Horrible relief pitching league-wide

Darviathar · 06-15-2013, 07:23 PM

I've been working on replaying 1980's seasons, starting with 1983 and playing through 1987 in OOTP 13 with pretty accurate results. I moved my franchise to OOTP 14 in April and am now a little over half way through 1989. I am at wit's end with the horrible relief pitching around the league since I moved to OOTP. No lead seems to be safe for anyone, even with elite closers on the hill. Guys that historically had ERA's in the mid 2's are well into the high fours and fives across the board. I have gone in and increased reliever stamina several times in an attempt to keep relievers "fresh" so that they aren't always pitching tired, but it doesn't seem to help. Historically, 1988 and 1989 were two seasons in which pitching thrived, and offenses shouldn't be setting records. This reminds me of a version of OOTP that I played about a decade ago that saw the same thing happen. Does anyone have any suggestions as how to fix this?

Charlie Hough · 06-15-2013, 09:53 PM

Are you using single-season recalc or three-year recalc?

Have you made sure that you're strictly using historical strategies, player creation modifiers, and other settings?

What ballpark factors are you using?

I suspect that there's a problem somewhere in your settings.

Darviathar · 06-16-2013, 01:29 AM

Quote:

Originally Posted by Charlie Hough

Are you using single-season recalc or three-year recalc?

Have you made sure that you're strictly using historical strategies, player creation modifiers, and other settings?

What ballpark factors are you using?

I suspect that there's a problem somewhere in your settings.

I'm using the historical settings for 1989 although I've changed a few to try to adjust what is near-steroid era offensive totals. I lowered BABIP from .286 to .275 after the all-star break. Lowered catcher and outfielder errors from 1.000 to .863 (side note: I assume lower that number results in fewer errors, and not vice versa?)

I'm using historical transactions, so I've not touched anything like player creation modifiers. Three year recalc. Historical strategies and ballpark factors have not been touched. Basically, I haven't changed anything anywhere but a few of the League Total modifiers. Now that you mention ballpark factors, I'm not sure I even know where to find that.

July 25 1989, looking at the historical simulation accuracy page:

Batting average: .265 (+4.4%)
2b: .045 (+4%)
3b: .007 (+10.4%)
HR: .024 (+9.8%)
BB: .093 (+10.6%)
K: .163 (-1.7%)
R/G: 4.72 (+14.1%)
BABIP .297 (+3.8%)

As you can see, offense is simply inflated league-wide.

NL teams have a slash line of .259/.323/.380
Historically it was .246/.312/.365

4.3 runs per game are being scored, while it was 3.94 historically.

Al slash line is .271/.346/.409
Historically it was .261/.326/.384

The totals in 1988 were somewhat off: 4.1 runs per game scored as to 3.88 historically NL and 4.7 to 4.36 in AL, but not quite as severe as what I'm seeing through about 100 games in 1989.

Charlie Hough · 06-17-2013, 07:50 PM

If you've changed league totals modifiers, I can almost guarantee that this is the source of your problem. These are not adjusted in the way that you would think intuitively. In fact, my understanding is that when you think you're lowering things, you're doing the opposite and vice versa

Have you read the manual sections on these and how to modify them? Read the manual and find the section that addresses this counter-intuitive aspect of things. I bet that's the problem.

Darviathar · 06-18-2013, 01:07 AM

I read about the inverse relationship between totals and stats, but that isn't what I've changed. I've only changed the modifiers, so this shouldn't be the problem.

Charlie Hough · 06-18-2013, 12:19 PM

I still suspect that the problem has to do with things that you've modified. That's usually the prime suspect in any case of statistical output being as skewed as you're experiencing.

But maybe this is something you can take up with the developers. Maybe you can post your league files through the usual support method so they can take a look.

JohnHoward · 06-18-2013, 04:38 PM

I would say one contributing factor is definitely three-year recalc since you have the 1987 outlier season still being rolled into the recalc for 1988 and 1989. For certain seasons when moving forward you really should not count them in recalc. I think there are three seasons where this is true: 1930, 1968, and 1987. You should do a "cold start" with 1931, 1969, and 1988 (actual season totals and modifiers set to 1.000, no recalc) and limited recalc for the following two seasons.

VanillaGorilla · 06-18-2013, 04:54 PM

Quote:

Originally Posted by JohnHoward

I would say one contributing factor is definitely three-year recalc since you have the 1987 outlier season still being rolled into the recalc for 1988 and 1989. For certain seasons when moving forward you really should not count them in recalc. I think there are three seasons where this is true: 1930, 1968, and 1987. You should do a "cold start" with 1931, 1969, and 1988 (actual season totals and modifiers set to 1.000, no recalc) and limited recalc for the following two seasons.

Nice call.

Darviathar · 06-20-2013, 08:33 PM

The 1987 season was indeed an outlier, and ERA's around my league jumped into orbit in comparison to 1985 and 1986. Almost everyone came back to earth in 1988, other than Darryl Strawberry, who hit an astonishing 55 home runs. I managed to win Orel Hershiser's rightful 1988 Cy Young (no 59 scoreless inning streak though) with a season that was easily better than what he did historically, so it isn't like hitting in general was still inflated from 1987. I probably wouldn't be worried about this problem if it didn't seem to be an OOTP 14 (as opposed to the 5 seasons I ran in OOTP 13) phenomenon. I play every game, so I can really "see" just how shaky bullpen's are. The 1988 Dodgers had an excellent bullpen, with Howell (2.08), Pena (1.91), Holton (1.70), Orosco (2.72) and Crews (3.14). I didn't get anywhere near that kind of performance. These guys were so bad all season long that I finished 7 games out. Holton was completely useless all season (5.75), while Jay Howell lost the closer job in May after setting what I consider an unofficial league record for walking the leadoff batter in save opportunities. I played one of the early OOTP's a decade ago and recall bullpens being universally bad for no obvious reason. Just hoping that isn't the case here.

VanillaGorilla · 06-20-2013, 11:07 PM

Quote:

Originally Posted by Darviathar

The 1988 Dodgers had an excellent bullpen, with Howell (2.08), Pena (1.91), Holton (1.70), Orosco (2.72) and Crews (3.14). I didn't get anywhere near that kind of performance. These guys were so bad all season long that I finished 7 games out. Holton was completely useless all season (5.75), while Jay Howell lost the closer job in May after setting what I consider an unofficial league record for walking the leadoff batter in save opportunities. I played one of the early OOTP's a decade ago and recall bullpens being universally bad for no obvious reason. Just hoping that isn't the case here.

To what John Howard offered, and Charlie Hough inquired, it is a skew of recalc.

All of the pitchers you listed, save Crews, had worse 1987 season than 1988. By using 3 year recalc you are telling the program to smooth the output from season to season, so when you compare the 1988 output of your sim to RL 1988 you are getting a discrepancy in indicidual performances.

Throw in the fact that we are talking RPs that face fewer batters than SPs, so a bad streak or two can affect the season stats much more.

At the heart of this is that you told the program to not base each year on each year's RL stats, but a composite of three years. When you do that you cannot expect the results to reflect the best season, only.

Brian Holton did have an ERA of 1.70 in 1988, but he never had another season even close to that one. You told the program to consider more than Holton's 1988 RL performance for your replay of the 1988 season. He is not going to pitch like 1988 because of your instruction.

On the bright side, he should be better in 1989 than he was IRL.

ADD: The bullpen/warm-up bug was real in previous versions and has been fixed in v 14.

Darviathar · 06-21-2013, 02:18 PM

I suppose you guys have now offered a ringing endorsement for the one-year recalc feature. I don't know why I haven't been using it all these seasons. You're right about my 1988 bullpen: Howell, Orosco and Holton were all pretty bad in 1987 historically. However, Holton had been consistently excellent for me for 3 years somehow, so I guess I just assumed he would have an even better season in what was his career year. He "made the club" out of spring training in 1985 (I like to look at spring stats for guys that are marginal players and take the one who performed best to at least start the season) and somehow went 4-2 with a 1.46 ERA in 61 innings. Followed that up with 4-1 2.76 in 65 in 1986, and 2-5 2.57 in 84 in 1987. The computer manager had been urging me to cast him as my closer for years. He has indeed been better for Baltimore in 1989 than he was historically, despite now being rated as a one star player. Pena has completely bottomed out in 1989, going from a guy that saved 60 games in 1987-1988 to one that is 3-5 4.91 at the beginning of September, 1989. He'll be the Mets problem next year at least.

VanillaGorilla · 06-21-2013, 04:25 PM

My ringing endorsement is for OOTP!

The fact that you can choose 1 year/3 year/5 year recalc to suit your desire for play (and all the other user selections offered) is a huge part of what makes this game great.

For my purposes, I use recalc (have experimented with various lengths, and I have liked playing with them all) with player development turned on. I get players that are based on their history, but I also get to employ the OOTP player development engine which adds the "fictional realism" that I want.

From what I gather, here, it looks like you might prefer using the one year recalc with player development turned off which will result in year by year performance that will be tied to the actual corresponding years.

Enjoy whatever way you choose to play.

Darviathar · 06-22-2013, 02:37 AM

Quote:

Originally Posted by VanillaGorilla

My ringing endorsement is for OOTP!

The fact that you can choose 1 year/3 year/5 year recalc to suit your desire for play (and all the other user selections offered) is a huge part of what makes this game great.

For my purposes, I use recalc (have experimented with various lengths, and I have liked playing with them all) with player development turned on. I get players that are based on their history, but I also get to employ the OOTP player development engine which adds the "fictional realism" that I want.

From what I gather, here, it looks like you might prefer using the one year recalc with player development turned off which will result in year by year performance that will be tied to the actual corresponding years.

Enjoy whatever way you choose to play.

Where is the recalc option? I don't recall ever seeing a 5 year option, honestly. Can I turn player development off this far into a franchise? 1989 should be done in a day or two, and I'd like to change these settings before beginning 1990.

VanillaGorilla · 06-22-2013, 04:59 AM

Quote:

Originally Posted by Darviathar

Where is the recalc option? I don't recall ever seeing a 5 year option, honestly.

League Setup>>>Historical

There is a drop down under where you check the box for recalc which gives you the 1/3/5 option.

Quote:

Originally Posted by Darviathar

Can I turn player development off this far into a franchise? 1989 should be done in a day or two, and I'd like to change these settings before beginning 1990.

Yes, you can turn it off at any time.

However.....

I do not know if the development (or relative lack of development, in some cases) that has taken place for you current players will carry over into upcoming years for those players, and if it does, just how much impact it will have and for how long, once you deselect player development.

Darviathar · 06-22-2013, 04:14 PM

I see the recalc section with 1-3-5, but not "player development off" anywhere. Is this on another screen, or have to be done during spring training, etc?

VanillaGorilla · 06-22-2013, 05:14 PM

Game Setup>>>Player and Picture Options

There is a box to check for "Disable Player Development"

Darviathar · 06-25-2013, 04:09 PM

Switched to One recalc.

Recalc player ratings based on real stats after each year is checked. Double weight of current year stats is checked. Base Ratings based on : real stats, potential on remaining peak seasons. Rookie fielding ratings and pitcher stamina on current season.

I suppose I was wrong in assuming this would give me the same ratings as a replay league with player ratings optimized for single-season replay? I am seeing some ratings so poor that I feel obligated to go in and manually change them. Bobby Thigpen, who saved 57 games with an ERA of 1.83 in 1990 is rated as a one star closer. The computer seems to have a very low opinion of him in general: He has sported ERA's of 6.46, 4.68, and 7.52 in the past three seasons. Eric Davis had a down 1990 compared to to 1989, but I could swear the computer is assigning his ratings based on his dismal 1991 season! I obviously made the mistake of not optimizing stats for single season replay when I started this franchise back in 1983. Is there any way to remedy that?

06-15-2013, 07:23 PM	#1
Darviathar Minors (Single A) Join Date: Feb 2011 Posts: 82	Horrible relief pitching league-wide I've been working on replaying 1980's seasons, starting with 1983 and playing through 1987 in OOTP 13 with pretty accurate results. I moved my franchise to OOTP 14 in April and am now a little over half way through 1989. I am at wit's end with the horrible relief pitching around the league since I moved to OOTP. No lead seems to be safe for anyone, even with elite closers on the hill. Guys that historically had ERA's in the mid 2's are well into the high fours and fives across the board. I have gone in and increased reliever stamina several times in an attempt to keep relievers "fresh" so that they aren't always pitching tired, but it doesn't seem to help. Historically, 1988 and 1989 were two seasons in which pitching thrived, and offenses shouldn't be setting records. This reminds me of a version of OOTP that I played about a decade ago that saw the same thing happen. Does anyone have any suggestions as how to fix this? Last edited by Darviathar; 06-15-2013 at 07:26 PM.

06-15-2013, 09:53 PM	#2
Charlie Hough Hall Of Famer Join Date: May 2006 Posts: 3,625	Are you using single-season recalc or three-year recalc? Have you made sure that you're strictly using historical strategies, player creation modifiers, and other settings? What ballpark factors are you using? I suspect that there's a problem somewhere in your settings. Last edited by Charlie Hough; 06-15-2013 at 11:14 PM.

06-21-2013, 02:18 PM	#11
Darviathar Minors (Single A) Join Date: Feb 2011 Posts: 82	I suppose you guys have now offered a ringing endorsement for the one-year recalc feature. I don't know why I haven't been using it all these seasons. You're right about my 1988 bullpen: Howell, Orosco and Holton were all pretty bad in 1987 historically. However, Holton had been consistently excellent for me for 3 years somehow, so I guess I just assumed he would have an even better season in what was his career year. He "made the club" out of spring training in 1985 (I like to look at spring stats for guys that are marginal players and take the one who performed best to at least start the season) and somehow went 4-2 with a 1.46 ERA in 61 innings. Followed that up with 4-1 2.76 in 65 in 1986, and 2-5 2.57 in 84 in 1987. The computer manager had been urging me to cast him as my closer for years. He has indeed been better for Baltimore in 1989 than he was historically, despite now being rated as a one star player. Pena has completely bottomed out in 1989, going from a guy that saved 60 games in 1987-1988 to one that is 3-5 4.91 at the beginning of September, 1989. He'll be the Mets problem next year at least. Last edited by Darviathar; 06-21-2013 at 02:19 PM.

06-17-2013, 07:50 PM	#4
Charlie Hough Hall Of Famer Join Date: May 2006 Posts: 3,625	If you've changed league totals modifiers, I can almost guarantee that this is the source of your problem. These are not adjusted in the way that you would think intuitively. In fact, my understanding is that when you think you're lowering things, you're doing the opposite and vice versa Have you read the manual sections on these and how to modify them? Read the manual and find the section that addresses this counter-intuitive aspect of things. I bet that's the problem.

06-18-2013, 01:07 AM	#5
Darviathar Minors (Single A) Join Date: Feb 2011 Posts: 82	I read about the inverse relationship between totals and stats, but that isn't what I've changed. I've only changed the modifiers, so this shouldn't be the problem.

06-18-2013, 12:19 PM	#6
Charlie Hough Hall Of Famer Join Date: May 2006 Posts: 3,625	I still suspect that the problem has to do with things that you've modified. That's usually the prime suspect in any case of statistical output being as skewed as you're experiencing. But maybe this is something you can take up with the developers. Maybe you can post your league files through the usual support method so they can take a look.

06-18-2013, 04:38 PM	#7
JohnHoward All Star Reserve Join Date: Jun 2005 Posts: 624	I would say one contributing factor is definitely three-year recalc since you have the 1987 outlier season still being rolled into the recalc for 1988 and 1989. For certain seasons when moving forward you really should not count them in recalc. I think there are three seasons where this is true: 1930, 1968, and 1987. You should do a "cold start" with 1931, 1969, and 1988 (actual season totals and modifiers set to 1.000, no recalc) and limited recalc for the following two seasons.

06-20-2013, 08:33 PM	#9
Darviathar Minors (Single A) Join Date: Feb 2011 Posts: 82	The 1987 season was indeed an outlier, and ERA's around my league jumped into orbit in comparison to 1985 and 1986. Almost everyone came back to earth in 1988, other than Darryl Strawberry, who hit an astonishing 55 home runs. I managed to win Orel Hershiser's rightful 1988 Cy Young (no 59 scoreless inning streak though) with a season that was easily better than what he did historically, so it isn't like hitting in general was still inflated from 1987. I probably wouldn't be worried about this problem if it didn't seem to be an OOTP 14 (as opposed to the 5 seasons I ran in OOTP 13) phenomenon. I play every game, so I can really "see" just how shaky bullpen's are. The 1988 Dodgers had an excellent bullpen, with Howell (2.08), Pena (1.91), Holton (1.70), Orosco (2.72) and Crews (3.14). I didn't get anywhere near that kind of performance. These guys were so bad all season long that I finished 7 games out. Holton was completely useless all season (5.75), while Jay Howell lost the closer job in May after setting what I consider an unofficial league record for walking the leadoff batter in save opportunities. I played one of the early OOTP's a decade ago and recall bullpens being universally bad for no obvious reason. Just hoping that isn't the case here.

06-21-2013, 04:25 PM	#12
VanillaGorilla All Star Starter Join Date: Dec 2011 Posts: 1,371 Blog Entries: 2	My ringing endorsement is for OOTP! The fact that you can choose 1 year/3 year/5 year recalc to suit your desire for play (and all the other user selections offered) is a huge part of what makes this game great. For my purposes, I use recalc (have experimented with various lengths, and I have liked playing with them all) with player development turned on. I get players that are based on their history, but I also get to employ the OOTP player development engine which adds the "fictional realism" that I want. From what I gather, here, it looks like you might prefer using the one year recalc with player development turned off which will result in year by year performance that will be tied to the actual corresponding years. Enjoy whatever way you choose to play.

06-22-2013, 04:14 PM	#15
Darviathar Minors (Single A) Join Date: Feb 2011 Posts: 82	I see the recalc section with 1-3-5, but not "player development off" anywhere. Is this on another screen, or have to be done during spring training, etc?

06-22-2013, 05:14 PM	#16
VanillaGorilla All Star Starter Join Date: Dec 2011 Posts: 1,371 Blog Entries: 2	Game Setup>>>Player and Picture Options There is a box to check for "Disable Player Development"

06-25-2013, 04:09 PM	#17
Darviathar Minors (Single A) Join Date: Feb 2011 Posts: 82	Switched to One recalc. Recalc player ratings based on real stats after each year is checked. Double weight of current year stats is checked. Base Ratings based on : real stats, potential on remaining peak seasons. Rookie fielding ratings and pitcher stamina on current season. I suppose I was wrong in assuming this would give me the same ratings as a replay league with player ratings optimized for single-season replay? I am seeing some ratings so poor that I feel obligated to go in and manually change them. Bobby Thigpen, who saved 57 games with an ERA of 1.83 in 1990 is rated as a one star closer. The computer seems to have a very low opinion of him in general: He has sported ERA's of 6.46, 4.68, and 7.52 in the past three seasons. Eric Davis had a down 1990 compared to to 1989, but I could swear the computer is assigning his ratings based on his dismal 1991 season! I obviously made the mistake of not optimizing stats for single season replay when I started this franchise back in 1983. Is there any way to remedy that?