2004 Lahman update on 11/16

arod23 · 11-15-2004, 07:17 PM

FYI: The Lahman database (www.baseball1.com) updated through 2004 is due out tomorrow the 16th.

Ankit et al: Are you planning on releasing an update to your CarAvg, NormCarAvg, etc databases?

Ankit · 11-16-2004, 11:14 AM

I most likely will be doing updates. AnkitDB will be updated first, I am not too sure about NormDB or CarAvgDB because I have found that with the development modifiers in OOTP 6, one can get better results using simply AnkitDB. There are too few users for NormDB for me to spend the requisite time in updating it.

Johnny99 · 11-16-2004, 02:32 PM

Quote:

Originally Posted by Ankit

I most likely will be doing updates. AnkitDB will be updated first, I am not too sure about NormDB or CarAvgDB because I have found that with the development modifiers in OOTP 6, one can get better results using simply AnkitDB. There are too few users for NormDB for me to spend the requisite time in updating it.

Ankit - Can you elaborate on your comment re the dev. modifiers? I have found the career avg database is a better overall representation of players career than the standard db, BUT, players come in at too high a talent level for the early stage of their career. Too many players have career high in homers in rookie year, for example. Any help is appreciated.

Ankit · 11-16-2004, 03:57 PM

Quote:

Originally Posted by Johnny99

Ankit - Can you elaborate on your comment re the dev. modifiers? I have found the career avg database is a better overall representation of players career than the standard db, BUT, players come in at too high a talent level for the early stage of their career. Too many players have career high in homers in rookie year, for example. Any help is appreciated.

Sure. I am running a historical league of my own in OOTP 6 after a very long time to see how well my DBs are working. I used AnkitDB to import players initially in 1901 and then have been using CarAvgDB since then. I am now in 1923 and have found that some players benefit more than they should and as a result are skewing things unneccesarily.

For example, when Ruth is imported as a LF in 1918, he comes in with a rating over 100 for HRs. Its expected that he should hit over 40 HRs by 1920 on a regular basis. More often than not, he doesn't and this is why: In CarAvgDB, many players imported from 1914 onwards will come in with a somewhat higher than warranted HR rating because they, IRL, hit more HRs after 1920 in their career due the Ruth Effect.

Rogers Hornsby, as a result imports with a HR rating of around 80 in 1915, when in actuality he doesn't begin to hit HRs in double-digits regularly until 1921. However, due to the CarAvgDB he will begin to hit HR earlier and thus taking them away from Ruth (due to how the League Totals work). Same story for other players including Jack Fournier, Ken Williams and others.

This effect is also true for other players in other ratings, including pitchers and strikeouts. My analysis is to use only the AnkitDB to import players and set the Development Modifiers to +50% for both pitchers and hitters. The "+50%" is not an official recommendation by me, try it out and see if you like it. I have yet to try it out. However, what this does is that it alleviates the problems of players not developing quickly enough and as a result, don't do as expected.

A long answer but I hope you understood my point. If you need more clarification on something, don't hesitate to ask...

Johnny99 · 11-16-2004, 04:53 PM

Quote:

Originally Posted by Ankit

Sure. I am running a historical league of my own in OOTP 6 after a very long time to see how well my DBs are working. I used AnkitDB to import players initially in 1901 and then have been using CarAvgDB since then. I am now in 1923 and have found that some players benefit more than they should and as a result are skewing things unneccesarily.

For example, when Ruth is imported as a LF in 1918, he comes in with a rating over 100 for HRs. Its expected that he should hit over 40 HRs by 1920 on a regular basis. More often than not, he doesn't and this is why: In CarAvgDB, many players imported from 1914 onwards will come in with a somewhat higher than warranted HR rating because they, IRL, hit more HRs after 1920 in their career due the Ruth Effect.

Rogers Hornsby, as a result imports with a HR rating of around 80 in 1915, when in actuality he doesn't begin to hit HRs in double-digits regularly until 1921. However, due to the CarAvgDB he will begin to hit HR earlier and thus taking them away from Ruth (due to how the League Totals work). Same story for other players including Jack Fournier, Ken Williams and others.

This effect is also true for other players in other ratings, including pitchers and strikeouts. My analysis is to use only the AnkitDB to import players and set the Development Modifiers to +50% for both pitchers and hitters. The "+50%" is not an official recommendation by me, try it out and see if you like it. I have yet to try it out. However, what this does is that it alleviates the problems of players not developing quickly enough and as a result, don't do as expected.

A long answer but I hope you understood my point. If you need more clarification on something, don't hesitate to ask...

thanks for the quick response. I also started a historical league using your standard db and importing rookies from the careeravg db, per your recommendation. However, I SLOWED development and continued using the careeravg db. It appears we had the same general thought, but approached it from opposite directions. I do not have enough results, though, to make any kind of assessment of this approach and have since moved on to another league. Your approach seems like the better way to go, though, if one can live with the vagaries of the somewhat random development engine. If you test this out, I would love to hear the results, especially the infamous Ruth problem. Good luck.

ctorg · 11-17-2004, 01:20 PM

Well, 11/16 came and went without a new database.

jmm8356 · 11-17-2004, 02:32 PM

Quote:

Originally Posted by ctorg

Well, 11/16 came and went without a new database.

I demand a refund!

Oh... wait...

ctorg · 11-18-2004, 10:23 AM

Still no new database. Hmm, I guess the new database is floating out in limbo with head to head.

arod23 · 11-18-2004, 11:21 AM

I hope you update your CareerAvg database, I for one have found it useful. I have also simmed historically and believe I have good results. Yes, Ruth doesn't dominate the league in homers like he did in real life (b/c others are also hitting them) but I don't consider that unrealistic. I also don't expect the sim to perfectly match history either, why sim then?

Also, using your regular DB or Lahman to import rookies seems to result in far less "realistic" figures, certainly a lot more randomness to who develops and who doesn't. I have found decent results when I decrease player devp like Johnny99 states (using career averages) and adjust league settings each year.

Anyway, thanks for your DBs and I hope you put together the CareerAvg DB too.

11-15-2004, 07:17 PM	#1
arod23 Major Leagues Join Date: Apr 2004 Location: Philadelphia, PA Posts: 378	2004 Lahman update on 11/16 FYI: The Lahman database (www.baseball1.com) updated through 2004 is due out tomorrow the 16th. Ankit et al: Are you planning on releasing an update to your CarAvg, NormCarAvg, etc databases?

11-16-2004, 11:14 AM	#2
Ankit All Star Reserve Join Date: Feb 2002 Location: Singapore Posts: 603	I most likely will be doing updates. AnkitDB will be updated first, I am not too sure about NormDB or CarAvgDB because I have found that with the development modifiers in OOTP 6, one can get better results using simply AnkitDB. There are too few users for NormDB for me to spend the requisite time in updating it. __________________

11-17-2004, 01:20 PM	#6
ctorg Global Moderator Join Date: Nov 2002 Location: Queens, NY Posts: 9,848	Well, 11/16 came and went without a new database. __________________ My music "When the trees blow back and forth, that's what makes the wind." - Steven Wright Fjord emena pancreas thorax fornicate marmalade morpheme proteolysis smaxa cabana offal srue vitriol grope hallelujah lentils

11-18-2004, 10:23 AM	#8
ctorg Global Moderator Join Date: Nov 2002 Location: Queens, NY Posts: 9,848	Still no new database. Hmm, I guess the new database is floating out in limbo with head to head. __________________ My music "When the trees blow back and forth, that's what makes the wind." - Steven Wright Fjord emena pancreas thorax fornicate marmalade morpheme proteolysis smaxa cabana offal srue vitriol grope hallelujah lentils

11-18-2004, 11:21 AM	#9
arod23 Major Leagues Join Date: Apr 2004 Location: Philadelphia, PA Posts: 378	I hope you update your CareerAvg database, I for one have found it useful. I have also simmed historically and believe I have good results. Yes, Ruth doesn't dominate the league in homers like he did in real life (b/c others are also hitting them) but I don't consider that unrealistic. I also don't expect the sim to perfectly match history either, why sim then? Also, using your regular DB or Lahman to import rookies seems to result in far less "realistic" figures, certainly a lot more randomness to who develops and who doesn't. I have found decent results when I decrease player devp like Johnny99 states (using career averages) and adjust league settings each year. Anyway, thanks for your DBs and I hope you put together the CareerAvg DB too.