2007 League Totals over time

atlbrave1 · 04-16-2007, 10:53 PM

I originally posted about this issue here:
http://www.ootpdevelopments.com/boar...d.php?t=145965

Things seem a bit dead over there so I thought I'd post here:

What started out as an examination of the OOTP's creation of "super pitchers" has turned into a test of OOTP's league totals.

Here's the test: I created a fictional 30 team league (1 league w/ DH, 1 w/out, no interleague - in other words - MLB), removed minors to increase sim speed and ran 25 years. I did this 3 separate times. Legaue totals were not changed, nor were the PCMs. Here are the results:

League Totals OOTP (2007 Settings)
167353 AB 44522 H 8919 2B 898 3B 5451 HR 16222 BB 1850 HB 31828 SO

Average of 25 Years
Test #1
167974 AB 43516 H 8898 2B 1190 3B 4495 HR 15558 BB 1820 HB 28044 SO
Test #2
267687 AB 43535 H 8990 2B 1213 3B 4374 HR 15808 BB 1744 HB 27721 SO
Test #3
167926 AB 43964 H 9120 2B 1202 3B 4494 HR 15874 BB 1792 HB 27432 SO

In all 3 test runs HR's were 20% lower than the league totals. In fact, not once in the 75 test seasons did the league HR total exceed 5200.

Also, in all 3 test runs, the seasons w/ the most HR's occurred early in the history (w/in the 1st 5 years) then tapered off into the 4300 HR range towards the end of the 25 years.

Any explanations? I have a couple (one that I don't like and one that I can't quite wrap my head around) but would like to hear from some of the experts - and also see if this can be replicated as well.

Curtis · 04-17-2007, 02:58 AM

No response to your questions (sorry), but I note that your triples are even further off than your homers, but on the high side. Strikeouts don't look good, either (off more than 10% low).

never_ends · 04-17-2007, 04:41 AM

Replicated yes, solution just yet is a no... I did a test league tonight with similar results to yours.

I wanted a 1960's type league with around 150 HR's ave per team so I set up a league, 10 AL and 10 NL. AAA, AA, A, R + Mexican Winter League.

Imported 1961 modifiers into league, wondered about the power rating modifier dropping to .754 but went ahead anyway... first 5 seasons the teams averaged 65 HR's per team (about 1,300 total for the 20 teams), about 100 HR's short per team of what they should have been. So... raised the league total modifier to compensate the pct needed (from 1.000 to 2.500) and sure enough the totals per team fell right into the 150-170 HR ave per team bracket for years, perfect! until...

25 years later a rookie named Cox comes up, shatters the HR record (was 60 held by two guys which was fine) by hitting 76... next season the guy hits .411 with 97 HR's!... so, maybe messing with the modifier was not a good idea. So, not sure how to solve the low HR problem I had at the start with the 1961 stock modifier settings, working on it.

RonCo · 04-17-2007, 06:18 AM

League totals are not truely league totals. You can tell this through the following test: double your league totals HR setting and run a test. You will see HR totals across the league drop heavily. If league totals were true counting totals, you would expect to see the league's HR count double when you doubled that value.

Stats are created by the various talents/ratings of players in the league, modulated by the league totals and then influenced by ballparks.

League totals are better thought of as average rate values by which players are compared. In other words, the average HR hitter in the league would hit HR at a rate of .032/AB (5451/167353). If you double it, the average HR hitter would hit them at a .064 clip. Now, OOTP players have ratings that correspond to rates. I don't know exactly what they are, but let's say that I have a player whos'e HR rate = .032/AB. In the original league, that player will hit about 18 HR in 550 AB. However, in the second league, his .032 expected rate will be modified down because an average HR hitter is expected to hit HR at a .064 clip, and he is only half that good. So I would expect about 8 HR from this guy if the HR league totals were doubled.

RonCo · 04-17-2007, 06:20 AM

So, ultimately, if you want to rais HRs in your league, you lower the League Totals.

Solonor · 04-17-2007, 06:36 AM

Even better than messing with the totals would be to pick a year that you like and import the settings at the top of the Strategy setup page, then use the League Totals Modifiers to bump up or down the numbers you want to wind up with. The modifiers always made more sense to me than the totals, since if you want the HR's to go up, you bump up the modifier for HR's.

RonCo · 04-17-2007, 09:12 AM

Quote:

Originally Posted by Solonor

Even better than messing with the totals would be to pick a year that you like and import the settings at the top of the Strategy setup page, then use the League Totals Modifiers to bump up or down the numbers you want to wind up with. The modifiers always made more sense to me than the totals, since if you want the HR's to go up, you bump up the modifier for HR's.

Even that isn't foolproof, since the actual stats will eventually be determined by the ratings/talents of the players.

atlbrave1 · 04-17-2007, 09:44 AM

If the league totals aren't the answer then how should I set up my league to get totals similar to the league totals over a long period of time?

Solonor · 04-17-2007, 10:45 AM

Quote:

Originally Posted by atlbrave1

If the league totals aren't the answer then how should I set up my league to get totals similar to the league totals over a long period of time?

As long as you don't change them every year, they should stay the same over time (within reason). You just pick a year you like, experiment with the modifiers until you find the ones that give you the results you want, then use those forever (or until you get sick of them).

League results don't (or shouldn't) vary greatly from year-to-year, all things being equal--i.e., you don't change a bunch of stuff regarding development or injuries or the draft pool or whatever.

That was a major problem with early versions of 2006...without making any changes to league setup, the league ERA would start to rise like crazy and the BA would sink (or vice versa, it's been a while). That was fixed a long time ago, though.

atlbrave1 · 04-17-2007, 11:37 AM

I agree entirely that the league totals annually should be close to the league totals at setup if nothing is changed (close to me would be 5% over a long period of time) -- my point is how do you explain what is occuring per my original post. HR's are 20% lower over a 25 yr period (not to mention the 10% reduction in K's and the 30% increase in 3B's.) I can accept some fluctuation but I don't think I should have to change the PCM's or the league totals to play in the "modern" era.

Solonor · 04-17-2007, 11:58 AM

No. There is no direct correlation between the poorly named (imho) "League Totals" and the actual results. You are not going to get league totals to equal "League Totals" no matter what you do. The "League Totals" are the seed numbers for the league. They are not a target number to be met. (See RonCo's explanation above.)

What I am saying is that once you get the results you like, the results won't vary much over time...not that they will come anywhere close to the "League Totals".

DougWyatt · 04-17-2007, 12:05 PM

Quote:

Originally Posted by RonCo

Even that isn't foolproof, since the actual stats will eventually be determined by the ratings/talents of the players.

This isn't true.

The league stats are guided by the League Totals (although as stated, not in any way the same).

League stats are not guided by ratings and talents of your players.

IE. If your League Totals generate X home runs, then your league will always generate approximately X home runs (unless you change the totals or modifiers). It doesn't matter if the whole league is rated 100/100 power, or 2/2 power.

The player ratings are for comparison.

atlbrave1 · 04-17-2007, 12:25 PM

OK -- So, I guess I am back to my question a couple of posts ago. What settings would I need to use to represent the "modern" era? Since the 1.000 PCMs do not generate these numbers over time, am I supposed to run multiple test runs to get today's numbers to reproduce on a consistent basis? These are the numbers I am looking to reproduce:
HR's: 5100 - 5900
2B: 8400 - 9400
3B: 850 - 950
BB: 15000 - 17000
SO: 30000 - 33500
These are all rough approximations of the default League totals +/- 5%. I'm not looking for the exact numbers year in, year out, I realize that the ratings do have an effect, but I would think think that +/- 5% is w/in the realm of possibility.

I have a feeling I know what is causing the numbers (especially HR's and K's) to be well below what the modern era statndards are. I think the A.I. is overvaluing 1) Speed, 2) Avoid K's, 3) Movement (for P) and over time, less and less power hitters are getting AB's, thereby reducing HR's & K's, along w/ pitchers w/ better movement ratings getting a majority of IP. Maybe I'm wrong, its just my guess.

DougWyatt · 04-17-2007, 12:30 PM

Quote:

Originally Posted by atlbrave1

OK -- So, I guess I am back to my question a couple of posts ago. What settings would I need to use to represent the "modern" era? Since the 1.000 PCMs do not generate these numbers over time, am I supposed to run multiple test runs to get today's numbers to reproduce on a consistent basis? These are the numbers I am looking to reproduce:
HR's: 5100 - 5900
2B: 8400 - 9400
3B: 850 - 950
BB: 15000 - 17000
SO: 30000 - 33500
These are all rough approximations of the default League totals +/- 5%. I'm not looking for the exact numbers year in, year out, I realize that the ratings do have an effect, but I would think think that +/- 5% is w/in the realm of possibility.

I have a feeling I know what is causing the numbers (especially HR's and K's) to be well below what the modern era statndards are. I think the A.I. is overvaluing 1) Speed, 2) Avoid K's, 3) Movement (for P) and over time, less and less power hitters are getting AB's, thereby reducing HR's & K's, along w/ pitchers w/ better movement ratings getting a majority of IP. Maybe I'm wrong, its just my guess.

This may be a good starting point (or exactly) for you.

http://www.ootpdevelopments.com/boar...37&postcount=1

atlbrave1 · 04-17-2007, 12:42 PM

Thanks for the link, but I am looking to recreate today's baseball. I have messed arounnd a bit w/ the league totals and modifiers to create a 1970's league (need to get back to that soon w/ the sweet retro uniform mod out there -- love that Padre brown), but I am just trying to create a league to model today's game and it seems to me that the settings provided are not reproducing that game over time.

RonCo · 04-17-2007, 01:59 PM

Quote:

Originally Posted by DougWyatt

This isn't true.

The league stats are guided by the League Totals (although as stated, not in any way the same).

League stats are not guided by ratings and talents of your players.

IE. If your League Totals generate X home runs, then your league will always generate approximately X home runs (unless you change the totals or modifiers). It doesn't matter if the whole league is rated 100/100 power, or 2/2 power.

The player ratings are for comparison.

I understand why you would think that, but I recommend you take about 5 minutes, set up a test league with a few teams, export rosters and change all power ratings to 1, then import again and run a year (while keeping the dafault league totals). I'll bet you a lunch that you get every team hitting 10 or less HR over the entire season.

I've done this several times.

RonCo · 04-17-2007, 02:01 PM

Here's output from my last test league on the matter

Code:

 Division 1
Team	W	L	PCT	GB	AVG	HR	R	AB	H	2B	3B	BB	K	OBP	SLG	OPS	SB
Miami	90	72	.556	-	.246	4	580	5574	1372	354	41	488	1021	.309	.327	.636	106
Chicago	82	80	.506	8.0	.232	8	542	5463	1269	352	38	524	994	.304	.315	.619	35
New Y	65	97	.401	25.0	.232	2	533	5441	1261	321	63	557	1083	.306	.315	.621	39
Nashv	63	99	.389	27.0	.234	6	501	5638	1322	322	28	545	1059	.308	.305	.612	30
Division 2
Team	W	L	PCT	GB	AVG	HR	R	AB	H	2B	3B	BB	K	OBP	SLG	OPS	SB
San B	115	47	.710	-	.253	7	686	5548	1405	305	55	507	1042	.318	.332	.650	346
Staten	87	75	.537	28.0	.249	7	621	5536	1376	353	48	541	1158	.319	.333	.652	105
Jackson	74	88	.457	41.0	.233	10	491	5469	1277	309	34	477	1114	.301	.308	.609	91
Milwau	72	90	.444	43.0	.238	2	518	5631	1342	330	45	448	1058	.299	.314	.613	131
Sub League 1 Totals
Team	W	L	PCT	GB	AVG	HR	R	AB	H	2B	3B	BB	K	OBP	SLG	OPS	SB
TOTALS					.240	46	4472	44300	10624	2646	352	4087	8529	.308	.319	.627	883
Sub League 2 Batting
Division 1
Team	W	L	PCT	GB	AVG	HR	R	AB	H	2B	3B	BB	K	OBP	SLG	OPS	SB
Charl	95	67	.586	-	.250	3	557	5617	1404	292	41	476	1000	.310	.318	.628	169
Balti	91	71	.562	4.0	.242	5	546	5508	1335	324	36	491	1076	.308	.317	.625	47
Manh	91	71	.562	4.0	.235	9	544	5520	1298	310	37	508	1036	.306	.310	.616	119
Long	85	77	.525	10.0	.218	3	450	5533	1204	275	38	438	1305	.281	.283	.563	232
Division 2
Team	W	L	PCT	GB	AVG	HR	R	AB	H	2B	3B	BB	K	OBP	SLG	OPS	SB
San J	79	83	.488	-	.242	4	548	5467	1321	308	52	470	1035	.306	.319	.625	133
Colum	73	89	.451	6.0	.223	2	475	5431	1212	323	47	463	1120	.289	.301	.590	87
Fresno	71	91	.438	8.0	.242	6	540	5509	1331	368	39	561	1267	.315	.326	.640	102
Las Ve	63	99	.389	16.0	.220	2	422	5498	1207	336	28	482	1217	.285	.292	.577	9
Sub League 2 Totals
Team	W	L	PCT	GB	AVG	HR	R	AB	H	2B	3B	BB	K	OBP	SLG	OPS	SB
TOTALS					.234	34	4082	44083	10312	2536	318	3889	9056	.300	.308	.608	898

Curtis · 04-17-2007, 03:57 PM

atlbrave, my recommendation is to set your league total MODIFIERS to .747 for triples, 1.224 for homers and 1.148 for strikeouts. Don't touch the others. Try another 25 year sim (or a few) and see how close that gets you.

DougWyatt · 04-17-2007, 07:10 PM

Quote:

Originally Posted by RonCo

I understand why you would think that, but I recommend you take about 5 minutes, set up a test league with a few teams, export rosters and change all power ratings to 1, then import again and run a year (while keeping the dafault league totals). I'll bet you a lunch that you get every team hitting 10 or less HR over the entire season.

I've done this several times.

Well shimmy, shimmy, bang-bang. I had no friggin' idea. Thank you, very much, for sharing that.

For as long as I've played OOTP, I've thought it was as I stated; and thought I had read numerous confirmations of that theory.

On the other hand, blech - why did you have to teach me that ? I now need to reteach the entire OOTP portion of my brain.

RonCo · 04-17-2007, 07:20 PM

Quote:

Originally Posted by DougWyatt

Well shimmy, shimmy, bang-bang. I had no friggin' idea. Thank you, very much, for sharing that.

For as long as I've played OOTP, I've thought it was as I stated; and thought I had read numerous confirmations of that theory.

On the other hand, blech - why did you have to teach me that ? I now need to reteach the entire OOTP portion of my brain.

I think very few people actually understand how league totals work. It's not intuitively obvious.

04-17-2007, 06:18 AM	#4
RonCo Hall Of Famer Join Date: Aug 2003 Posts: 10,508	League totals are not truely league totals. You can tell this through the following test: double your league totals HR setting and run a test. You will see HR totals across the league drop heavily. If league totals were true counting totals, you would expect to see the league's HR count double when you doubled that value. Stats are created by the various talents/ratings of players in the league, modulated by the league totals and then influenced by ballparks. League totals are better thought of as average rate values by which players are compared. In other words, the average HR hitter in the league would hit HR at a rate of .032/AB (5451/167353). If you double it, the average HR hitter would hit them at a .064 clip. Now, OOTP players have ratings that correspond to rates. I don't know exactly what they are, but let's say that I have a player whos'e HR rate = .032/AB. In the original league, that player will hit about 18 HR in 550 AB. However, in the second league, his .032 expected rate will be modified down because an average HR hitter is expected to hit HR at a .064 clip, and he is only half that good. So I would expect about 8 HR from this guy if the HR league totals were doubled. Last edited by RonCo; 04-17-2007 at 04:49 PM.

04-17-2007, 06:36 AM	#6
Solonor Hall Of Famer Join Date: Mar 2003 Location: Maine Posts: 4,467	Even better than messing with the totals would be to pick a year that you like and import the settings at the top of the Strategy setup page, then use the League Totals Modifiers to bump up or down the numbers you want to wind up with. The modifiers always made more sense to me than the totals, since if you want the HR's to go up, you bump up the modifier for HR's. __________________ Solonor's Groovy Computer Baseball League - Making baseball a hobbit since 2003! "Beings will come, Frodo. The one constant through all the years has been baseball. Middle Earth has rolled by like an army of Mumakil. It has been erased like a slate, rebuilt and erased again. But baseball has marked the time. This field, this game: it's a part of our past, Frodo. It reminds of us of all that once was good and it could be again. Oh... beings will come Frodo. Beings will most definitely come." - Gladden Field of Dreams

04-17-2007, 11:58 AM	#11
Solonor Hall Of Famer Join Date: Mar 2003 Location: Maine Posts: 4,467	No. There is no direct correlation between the poorly named (imho) "League Totals" and the actual results. You are not going to get league totals to equal "League Totals" no matter what you do. The "League Totals" are the seed numbers for the league. They are not a target number to be met. (See RonCo's explanation above.) What I am saying is that once you get the results you like, the results won't vary much over time...not that they will come anywhere close to the "League Totals". __________________ Solonor's Groovy Computer Baseball League - Making baseball a hobbit since 2003! "Beings will come, Frodo. The one constant through all the years has been baseball. Middle Earth has rolled by like an army of Mumakil. It has been erased like a slate, rebuilt and erased again. But baseball has marked the time. This field, this game: it's a part of our past, Frodo. It reminds of us of all that once was good and it could be again. Oh... beings will come Frodo. Beings will most definitely come." - Gladden Field of Dreams

04-16-2007, 10:53 PM	#1
atlbrave1 Major Leagues Join Date: Apr 2002 Posts: 300	2007 League Totals over time I originally posted about this issue here: http://www.ootpdevelopments.com/boar...d.php?t=145965 Things seem a bit dead over there so I thought I'd post here: What started out as an examination of the OOTP's creation of "super pitchers" has turned into a test of OOTP's league totals. Here's the test: I created a fictional 30 team league (1 league w/ DH, 1 w/out, no interleague - in other words - MLB), removed minors to increase sim speed and ran 25 years. I did this 3 separate times. Legaue totals were not changed, nor were the PCMs. Here are the results: League Totals OOTP (2007 Settings) 167353 AB 44522 H 8919 2B 898 3B 5451 HR 16222 BB 1850 HB 31828 SO Average of 25 Years Test #1 167974 AB 43516 H 8898 2B 1190 3B 4495 HR 15558 BB 1820 HB 28044 SO Test #2 267687 AB 43535 H 8990 2B 1213 3B 4374 HR 15808 BB 1744 HB 27721 SO Test #3 167926 AB 43964 H 9120 2B 1202 3B 4494 HR 15874 BB 1792 HB 27432 SO In all 3 test runs HR's were 20% lower than the league totals. In fact, not once in the 75 test seasons did the league HR total exceed 5200. Also, in all 3 test runs, the seasons w/ the most HR's occurred early in the history (w/in the 1st 5 years) then tapered off into the 4300 HR range towards the end of the 25 years. Any explanations? I have a couple (one that I don't like and one that I can't quite wrap my head around) but would like to hear from some of the experts - and also see if this can be replicated as well.

04-17-2007, 02:58 AM	#2
Curtis Hall Of Famer Join Date: Jul 2006 Location: Watertown, New York Posts: 4,567	No response to your questions (sorry), but I note that your triples are even further off than your homers, but on the high side. Strikeouts don't look good, either (off more than 10% low).

04-17-2007, 04:41 AM	#3
never_ends Minors (Double A) Join Date: Oct 2005 Location: Nevada Posts: 133	Replicated yes, solution just yet is a no... I did a test league tonight with similar results to yours. I wanted a 1960's type league with around 150 HR's ave per team so I set up a league, 10 AL and 10 NL. AAA, AA, A, R + Mexican Winter League. Imported 1961 modifiers into league, wondered about the power rating modifier dropping to .754 but went ahead anyway... first 5 seasons the teams averaged 65 HR's per team (about 1,300 total for the 20 teams), about 100 HR's short per team of what they should have been. So... raised the league total modifier to compensate the pct needed (from 1.000 to 2.500) and sure enough the totals per team fell right into the 150-170 HR ave per team bracket for years, perfect! until... 25 years later a rookie named Cox comes up, shatters the HR record (was 60 held by two guys which was fine) by hitting 76... next season the guy hits .411 with 97 HR's!... so, maybe messing with the modifier was not a good idea. So, not sure how to solve the low HR problem I had at the start with the 1961 stock modifier settings, working on it.

04-17-2007, 06:20 AM	#5
RonCo Hall Of Famer Join Date: Aug 2003 Posts: 10,508	So, ultimately, if you want to rais HRs in your league, you lower the League Totals.

04-17-2007, 09:44 AM	#8
atlbrave1 Major Leagues Join Date: Apr 2002 Posts: 300	If the league totals aren't the answer then how should I set up my league to get totals similar to the league totals over a long period of time?

04-17-2007, 11:37 AM	#10
atlbrave1 Major Leagues Join Date: Apr 2002 Posts: 300	I agree entirely that the league totals annually should be close to the league totals at setup if nothing is changed (close to me would be 5% over a long period of time) -- my point is how do you explain what is occuring per my original post. HR's are 20% lower over a 25 yr period (not to mention the 10% reduction in K's and the 30% increase in 3B's.) I can accept some fluctuation but I don't think I should have to change the PCM's or the league totals to play in the "modern" era.

04-17-2007, 12:25 PM	#13
atlbrave1 Major Leagues Join Date: Apr 2002 Posts: 300	OK -- So, I guess I am back to my question a couple of posts ago. What settings would I need to use to represent the "modern" era? Since the 1.000 PCMs do not generate these numbers over time, am I supposed to run multiple test runs to get today's numbers to reproduce on a consistent basis? These are the numbers I am looking to reproduce: HR's: 5100 - 5900 2B: 8400 - 9400 3B: 850 - 950 BB: 15000 - 17000 SO: 30000 - 33500 These are all rough approximations of the default League totals +/- 5%. I'm not looking for the exact numbers year in, year out, I realize that the ratings do have an effect, but I would think think that +/- 5% is w/in the realm of possibility. I have a feeling I know what is causing the numbers (especially HR's and K's) to be well below what the modern era statndards are. I think the A.I. is overvaluing 1) Speed, 2) Avoid K's, 3) Movement (for P) and over time, less and less power hitters are getting AB's, thereby reducing HR's & K's, along w/ pitchers w/ better movement ratings getting a majority of IP. Maybe I'm wrong, its just my guess.

04-17-2007, 12:42 PM	#15
atlbrave1 Major Leagues Join Date: Apr 2002 Posts: 300	Thanks for the link, but I am looking to recreate today's baseball. I have messed arounnd a bit w/ the league totals and modifiers to create a 1970's league (need to get back to that soon w/ the sweet retro uniform mod out there -- love that Padre brown), but I am just trying to create a league to model today's game and it seems to me that the settings provided are not reproducing that game over time.

04-17-2007, 03:57 PM	#18
Curtis Hall Of Famer Join Date: Jul 2006 Location: Watertown, New York Posts: 4,567	atlbrave, my recommendation is to set your league total MODIFIERS to .747 for triples, 1.224 for homers and 1.148 for strikeouts. Don't touch the others. Try another 25 year sim (or a few) and see how close that gets you.