No longer able to use a certain year as base year in Random Debut.

[actionjackson]

Originally Posted by pstrickert

Sorry to hijack this thread. But let me add a quick note. I checked the real life BABIP for the Oakland pitching staff from 1970-75. Each year, they were number one or number two league-wide in lowest BABIP. So, did the A’s just get lucky six years in a row? The Oakland dynasty of the early ‘70s was nothing but luck?

Don't be too hard on yourself pstrickert. Technically, this discussion has been brought up in this thread before, so I wouldn't call it hijacking. Also, I'm not sure anybody can ever convince me that a low BABIP is 100% luck derived. I don't think we'll ever know how much credit to give the defense and how much credit to ascribe to luck, but I can pretty much guarantee you that there's not a 100% figure in either camp.

[Markus Heinsohn]

OK, I have added a new option, called "Lock league totals" for OOTP 19. If this is enabled, it will not touch the total stats but still recalc the modifiers (if enabled) prior to each season. That should properly accomplish what you are looking for

[Reed]

I was thinking of skipping 19 or waiting until mid season to buy but Markus just got me juiced to buy as soon as possible!

[actionjackson]

I've just run test sims in both OOTP16 and OOTP18 with the same settings (where possible, obviously that would exclude settings that have been added since OOTP16 finished it's development cycle) over a forty season span (1946-1985) setting the game up to auto-import LTMs for each season as they go whizzing past. I did not tie the stats to one particular season as this thread has been discussing. I wanted to see what would happen with a normal random debut setup.

I thought the results were interesting. It appears to me that OOTP18 runs quite a bit hotter than OOTP16 when it comes to extreme seasons. I'll leave it up to the reader to decide whether or not it's an issue. I used Real Stats with 3 year double weighted recalc for both of these test runs.

First the results for the OOTP16 test run. These numbers are the number of individual player seasons at or above the figure indicated for the particular metric. They also include instances where the number rounded up to the figure indicated:

AVG - .350 or higher - 41 player seasons
OBP - .450 or higher - 16 player seasons
SLG - .600 or higher - 59 player seasons
OPS - 1.050 or higher - 26 player seasons
WAR - 10.0 or higher - 12 player seasons

Now the results for the OOTP18 test run results. The same conditions apply as in the OOTP16 test run with rounding up etc.

AVG - .350 or higher - 70 player seasons
OBP - .450 or higher - 53 player seasons
SLG - .600 or higher - 97 player seasons
OPS - 1.050 or higher - 49 player seasons
WAR - 10.0 or higher - 34 player seasons

Now for the figures of the last player (rankings run from 1-100, so this is #100) in the rankings in each category in the OOTP18 test run, and how many players exceeded this value in the OOTP16 test run:

AVG - #100: .3458, only 55 players in OOTP16 managed to exceed this value
OBP - #100: .4362, only 46 players in OOTP16 managed to exceed this value
SLG - #100: .5972, only 65 players in OOTP16 managed to exceed this value
OPS - #100: 1.0148, only 52 players in OOTP16 managed to exceed this value
WAR - #100: 8.67, only 52 players in OOTP16 managed to exceed this value

There's no doubt in my mind that things appear to be running hotter in OOTP18 than OOTP16 when it comes to extreme seasons. More testing would help to back this up further, but it pretty much doesn't matter what stat you look at (pitching as well), there appear to be more extreme seasons in OOTP18 than OOTP16. My question is, what has changed that would make this so? The brains behind the game know better than this peon what could've changed between 16 and 18 that would make this so. So I leave it to you guys. I can do more tests if you like, but since this one shows values that are in line with what's being discussed in this thread, I think it's worth checking out.

In general the league-wide numbers look fine: 1968 is deadballish as it should be, and offense surges in expansion years due to watered down pitching as it should. Lots of green (as opposed to red) on the "Historical Simulation Accuracy" page, which suggests things are basically proceeding as they should on a league-wide basis.

[David Watts]

Originally Posted by actionjackson

I've just run test sims in both OOTP16 and OOTP18 with the same settings (where possible, obviously that would exclude settings that have been added since OOTP16 finished it's development cycle) over a forty season span (1946-1985) setting the game up to auto-import LTMs for each season as they go whizzing past. I did not tie the stats to one particular season as this thread has been discussing. I wanted to see what would happen with a normal random debut setup.

I thought the results were interesting. It appears to me that OOTP18 runs quite a bit hotter than OOTP16 when it comes to extreme seasons. I'll leave it up to the reader to decide whether or not it's an issue. I used Real Stats with 3 year double weighted recalc for both of these test runs.

First the results for the OOTP16 test run. These numbers are the number of individual player seasons at or above the figure indicated for the particular metric. They also include instances where the number rounded up to the figure indicated:

AVG - .350 or higher - 41 player seasons
OBP - .450 or higher - 16 player seasons
SLG - .600 or higher - 59 player seasons
OPS - 1.050 or higher - 26 player seasons
WAR - 10.0 or higher - 12 player seasons

Now the results for the OOTP18 test run results. The same conditions apply as in the OOTP16 test run with rounding up etc.

AVG - .350 or higher - 70 player seasons
OBP - .450 or higher - 53 player seasons
SLG - .600 or higher - 97 player seasons
OPS - 1.050 or higher - 49 player seasons
WAR - 10.0 or higher - 34 player seasons

Now for the figures of the last player (rankings run from 1-100, so this is #100) in the rankings in each category in the OOTP18 test run, and how many players exceeded this value in the OOTP16 test run:

AVG - #100: .3458, only 55 players in OOTP16 managed to exceed this value
OBP - #100: .4362, only 46 players in OOTP16 managed to exceed this value
SLG - #100: .5972, only 65 players in OOTP16 managed to exceed this value
OPS - #100: 1.0148, only 52 players in OOTP16 managed to exceed this value
WAR - #100: 8.67, only 52 players in OOTP16 managed to exceed this value

There's no doubt in my mind that things appear to be running hotter in OOTP18 than OOTP16 when it comes to extreme seasons. More testing would help to back this up further, but it pretty much doesn't matter what stat you look at (pitching as well), there appear to be more extreme seasons in OOTP18 than OOTP16. My question is, what has changed that would make this so? The brains behind the game know better than this peon what could've changed between 16 and 18 that would make this so. So I leave it to you guys. I can do more tests if you like, but since this one shows values that are in line with what's being discussed in this thread, I think it's worth checking out.

In general the league-wide numbers look fine: 1968 is deadballish as it should be, and offense surges in expansion years due to watered down pitching as it should. Lots of green (as opposed to red) on the "Historical Simulation Accuracy" page, which suggests things are basically proceeding as they should on a league-wide basis.

So I'm not the only one seeing this kind of stuff. Would still love to know how many more extreme bad seasons are occurring with 18, but that would take a lot of time to cipher out.. Just have to figure to make league numbers balance out, some guys have to be really bad as well.

[pstrickert]

This is helpful data, actionjackson. Glad to see you’re on the beta team. You know, of course, that you’ll have to do double the work since David Watts refuses to help us out.

[Reed]

I just simmed from 1901 to 1950 and had a lot of players hitting over 400. Hornsby hit 448 followed by a 476 season. I hope this is carefully checked for 19. The league wide stats were good though.

[actionjackson]

Originally Posted by Reed

I just simmed from 1901 to 1950 and had a lot of players hitting over 400. Hornsby hit 448 followed by a 476 season. I hope this is carefully checked for 19. The league wide stats were good though.

28 players have had a single season batting average of over .400 since 1876. Only 13 of those player seasons happened between 1901 and 1950. How many is a lot?

[David Watts]

Originally Posted by Reed

I just simmed from 1901 to 1950 and had a lot of players hitting over 400. Hornsby hit 448 followed by a 476 season. I hope this is carefully checked for 19. The league wide stats were good though.

How many hits did Hornsby have in the .476 season

[Reed]

Hornsby played 134 games and had 241 hits. 1924 season
68 player seasons at .400 or higher, some players like Cobb and Hornsby had multiple seasons. HRs and other stats seem reasonable.
I play 1 year recalc, no development engine, players miss season according to history and pretty much default settings. Under historical simulation accuracy it was green for AVG for every year.
I normally fast sim a league before I play it and didn’t see this problem before so I wonder if it has something to do with the latest build or I guess it’s possible I screwed something up, but if I did, I haven’t found it yet.

[David Watts]

Originally Posted by Reed

Hornsby played 134 games and had 241 hits. 1924 season
68 player seasons at .400 or higher, some players like Cobb and Hornsby had multiple seasons. HRs and other stats seem reasonable.
I play 1 year recalc, no development engine, players miss season according to history and pretty much default settings. Under historical simulation accuracy it was green for AVG for every year.
I normally fast sim a league before I play it and didn’t see this problem before so I wonder if it has something to do with the latest build or I guess it’s possible I screwed something up, but if I did, I haven’t found it yet.

I had Hornsby hit .485 in a real lineups/real transaction fast sim. 1 year recalc as well.

[actionjackson]

Originally Posted by Reed

Hornsby played 134 games and had 241 hits. 1924 season
68 player seasons at .400 or higher, some players like Cobb and Hornsby had multiple seasons. HRs and other stats seem reasonable.
I play 1 year recalc, no development engine, players miss season according to history and pretty much default settings. Under historical simulation accuracy it was green for AVG for every year.
I normally fast sim a league before I play it and didn’t see this problem before so I wonder if it has something to do with the latest build or I guess it’s possible I screwed something up, but if I did, I haven’t found it yet.

68? Whoa! That's over five times as often as real life (13). This just further backs up what we've been seeing: League-wide totals are fine, but something in the distribution of those league-wide totals is off.

[JaBurns]

Thank you to everyone in this thread for your comments and insights; I very much enjoy playing historical random debut but have not nearly the experience or the talents of you gentlemen. I play a number of seasons of random debut always starting in 1901 and let all settings progress yearly. I usually get 10 to 15 seasons in then see something that gives me a “better” way to run my league. Off I go and start again until my next revelation, as Mr. Watts says rinse and repeat but mine is with an entirely new setup. Any improvements you can suggest are more than welcome, just started my latest and greatest idea in 1901 this morning. Hope you have all the bugs worked out by 22 March 18 for OOTP 19.

[actionjackson]

Bumpity, bump, bump, bump. Any word as to whether "hot" offense will get looked at? Markus' new addition to the Stats & AI page should be really helpful for us weirdos that like to do this, but individual seasons are still running crazy whether you tie stats to a historical season or play historical straight up as it comes, as my post details two posts below the post where Markus announced said addition.

[Lukas Berger]

Originally Posted by actionjackson

Bumpity, bump, bump, bump. Any word as to whether "hot" offense will get looked at? Markus' new addition to the Stats & AI page should be really helpful for us weirdos that like to do this, but individual seasons are still running crazy whether you tie stats to a historical season or play historical straight up as it comes, as my post details two posts below the post where Markus announced said addition.

Yes, it's getting looked at for sure.

Markus has already made some changes that in theory should fix this, and we'll test them out as thoroughly as possible and see what we find.

It's possible what we find won't be the same as what you're seeing if this is primarily a settings thing. We'll do the best we can though.

[David Watts]

Originally Posted by Lukas Berger

Yes, it's getting looked at for sure.

Markus has already made some changes that in theory should fix this, and we'll test them out as thoroughly as possible and see what we find.

It's possible what we find won't be the same as what you're seeing if this is primarily a settings thing. We'll do the best we can though.

Can I ask you this, are y'all looking into the problem of players having way too many hits? This occurs even in straight historical play with development turned off. See my detailed posts in recalc thread. 240-260 hits in a OOTP season is pretty much a regular occurrence with 18. This of course leads to way too many guys finishing with high batting averages. .400 isn't even something to celebrate.

[Lukas Berger]

Originally Posted by David Watts

Can I ask you this, are y'all looking into the problem of players having way too many hits? This occurs even in straight historical play with development turned off. See my detailed posts in recalc thread. 240-260 hits in a OOTP season is pretty much a regular occurrence with 18. This of course leads to way too many guys finishing with high batting averages. .400 isn't even something to celebrate.

Yeah, we'll look at everything regarding statistical distribution in both historical and regular play for sure.

[pstrickert]

Originally Posted by David Watts

Can I ask you this, are y'all looking into the problem of players having way too many hits? This occurs even in straight historical play with development turned off. See my detailed posts in recalc thread. 240-260 hits in a OOTP season is pretty much a regular occurrence with 18. This of course leads to way too many guys finishing with high batting averages. .400 isn't even something to celebrate.

I’ll definitely be scrutinizing this area during beta, David.

[David Watts]

Originally Posted by pstrickert

I’ll definitely be scrutinizing this area during beta, David.

So you are back on beta? I thought I heard you weren't doing it this year. Oh man, this gives me a big boost in the hope department.

[pstrickert]

At the risk of being called Brett Favre, yes, I unretired.