess I'll have to keep fixing it manually. Thanks for letting me know.

Lukas Berger · 04-10-2018, 03:16 PM

Quote:

Originally Posted by Skrabec1

Hi,

When creating a fictional league and adding historical teams, adding the 1998 New York Yankees and Jorge Posada is not on the roster-

Thanks

Neil

Quote:

Originally Posted by Spritze

Please read the two previous posts to yours.

Reading is an important tool. Try it out?

Posada doesn't show up at all in historical exhibition either, at least not in 1998. So I'm pretty sure something is wrong here. There's another report on this here: http://www.ootpdevelopments.com/boar...93#post4309793

Any ideas from anyone in the know, is this a Markus issue I'd need to PT or is it a db issue in some way, shape or form?

Spritze · 04-10-2018, 07:37 PM

He is in the database properly. A regular boring old school historical league includes hisownself. It is only when gamers are importing or exhibitioning or custom leaguing etc when the challenge kicks in.

Bottom line this is not a database issue more than likely.

See the attachment from a unhorsed around with historical league that is attachmentally attached to this missive.

DB930 · 04-10-2018, 09:17 PM

John W Russell (russel005joh), the 1980s backup catcher, doesn't have the J in his first name capitalized.

actionjackson · 04-13-2018, 04:52 PM

Spritze, I've noticed that there seems to have been a (somewhat massive) downgrade in the pre-defined draft values for position players which means that some very good position players are falling much further in the amateur draft than they did in previous versions. For example:

Johnny Evers (OOTP18 pre-defined draft value): 1784, (OOTP19 pre-defined draft value): 598

Joe Tinker (OOTP18 pre-defined draft value): 1806, (OOTP19 pre-defined draft value): 353

As you can see, that's quite the dropoff. If this was a directive from the powers that be, there's probably not much that can be done, but this does make it easier for the human player to pick off some really good position players lower down in the draft. I believe pitchers have also had their pre-defined draft values upgraded somewhat (not nearly to the extent that position players have had theirs downgraded), which complicates the problem a little further. For example:

George Mullin (OOTP18 pre-defined draft value): 1958, (OOTP19 pre-defined draft value): 2331

Addie Joss (OOTP18 pre-defined draft value): 1102, (OOTP19 pre-defined draft value): 1587

I'm sure I could find many more examples, but these are just four from the 1901 draft (players that debuted in 1902 in MLB).

Spritze · 04-13-2018, 07:56 PM

The numbers don't mean anything by themselves..

Tinker at 353 is #13 of 138 in his draft class.
at 1806 he was #13 in his draft class.

Mullin was #1 and still is.
Joss was #2 and still is.

The changes were made to make sure Ruth is #1 in his draft class (lots of complaints that he was #3) which of course effects everyone else. Since he started as a pitcher all pitchers get a bump.

Damned if you do, damned if you don't.

During the deadball era pitchers have more value than batters
During the steroid era it is the other way round.

MightyVotto · 04-15-2018, 09:54 PM

I'm noticing an old, previously squashed bug appearing again with historical replay leagues. I started a replay league in 1901 and simmed to 1909 with the intent of playing out the 1909 season. I'm using real transactions but not real lineups.

Everything's fine for the players who were imported in from the beginning in 1901. However, once new players come into the league beginning in 1902 and onward, they get imported in with their 2nd year's stat ratings rather than their first year.

For example, Home Run Baker's debut season is 1908. His ratings look more like his 1909 stats. When I transition to 1909, his ratings match up more with his 1910 real life stats. I'm seeing this for all players who came into the league after 1901.

The season ages for all players in the Real Life Stats tab are always one year off from what their BBRef profile says as well. I'm wondering if this may be causing the mismatched imports. In 1909, Baker was 23. The Real Life Stats tab displays 22 for him for that year.

This is annoying if you want to play replay leagues since the results won't mirror historical results. Ruth's 1919 will actually be his 1920 and on and on.

Elf Marine · 04-16-2018, 02:01 AM

I'm being GM for San Diego since toured the ballpark less than a week before the season started. In my news in 2018 Cleveland had a 10 game winning streak and I noticed Francona is the Manager (last I remembered him he was in Boston). So, I looked to see his history and noticed an oddity for last year. It says the Indians won 102 games and finished 1st. Yet, it also says they didn't qualify for the playoffs. Sounds like a historical type issue so I came here. Now I'll see if I can upload a screenshot.

Spritze · 04-19-2018, 07:51 PM

Draft Value

This is for the new guy as it is a good example of pretty much everything.

Draft Value is simply a rookies future "value" in an OOTP universe that uses a rookie draft. There is no hint from the development team as to what it covers nor how it is used so the next bit is just a guess.
This value does not apply to inaugural drafts as many/most of those players are far removed from their rookie year.
This value does not apply to random debut league drafts for the same reason.
This value does not apply to expansion drafts for the same reasons.
This value does not apply to custom league drafts again for the same reasons although depending on what all is customized in some instances it could, I just think it does not as that would be far too complicated to program for.

That does not keep gamers from whining about non-sensical results when they choose to enable it for leagues to whom it probably does not apply. It is best to ignore those complaints if you can. I have tried many ways to explain to the gamers in the historical forums that draft value does not mean what they assume it means and that it only applies correctly to rookie drafts. They would like it to apply to anything with the word draft in it (including their beer) but alas and alack this is not so.

Now on to the things that make up the value. Batters and Pitchers are calculated separately. Currently the average (1871-2017) batter has an 80 draft value and the average pitcher stands at 70. The top 10 all-time values have 9 hitters and one pitcher. The top 100 values have 60 hitters and 40 pitchers. In any year the top values change depending on whether pitchers or batters are putting up better stats so in the deadball era more pitchers lead the way and in the steroid era more batters do. There are only two parts to the draft value. Longevity is the first and largest part which consists of BFP and Seasons Played for pitchers and PA and Seasons Played for batters. The secondary part is how well a batter hit and a pitcher pitched. Basically this is career OPS for batters and runs allowed along with OBP for pitchers. You will notice that defense plays no part. Until OOTP19 defensive ratings were rudimentary. Now that OOTP has them in the fielding file they could be added easily for OOTP20 or for an OOTP19 update. (This has been added.) WAR is not used because it did not yet exist when draft value was created and because it cannot be applied to the entire database of MLB, MiLB, negro, Cuban, Japanese and other players. Plus draft value is longevity based not peak value based. That could be changed of course since it is an entirely made up value. It can be changed to anything you want it to be.

The formulas have not changed much between OOTP15 and 19. Apparently I made some sort of error in OOTP18 that resulted in weirdly inflated batter stats. Some of the posters on the historical forum liked the aberration but it was just a non-reproducible error. Hopefully they will get over it in time. I have many backups of the draft values formulas going back to the very first one. None of them produce those weird results and since batters and pitchers produce similar results (80 and 70) overall it seems like all is as it should be.

This value also points out the fact that neither OOTP nor the database knows names. Gamers often point out that some player should move two spots up in the draft or something. This is usually based on name recognition rather than data or sometimes the data does not match their expectations. Draft Value just lets the chips fall where they may. The formulas apply to every player equally. There is no earthly use in trying to meet pre-conceived notions since each gamer has a different set of those. So when "errors" are pointed out you'll just need to ignore the whine and look at the overall concepts and results. For draft value the concept is longevity and the overall results (80 and 70) seem acceptable.

I always assume any poster of errors is correct and investigate each post. The first thing to do is to see if you can recreate their challenge using plain vanilla out-of-the-box settings. If you cannot then the challenge is not database related but game settings related. Only the Development Team can address those so I ask the error poster to let them know about those issues as I do not know that this portion of the bug reports forum gets reviewed by OOTP often. Plus it is nice to make it easy on them to find game related things in just one place. If the issue is database related then simply fix it right now.

From tests i have run it appears this value is used to determine a grouping of the top three players a team might find useful. One of the three becomes that teams draft pick. Then it is on to the next teams set of needs. This allows for each teams AI to draft based on that teams needs. So draft value is not locked in 1 to 150 but wiggles around a bit depending on whatever it depends on. Which is unknown.

Depending on team and league settings and choices players may under or over perform based on their draft value. This value was added to the master.csv years ago due to AI limitations of the time. Now the AI makes much better decisions than this silly artifact left over from by-gone days. If I had a vote I'd say dump it as redundant and redonkulous.

Spritze

Reed · 04-19-2018, 08:06 PM

Thank you for this detailed explanation. I wish the manual was a little more clear since I learned through trial and error that it needs to be off for inaugural drafts. Maybe it is in the manual and I just missed it.

Spritze · 04-20-2018, 11:41 AM

The manual has many errors and omissions, misrepresentations and misguided assumptions. It has not been updated much and whoever is in charge of it has never posted any questions about anything historical in at least 5 years. Take it with 127,843 grains of salt. Or one banana.

Reed · 04-20-2018, 12:23 PM

Reread the manual and it says for realistic historical inaugural drafts that pre defined draft values should be checked. I went ahead and posted a bug that the manual is wrong and suggested the game engine disregard that check block during inaugural drafts.

Connelly · 04-21-2018, 04:18 AM

In the 1955 historical game....many pitchers listed as 1st base, or OF...Don Larson,Gary Bell, etc....I've counted at least 3 dozen....I do change them when I find them. Just wanted to report it. I'm sure a it's prob been reported, but I wanted to make sure. Thanks

Spritze · 04-21-2018, 10:57 AM

Yes, there is an unresolved bug report on this. The Development Team gave it a look but could not find the flaw (yet). I also noticed that many other position players get turned into first basemen even though they never played there IRL.

NickG · 04-22-2018, 02:20 PM

Generally speaking, when importing for a season where players missed time because they went off to war, players will import as free agents rather than on the teams they would have been on otherwise if you do not check "players miss time according to history". For example, if you start a league in 1953, Willie Mays imports as a free agent, rather than as a member of the Giants. While I'd love to snap up Mays, it seems like guys who miss seasons according to history should be imported as though they were on the team they were on previously, with whatever ratings they would have had if they had been imported as of their last completed season. That probably isn't practical to code, though.

Spritze · 04-22-2018, 05:52 PM

Quote:

Originally Posted by NickG

Generally speaking, when importing for a season where players missed time because they went off to war, players will import as free agents rather than on the teams they would have been on otherwise if you do not check "players miss time according to history". For example, if you start a league in 1953, Willie Mays imports as a free agent, rather than as a member of the Giants. While I'd love to snap up Mays, it seems like guys who miss seasons according to history should be imported as though they were on the team they were on previously, with whatever ratings they would have had if they had been imported as of their last completed season. That probably isn't practical to code, though.

Neutered stats have this feature since they are not real.

NickG · 04-22-2018, 08:58 PM

Here's a screenshot of what I'm talking about. This is a freshly started game in 1953, and here's the list of free agents on day 1 of that game:

https://drive.google.com/file/d/1Wn7...ew?usp=sharing

sprague · 04-23-2018, 01:42 AM

Thought i should post here also to be seen
Had a thread historical potential.

Started a game no minors 2013
About 20 players came in with 250 contact ratings.
Screenshot in thread.

Going to test out a few other years today.

clamel · 04-25-2018, 01:49 PM

This bloke came to the Cubs rookie league in 1992, played in Cubs minors until 1997.
He doesn't pop up until 1997 and that's his Major debut year. Then as shown he has NO minor league stats prior to 1997 and not after either. His pro career shows he started in 1997 directly in Cubs.
Doesn't he exist in the Minor League DB ?

Note I started this run in 1990 so should have been imported in 1992 I think in the minors. I have full Minors.

Spritze · 04-25-2018, 02:11 PM

Starts in 1992 in my test league as a FA since his first year was on a split squad Cubs/Diamondbacks which is not currently supported.

04-10-2018, 07:37 PM	#103
Spritze OOTP Historical Czar Join Date: Dec 2001 Location: Bothell Wa Posts: 7,253	He is in the database properly. A regular boring old school historical league includes hisownself. It is only when gamers are importing or exhibitioning or custom leaguing etc when the challenge kicks in. Bottom line this is not a database issue more than likely. See the attachment from a unhorsed around with historical league that is attachmentally attached to this missive. Attached Images __________________ It's madness, madness, I tell you! For the love of God, don't do it!

04-13-2018, 07:56 PM	#106
Spritze OOTP Historical Czar Join Date: Dec 2001 Location: Bothell Wa Posts: 7,253	The numbers don't mean anything by themselves.. Tinker at 353 is #13 of 138 in his draft class. at 1806 he was #13 in his draft class. Mullin was #1 and still is. Joss was #2 and still is. The changes were made to make sure Ruth is #1 in his draft class (lots of complaints that he was #3) which of course effects everyone else. Since he started as a pitcher all pitchers get a bump. Damned if you do, damned if you don't. During the deadball era pitchers have more value than batters During the steroid era it is the other way round. __________________ It's madness, madness, I tell you! For the love of God, don't do it! Last edited by Spritze; 04-13-2018 at 08:15 PM.

04-15-2018, 09:54 PM	#107
MightyVotto Minors (Double A) Join Date: Aug 2011 Posts: 125	Old Historical Import Bug Back I'm noticing an old, previously squashed bug appearing again with historical replay leagues. I started a replay league in 1901 and simmed to 1909 with the intent of playing out the 1909 season. I'm using real transactions but not real lineups. Everything's fine for the players who were imported in from the beginning in 1901. However, once new players come into the league beginning in 1902 and onward, they get imported in with their 2nd year's stat ratings rather than their first year. For example, Home Run Baker's debut season is 1908. His ratings look more like his 1909 stats. When I transition to 1909, his ratings match up more with his 1910 real life stats. I'm seeing this for all players who came into the league after 1901. The season ages for all players in the Real Life Stats tab are always one year off from what their BBRef profile says as well. I'm wondering if this may be causing the mismatched imports. In 1909, Baker was 23. The Real Life Stats tab displays 22 for him for that year. This is annoying if you want to play replay leagues since the results won't mirror historical results. Ruth's 1919 will actually be his 1920 and on and on.

04-16-2018, 02:01 AM	#108
Elf Marine Minors (Single A) Join Date: Mar 2016 Location: Houston, TX Posts: 70	Team won 102 Games in 2017 yet says didn't make playoffs I'm being GM for San Diego since toured the ballpark less than a week before the season started. In my news in 2018 Cleveland had a 10 game winning streak and I noticed Francona is the Manager (last I remembered him he was in Boston). So, I looked to see his history and noticed an oddity for last year. It says the Indians won 102 games and finished 1st. Yet, it also says they didn't qualify for the playoffs. Sounds like a historical type issue so I came here. Now I'll see if I can upload a screenshot. Attached Images

04-19-2018, 07:51 PM	#109
Spritze OOTP Historical Czar Join Date: Dec 2001 Location: Bothell Wa Posts: 7,253	Draft Value This is for the new guy as it is a good example of pretty much everything. Draft Value is simply a rookies future "value" in an OOTP universe that uses a rookie draft. There is no hint from the development team as to what it covers nor how it is used so the next bit is just a guess. This value does not apply to inaugural drafts as many/most of those players are far removed from their rookie year. This value does not apply to random debut league drafts for the same reason. This value does not apply to expansion drafts for the same reasons. This value does not apply to custom league drafts again for the same reasons although depending on what all is customized in some instances it could, I just think it does not as that would be far too complicated to program for. That does not keep gamers from whining about non-sensical results when they choose to enable it for leagues to whom it probably does not apply. It is best to ignore those complaints if you can. I have tried many ways to explain to the gamers in the historical forums that draft value does not mean what they assume it means and that it only applies correctly to rookie drafts. They would like it to apply to anything with the word draft in it (including their beer) but alas and alack this is not so. Now on to the things that make up the value. Batters and Pitchers are calculated separately. Currently the average (1871-2017) batter has an 80 draft value and the average pitcher stands at 70. The top 10 all-time values have 9 hitters and one pitcher. The top 100 values have 60 hitters and 40 pitchers. In any year the top values change depending on whether pitchers or batters are putting up better stats so in the deadball era more pitchers lead the way and in the steroid era more batters do. There are only two parts to the draft value. Longevity is the first and largest part which consists of BFP and Seasons Played for pitchers and PA and Seasons Played for batters. The secondary part is how well a batter hit and a pitcher pitched. Basically this is career OPS for batters and runs allowed along with OBP for pitchers. You will notice that defense plays no part. Until OOTP19 defensive ratings were rudimentary. Now that OOTP has them in the fielding file they could be added easily for OOTP20 or for an OOTP19 update. (This has been added.) WAR is not used because it did not yet exist when draft value was created and because it cannot be applied to the entire database of MLB, MiLB, negro, Cuban, Japanese and other players. Plus draft value is longevity based not peak value based. That could be changed of course since it is an entirely made up value. It can be changed to anything you want it to be. The formulas have not changed much between OOTP15 and 19. Apparently I made some sort of error in OOTP18 that resulted in weirdly inflated batter stats. Some of the posters on the historical forum liked the aberration but it was just a non-reproducible error. Hopefully they will get over it in time. I have many backups of the draft values formulas going back to the very first one. None of them produce those weird results and since batters and pitchers produce similar results (80 and 70) overall it seems like all is as it should be. This value also points out the fact that neither OOTP nor the database knows names. Gamers often point out that some player should move two spots up in the draft or something. This is usually based on name recognition rather than data or sometimes the data does not match their expectations. Draft Value just lets the chips fall where they may. The formulas apply to every player equally. There is no earthly use in trying to meet pre-conceived notions since each gamer has a different set of those. So when "errors" are pointed out you'll just need to ignore the whine and look at the overall concepts and results. For draft value the concept is longevity and the overall results (80 and 70) seem acceptable. I always assume any poster of errors is correct and investigate each post. The first thing to do is to see if you can recreate their challenge using plain vanilla out-of-the-box settings. If you cannot then the challenge is not database related but game settings related. Only the Development Team can address those so I ask the error poster to let them know about those issues as I do not know that this portion of the bug reports forum gets reviewed by OOTP often. Plus it is nice to make it easy on them to find game related things in just one place. If the issue is database related then simply fix it right now. From tests i have run it appears this value is used to determine a grouping of the top three players a team might find useful. One of the three becomes that teams draft pick. Then it is on to the next teams set of needs. This allows for each teams AI to draft based on that teams needs. So draft value is not locked in 1 to 150 but wiggles around a bit depending on whatever it depends on. Which is unknown. Depending on team and league settings and choices players may under or over perform based on their draft value. This value was added to the master.csv years ago due to AI limitations of the time. Now the AI makes much better decisions than this silly artifact left over from by-gone days. If I had a vote I'd say dump it as redundant and redonkulous. Spritze __________________ It's madness, madness, I tell you! For the love of God, don't do it! Last edited by Spritze; 04-25-2018 at 08:50 PM.

04-10-2018, 09:17 PM	#104
DB930 Major Leagues Join Date: Mar 2003 Posts: 302	John W Russell (russel005joh), the 1980s backup catcher, doesn't have the J in his first name capitalized.

04-13-2018, 04:52 PM	#105
actionjackson Hall Of Famer Join Date: Apr 2009 Location: Toronto, ON Posts: 6,163	Spritze, I've noticed that there seems to have been a (somewhat massive) downgrade in the pre-defined draft values for position players which means that some very good position players are falling much further in the amateur draft than they did in previous versions. For example: Johnny Evers (OOTP18 pre-defined draft value): 1784, (OOTP19 pre-defined draft value): 598 Joe Tinker (OOTP18 pre-defined draft value): 1806, (OOTP19 pre-defined draft value): 353 As you can see, that's quite the dropoff. If this was a directive from the powers that be, there's probably not much that can be done, but this does make it easier for the human player to pick off some really good position players lower down in the draft. I believe pitchers have also had their pre-defined draft values upgraded somewhat (not nearly to the extent that position players have had theirs downgraded), which complicates the problem a little further. For example: George Mullin (OOTP18 pre-defined draft value): 1958, (OOTP19 pre-defined draft value): 2331 Addie Joss (OOTP18 pre-defined draft value): 1102, (OOTP19 pre-defined draft value): 1587 I'm sure I could find many more examples, but these are just four from the 1901 draft (players that debuted in 1902 in MLB).

04-19-2018, 08:06 PM	#110
Reed Hall Of Famer Join Date: Oct 2013 Posts: 2,339	Thank you for this detailed explanation. I wish the manual was a little more clear since I learned through trial and error that it needs to be off for inaugural drafts. Maybe it is in the manual and I just missed it.

04-20-2018, 11:41 AM	#111
Spritze OOTP Historical Czar Join Date: Dec 2001 Location: Bothell Wa Posts: 7,253	The manual has many errors and omissions, misrepresentations and misguided assumptions. It has not been updated much and whoever is in charge of it has never posted any questions about anything historical in at least 5 years. Take it with 127,843 grains of salt. Or one banana. __________________ It's madness, madness, I tell you! For the love of God, don't do it!

04-20-2018, 12:23 PM	#112
Reed Hall Of Famer Join Date: Oct 2013 Posts: 2,339	Reread the manual and it says for realistic historical inaugural drafts that pre defined draft values should be checked. I went ahead and posted a bug that the manual is wrong and suggested the game engine disregard that check block during inaugural drafts.

04-21-2018, 04:18 AM	#113
Connelly Bat Boy Join Date: Nov 2014 Posts: 6	In the 1955 historical game....many pitchers listed as 1st base, or OF...Don Larson,Gary Bell, etc....I've counted at least 3 dozen....I do change them when I find them. Just wanted to report it. I'm sure a it's prob been reported, but I wanted to make sure. Thanks

04-21-2018, 10:57 AM	#114
Spritze OOTP Historical Czar Join Date: Dec 2001 Location: Bothell Wa Posts: 7,253	Yes, there is an unresolved bug report on this. The Development Team gave it a look but could not find the flaw (yet). I also noticed that many other position players get turned into first basemen even though they never played there IRL. __________________ It's madness, madness, I tell you! For the love of God, don't do it!

04-22-2018, 02:20 PM	#115
NickG Minors (Double A) Join Date: Mar 2002 Posts: 138	Generally speaking, when importing for a season where players missed time because they went off to war, players will import as free agents rather than on the teams they would have been on otherwise if you do not check "players miss time according to history". For example, if you start a league in 1953, Willie Mays imports as a free agent, rather than as a member of the Giants. While I'd love to snap up Mays, it seems like guys who miss seasons according to history should be imported as though they were on the team they were on previously, with whatever ratings they would have had if they had been imported as of their last completed season. That probably isn't practical to code, though. Last edited by NickG; 04-22-2018 at 02:23 PM.

04-22-2018, 08:58 PM	#117
NickG Minors (Double A) Join Date: Mar 2002 Posts: 138	Here's a screenshot of what I'm talking about. This is a freshly started game in 1953, and here's the list of free agents on day 1 of that game: https://drive.google.com/file/d/1Wn7...ew?usp=sharing Last edited by NickG; 04-22-2018 at 09:03 PM.

04-23-2018, 01:42 AM	#118
sprague All Star Starter Join Date: Aug 2011 Posts: 1,947	Thought i should post here also to be seen Had a thread historical potential. Started a game no minors 2013 About 20 players came in with 250 contact ratings. Screenshot in thread. Going to test out a few other years today.

04-25-2018, 01:49 PM	#119
clamel All Star Starter Join Date: Apr 2008 Location: Across the Pond Posts: 1,037	Geremi Gonzalez This bloke came to the Cubs rookie league in 1992, played in Cubs minors until 1997. He doesn't pop up until 1997 and that's his Major debut year. Then as shown he has NO minor league stats prior to 1997 and not after either. His pro career shows he started in 1997 directly in Cubs. Doesn't he exist in the Minor League DB ? Note I started this run in 1990 so should have been imported in 1992 I think in the minors. I have full Minors. Attached Images

04-25-2018, 02:11 PM	#120
Spritze OOTP Historical Czar Join Date: Dec 2001 Location: Bothell Wa Posts: 7,253	Starts in 1992 in my test league as a FA since his first year was on a split squad Cubs/Diamondbacks which is not currently supported. __________________ It's madness, madness, I tell you! For the love of God, don't do it!