High negro player values

sprague · 05-02-2022, 06:10 AM

While I have been commenting on the below thread in the bug reports section to show the massively high values for negro pitchers
https://forums.ootpdevelopments.com/...d.php?t=337005

it is also batters that seem to be gaining a big value jump. Here I want to show you the list for 1920 (any year can be used) to see the majority of the best players all be from the negro leagues.

And then some of the edit pages to see just how high rated (especially those potential ratings) they are

sprague · 05-02-2022, 06:12 AM

Next 3, look at the values for Heavy Johnson, that is far far better than Ty Cobb's potentials as we will see.

sprague · 05-02-2022, 06:15 AM

And the top 3 mlb players (why does hornsby have a higher potential rating than Cobb surprises me).

The negro players used to be VERY underated, now it seems the new formula has made them slightly over rated. They should be good players with high values but the formula right now I think is tweeking them just a bit much.

luckymann · 05-02-2022, 06:27 AM

Have you simmed these out to say 1960 and seen how the stat outputs come out?

I did a full 1-year recalc sim from 1920-2020, which I commented on HERE. Compared to my hand-curated profiles in a full 1890-2020 sim I did on 22, these numbers are generally lower and far fewer players made the HoF, while none led any all-time stat category. I'd be interested to see how yours pan out.

As LSt has pointed out in the other thread, the obvious next step is for the stats to be converted to Seamheads. But for now I think we just have to be thankful they are no longer treated as scrubs and are at least usable in a game for most OOTPers.

David Watts · 05-02-2022, 09:18 AM

Quote:

Originally Posted by luckymann

Have you simmed these out to say 1960 and seen how the stat outputs come out?

I did a full 1-year recalc sim from 1920-2020, which I commented on HERE. Compared to my hand-curated profiles in a full 1890-2020 sim I did on 22, these numbers are generally lower and far fewer players made the HoF, while none led any all-time stat category. I'd be interested to see how yours pan out.

As LSt has pointed out in the other thread, the obvious next step is for the stats to be converted to Seamheads. But for now I think we just have to be thankful they are no longer treated as scrubs and are at least usable in a game for most OOTPers.

Off topic, but figured it's easier to ask you here. How long does it take for the game to sim out seasons when you get to the point where there are like 30 to 40 leagues involved? By the time you reached the 70's and 80's was the your sim creeping? Just curious, thanks.

luckymann · 05-02-2022, 09:51 AM

Quote:

Originally Posted by David Watts

Off topic, but figured it's easier to ask you here. How long does it take for the game to sim out seasons when you get to the point where there are like 30 to 40 leagues involved? By the time you reached the 70's and 80's was the your sim creeping? Just curious, thanks.

It certainly slowed down noticeably as it progressed. I'm on a 2021 MacBook Pro with plenty of juice and it was taking roughly 5 mins per season toward the end. Another reason why I'd love to see these guys able to be used without having to invoke full historical minors.

sprague · 05-02-2022, 10:02 AM

Quote:

Originally Posted by luckymann

It certainly slowed down noticeably as it progressed. I'm on a 2021 MacBook Pro with plenty of juice and it was taking roughly 5 mins per season toward the end. Another reason why I'd love to see these guys able to be used without having to invoke full historical minors.

Free the Minor League Database has been my rallying cry since it came out.
There is no reason for it not to be editable for leagues, subleagues, teams, minors just like the regular non-minor league DB

David Watts · 05-02-2022, 10:34 AM

I was really excited back when they added historical minors. I quickly realized I didn't really want historical minors, but instead, I wanted the extra players, so I could do historical whatif type leagues and use the normal or high modern day injury settings. Historical minors is so darn massive.

I'd also love to know whether recalc gets done in any form on the players in the historical minors DB, or if recalc is strictly for the major league players? It seems like the weaken and adjust settings would pretty much ruin a lot of minor league player.

LansdowneSt · 05-02-2022, 01:56 PM

Quote:

Originally Posted by sprague

Free the Minor League Database

That sir, is a very good, very OOTP-niche bumper sticker. I like it.

I've been thinking about the NeL and the MiLB csv files ever since David mentioned separating them it in the other thread. I had been thinking about a third csv as the answer but what if the NeL players just went on the MLB master sheet as players? If the color line is on there is no "harm" to historic pre-Jackie Robinson sims. The NeL FAs have nowhere to play if the color line is on and there is no NeL toggled on. So, no effect on play as it currently is.

If the NeL league's toggle is on, they all go there as they will. I'm not concerned about them being FAs in such a scenario. I think the NeL teams came and went so much that getting them on start of and end of season rosters is a challenge anyway. If pre-1900 is a free for all, I'd trade this free for all scenario for having them in RDs and ensuring equal treatment of the partial stats nerf or any other (if any) unique treatment of players on that MLB csv.

If the color line is off and there is no NeL, the NeL'ers come into the AL/NL as Free Agents. If the color line is off and the NeL is there, it's a place for surplus players to play and have their contracts purchased from. But having the NeL'ers on the Master sheet would allow them on Random Debuts.

At some point when things calm down I expect all this will be looked at... even while those "Real Life" stats informing their abilities get updated to Seamhead quality.

Lastly, like luckymann said, I've found they play more evenly than the one-year look at star potentials indicate but I like 3- and 5-year recalcs to flatten them out a bit with retire to history so no one runs too off for too long. It is true that there are some definite shooting star players taking Cy Youngs but for now I think that is what it is. I wish the stats would be adjusted in a 23 patch late in the year so we could see how that helps. Maybe that will come, fingers crossed.

As for time of the sims, be sure to turn off all the stats you don't look at come the end. No box scores, no 3D replays, all of that off. For a 1903-1970-ish (I wait until full HOF balloting for them has had time to run), it takes me 4-5 hours. I tend to run those checks overnight or while at work.

sprague · 05-02-2022, 03:53 PM

Hi L,
That was a terrific answer and idea you made in the negro thread. I have long wanted the minor league players available for any format. You seem to be on the right track.
The big problem when it comes to the values for players is that the game is generating values based on actual stats that are in the database. Which means each league must have a MLB equivalent adjustment when creating the values.
Spritze's DB that I am sure you know well, had not real stat values for non MLB players or years, but were equivalents that the game would turn and create as if they were in an MLB environment.
I wonder- best of both worlds here.
The DB has one set of columns for real stats of the player, that is only to come into the game as the player history, but then is a second column of "stats" that functions like what Chuck's dB does, values for the player generation, and the AI only looks at these values. they take the real stats only as an upload for history.

Just thinking aloud

LansdowneSt · 05-02-2022, 04:58 PM

Yup, I'm familiar with how Spritze set up his NeL imported players to use Major League Equivalents (MLEs) but alas, I must confess that I've never run a game off his database or imported individuals from it - though I know many people have. Overall, I think MLE's are tricky. I've looked at some and while I've posted this link before, I will again in case new folks come across this thread. These MLEs are the one's I'm most familiar with:

https://horsehidedragnet.wordpress.com/negro-leagues/

I consider myself a practical fellow but admit my MLE aversion is uncharacteristically traditional. It's not that I'm not open to it and it's possible that that is how it will end up for the betterment of gameplay. I reckon I'd be okay with that under-the-hood adjustment if, as you say, the Seamhead stats still populated the history page. Overall though, I think that I'd first like a run at seeing how well those updated Seamhead stats would inform the player abilities rather than the outdated stats.

If there are a few players on the extremes post-SH stat update, that might be okay in the end. Afterall, no one feels the need to do an MLE for Ross Barnes or Old Hoss despite the fact the rules of baseball back then - as this fun site shows (https://www.baseball-almanac.com/rulechng.shtml) - were so different from our conception of the game. So much so, MLB - unlike bRef and OOTP - doesn't even recognize the NA as a major league. So, if I can live with player abilities coming from MLB-credited underhand pitching and 9 balls, I can live with a few odd stat ratios from NeL'ers in the game. That said, some allowance will need to be made from missing stats including ones you mentioned like pitcher HR/9 or batter K's, etc. where the stats are incomplete, can't just be left missing and so will need stand-ins. But maybe the game does that some for the 19th century players as it is and there's a process for it already.

Thinking aloud a bit myself and being practical about the order of future events, if the Seamhead stats are needed if only for the history tab, they'll need to be gotten either way. All the more reason to let them run, informing the players before turning to the work of MLEs. Esp. as one considers that Spritze's MLE's at the moment are based off the current in-game NeL stats and would also need to be updated to reflect current research/numbers.

Anyway, I understand the appeal an MLE approach has, even if I take the other side for now.

Well, I'm back from work and time to check the 1903-2000 sim that ran while I was out. Color line dropped - this time 5-year recalc instead of my default 3-, and DHs enabled for an additional offensive slot in both leagues...

luckymann · 05-03-2022, 08:23 AM

Quote:

Originally Posted by LansdowneSt

Yup, I'm familiar with how Spritze set up his NeL imported players to use Major League Equivalents (MLEs) but alas, I must confess that I've never run a game off his database or imported individuals from it - though I know many people have. Overall, I think MLE's are tricky. I've looked at some and while I've posted this link before, I will again in case new folks come across this thread. These MLEs are the one's I'm most familiar with:

https://horsehidedragnet.wordpress.com/negro-leagues/

I consider myself a practical fellow but admit my MLE aversion is uncharacteristically traditional. It's not that I'm not open to it and it's possible that that is how it will end up for the betterment of gameplay. I reckon I'd be okay with that under-the-hood adjustment if, as you say, the Seamhead stats still populated the history page. Overall though, I think that I'd first like a run at seeing how well those updated Seamhead stats would inform the player abilities rather than the outdated stats.

If there are a few players on the extremes post-SH stat update, that might be okay in the end. Afterall, no one feels the need to do an MLE for Ross Barnes or Old Hoss despite the fact the rules of baseball back then - as this fun site shows (https://www.baseball-almanac.com/rulechng.shtml) - were so different from our conception of the game. So much so, MLB - unlike bRef and OOTP - doesn't even recognize the NA as a major league. So, if I can live with player abilities coming from MLB-credited underhand pitching and 9 balls, I can live with a few odd stat ratios from NeL'ers in the game. That said, some allowance will need to be made from missing stats including ones you mentioned like pitcher HR/9 or batter K's, etc. where the stats are incomplete, can't just be left missing and so will need stand-ins. But maybe the game does that some for the 19th century players as it is and there's a process for it already.

Thinking aloud a bit myself and being practical about the order of future events, if the Seamhead stats are needed if only for the history tab, they'll need to be gotten either way. All the more reason to let them run, informing the players before turning to the work of MLEs. Esp. as one considers that Spritze's MLE's at the moment are based off the current in-game NeL stats and would also need to be updated to reflect current research/numbers.

Anyway, I understand the appeal an MLE approach has, even if I take the other side for now.

Well, I'm back from work and time to check the 1903-2000 sim that ran while I was out. Color line dropped - this time 5-year recalc instead of my default 3-, and DHs enabled for an additional offensive slot in both leagues...

I am a more staunch fan of the MLEs than my friend LSt, but as I have said on numerous occasions, I feel Eric is a bit too harsh in a lot of his adjustments. But I also feel those SH raw stats are probably a bit too strong in an MLB context, and also have more holes than I like to see. Ergo why I hybridize them in my curated profiles.

But even with my acknowledgment of their general viability, I don't believe MLEs should form a part of the general NeL player stats. To bring subjectivity into it on a game-wide basis is fraught with danger. Raw SH stats are what's needed IMHO, along with these players being made fully-fledged MLB players and included in that DB so they can be included in RD and used without having to invoke full historical MiLB. I have no doubt this should be the next move undertaken.

LSt, I'd love to see some of those 3-yr recalc sim leaderboards if you get the chance.

sprague · 05-03-2022, 02:01 PM

Thanks to all of you for joining in on this discussion. It is important to wind up with a true accurate sim of baseball history.

That the negro values are high by themselves are not a problem, if the only league running was the negro league. That is why it is fine players in 1871 have high values, there is no one else they are matched up to.
This becomes an issue when one type of league combines with players from another league used as the over all baseline- in this case MLB

I have created a few of these myself over the years. Where do i have Pop Lloyd compared to Honus Wagner, or Satchel Paige to Lefty Grove. I have used a few different stat equivalent outlets there to help guide the values I have used.
It would be like if I just took straight hockey stats for the Soviet Leagues of the 1970's and used the wrong equivalents. How good of a defenseman was Vladimir Lutchenko if he played in the NHL for example, and you need a variety of stats (from russia, international, canada cups, then the right tweek of it all to produce the honest result) The negro players will require something similar.

The negro players should be good, but if the top 15 batters turn out to be negro players then its out of whack for the reality of the day. So that is the challenge, how to get the AI to create equivalents for them that will showcase their skills, let them be the type of good or great player in an MLB environment yet not overtake what would be the likely historical reality. Josh Gibson should lead in home runs, but Jimmy Foxx or Hank Greenberg should wind up 2 3 for example, not 9 or 10.
You get my drift

LansdowneSt · 05-03-2022, 02:40 PM

Yup, agree with all that, sprague.

After things settle with 23's release, I'd like to learn more from the development team on if it treats players that have "MLB" stats but still reside on the MiLB csv file differently when creating the ratings. I think we all agree the NeL real stats are in desperate need of upgrade, but even I'll put that to the side and look at... say, Heliodoro Diaz.

In my recent sim (recalc every year, 5-year span, retire to history), Helio won the 1928 Cy Young for the Red Sox going 24-8, 3.04 ERA, 302 IP, 132K, 3.5 BB/9 and 3.9 K/9 but that same year's OOTP "real life stats" have him at 3-14, 4.56 ERA, 140 IP, 100K, 3.5 BB/9 and 6.4 K/9. His surrounding years aren't spectacular either. That was league average for that NeL season.

I'm not sure if I used bRef's Stathead and looked up a comparable pitcher in the AL or NL for those real stats, I'd get someone that ought to be a CY Young winner. It's a sim, so there's a chance I would. But even anecdotally, I think something isn't being considered by the AI. By that I don't mean a league quality dampener for the NeL. I mean, if Helio pitched that line for the Browns or the Sox in real life and the AI was converting those legit MLB stats to game performance... I'm not sure he'd be a CY Young winner. So, I'm inclined to rack my brain thinking of a technical AI translation problem before moving on to it being an NeL problem. For example, the same issue could arise if we took the Pacific Coast League and changed those stats to major league. Would the AI treat the PCL stat line X differently than the MLB .csv stat line Y... even if X = Y. Know what I mean?

David Watts · 05-03-2022, 03:04 PM

Does anyone know for sure that recalc gets run on the players outside the Major League DB?

LansdowneSt · 05-03-2022, 03:21 PM

Quote:

Originally Posted by David Watts

Does anyone know for sure that recalc gets run on the players outside the Major League DB?

I'm not sure. That and the "make bad" thresholds are the first places to look at though. They might only be applied to stats in the MLB csv files and the "levels below" and development engine alone might have been doing the heavy lifting for those in the MiLB folders. But that's what I too was getting at. Is there something the AI does to the MLB players as a matter of course because of where the data sits vice the now-MLB-level but MiLB csv guys.

Hoiles · 05-03-2022, 10:05 PM

Did some further sleuthing and made my way onto the thread. It's pretty evident the NeL pitchers get a huge boost to their movement and control ratings for some reason.

Looking at the single-season 1927 import at defaults, (single season to avoid effects of other seasons and recalcs)

Lefty Grove:
Imported resulting stats (in a modern day MLB environment):
227IP, 6 HR, 64 BB, 157 SO., 3.03 FIP
Real-life (pretty reliable)
262IP, 6 HR, 79 BB, 174 SO, 3.04 FIP

Heliodoro Diaz:
Imported resulting stats (in a modern day MLB environment):
231IP, 0 HR, 46 BB, 175 SO., 2.35 FIP
"Real-life" (according to OOTP database):
172IP, 6 HR, 56 BB, 112 SO, 3.36 FIP

Grove's stats are very close to his real-life ones. Diaz actually had a solid year IRL, but gets a huge boost in ratings based on the assumption that he wouldn't allow a home run in a full season. In fact, 22 out of the 64 NeL starters had a 80 movement rating, whereas 1 of the 193 MLB starters had a 75 movement rating (none 80). The NeL starters' control ratings were also noticeably higher, though to a lesser extent. The NeL hitters don't get a similar obvious boost.

Looking at the league totals in the Stats and AI section, the MLB and NeL had pretty comparable league stats (in fact, MLB had slightly more offense, so Grove should have gotten a bit more of a boost). So it doesn't seem to be a normalization vs league average type of thing.

I suspect we can get a better MLB/NeL balance by fiddling with the adjust/weaken settings, though that might affect the sim quality of the individual leagues themselves.

luckymann · 05-03-2022, 11:58 PM

So I'm just having a bit of a muck about with 23 after the patch and thought I'd bring things back a notch.

I have run a sim from 1915 to 1948 with the color bar on and just focusing on NeL. 5-year recalc / double-weight / development OFF / TCR 10. With this - as with all my sims - I have set hitter "make bad" to 120/100 and pitcher "make bad" to 25/18.

Here's how the leaderboards came out:

Nothing obviously untoward there. Although I'd love to know how, if the random LTMs that get applied to minor leagues are in effect here, how is the game expected to replicate NeL seasons with even the remotest level of realism and accuracy? No doubt you can just enter them in manually and do an autocalc if you feel so inclined, but still...

For mine, among the many questions I have, that of how recalc is applied is right at the top of the list. So the only way I can think of to at least get some idea in the absence of any official word is to look at the ratings of a few players' scouting from this sim.

So here is Charlie Smith under 5-yr / double-weight. Interesting the actual report says he is a pitcher.

He shows no MLB level stats in his IRL tab. It looks like he got two recalcs in '25 and '26 then nothing. Here are his in-game IRLs.

And here's Gentry Jessup (who also, it seems, gets imported automatically a second time as "Joseph" in addition to "Gentry". I have chosen these two because their short careers allow us to see the range in one screenshot.

He also seems to get just two OSA updates in each of his first two seasons, then nothing.

Here are his IRLs as per the game.

What I'd like to know here is, if dev is OFF and no recalc is being done, how is he being scouted from that point on to generate his in-game ratings? Or am I reading this report incorrectly?

LansdowneSt · 05-04-2022, 12:33 AM

Quote:

Originally Posted by Hoiles

Diaz actually had a solid year IRL, but gets a huge boost in ratings based on the assumption that he wouldn't allow a home run in a full season. In fact, 22 out of the 64 NeL starters had a 80 movement rating, whereas 1 of the 193 MLB starters had a 75 movement rating (none 80). The NeL starters' control ratings were also noticeably higher, though to a lesser extent.

I haven't looked at control in particular but as sprague pointed out, the HR/9 is a hole in the records so that's the core of movement issue. More than a third of the NeL pitchers, eh? Well, that sure helps. I'm sure there's a rationale that can be applied to fill that HR/9 hole in the records that can be done the same time the stats/bio info are updated.

italyprof · 05-09-2022, 06:51 AM

Quote:

Originally Posted by sprague

Free the Minor League Database has been my rallying cry since it came out.
There is no reason for it not to be editable for leagues, subleagues, teams, minors just like the regular non-minor league DB

Totally agree. We would like to be able to:

1. Play ONLY one or more minor leagues without majors, and to he able to configure the saved League as we choose as a baseball world.

2. Have individual players available in the draft for normal replays and for random debut if we want, by just clicking on an option.

05-02-2022, 06:10 AM	#1
sprague All Star Starter Join Date: Aug 2011 Posts: 1,947	High negro player values While I have been commenting on the below thread in the bug reports section to show the massively high values for negro pitchers https://forums.ootpdevelopments.com/...d.php?t=337005 it is also batters that seem to be gaining a big value jump. Here I want to show you the list for 1920 (any year can be used) to see the majority of the best players all be from the negro leagues. And then some of the edit pages to see just how high rated (especially those potential ratings) they are Attached Images

05-02-2022, 06:12 AM	#2
sprague All Star Starter Join Date: Aug 2011 Posts: 1,947	Next 3, look at the values for Heavy Johnson, that is far far better than Ty Cobb's potentials as we will see. Attached Images

05-02-2022, 06:15 AM	#3
sprague All Star Starter Join Date: Aug 2011 Posts: 1,947	And the top 3 mlb players (why does hornsby have a higher potential rating than Cobb surprises me). The negro players used to be VERY underated, now it seems the new formula has made them slightly over rated. They should be good players with high values but the formula right now I think is tweeking them just a bit much. Attached Images

05-02-2022, 06:27 AM	#4
luckymann Hall Of Famer Join Date: Nov 2019 Posts: 13,012	Have you simmed these out to say 1960 and seen how the stat outputs come out? I did a full 1-year recalc sim from 1920-2020, which I commented on HERE. Compared to my hand-curated profiles in a full 1890-2020 sim I did on 22, these numbers are generally lower and far fewer players made the HoF, while none led any all-time stat category. I'd be interested to see how yours pan out. As LSt has pointed out in the other thread, the obvious next step is for the stats to be converted to Seamheads. But for now I think we just have to be thankful they are no longer treated as scrubs and are at least usable in a game for most OOTPers. __________________ HISTORICAL DO-OVERS A'S RED SOX DODGERS CUSTOM SAVES ECLIPSE LEAGUE MOON SHOT LEAGUE EVERYMAN LEAGUE GULF LEAGUE USBA Last edited by luckymann; 05-02-2022 at 06:29 AM.

05-02-2022, 10:34 AM	#8
David Watts Hall Of Famer Join Date: Apr 2002 Location: Looking for a place called Leehofooks Posts: 9,597 Infractions: 0/1 (1)	I was really excited back when they added historical minors. I quickly realized I didn't really want historical minors, but instead, I wanted the extra players, so I could do historical whatif type leagues and use the normal or high modern day injury settings. Historical minors is so darn massive. I'd also love to know whether recalc gets done in any form on the players in the historical minors DB, or if recalc is strictly for the major league players? It seems like the weaken and adjust settings would pretty much ruin a lot of minor league player. Last edited by David Watts; 05-02-2022 at 12:20 PM.

05-02-2022, 03:53 PM	#10
sprague All Star Starter Join Date: Aug 2011 Posts: 1,947	Hi L, That was a terrific answer and idea you made in the negro thread. I have long wanted the minor league players available for any format. You seem to be on the right track. The big problem when it comes to the values for players is that the game is generating values based on actual stats that are in the database. Which means each league must have a MLB equivalent adjustment when creating the values. Spritze's DB that I am sure you know well, had not real stat values for non MLB players or years, but were equivalents that the game would turn and create as if they were in an MLB environment. I wonder- best of both worlds here. The DB has one set of columns for real stats of the player, that is only to come into the game as the player history, but then is a second column of "stats" that functions like what Chuck's dB does, values for the player generation, and the AI only looks at these values. they take the real stats only as an upload for history. Just thinking aloud

05-02-2022, 04:58 PM	#11
LansdowneSt Global Moderator Join Date: Feb 2004 Location: From Duxbury, Mass residing Baltimore Posts: 7,167	Yup, I'm familiar with how Spritze set up his NeL imported players to use Major League Equivalents (MLEs) but alas, I must confess that I've never run a game off his database or imported individuals from it - though I know many people have. Overall, I think MLE's are tricky. I've looked at some and while I've posted this link before, I will again in case new folks come across this thread. These MLEs are the one's I'm most familiar with: https://horsehidedragnet.wordpress.com/negro-leagues/ I consider myself a practical fellow but admit my MLE aversion is uncharacteristically traditional. It's not that I'm not open to it and it's possible that that is how it will end up for the betterment of gameplay. I reckon I'd be okay with that under-the-hood adjustment if, as you say, the Seamhead stats still populated the history page. Overall though, I think that I'd first like a run at seeing how well those updated Seamhead stats would inform the player abilities rather than the outdated stats. If there are a few players on the extremes post-SH stat update, that might be okay in the end. Afterall, no one feels the need to do an MLE for Ross Barnes or Old Hoss despite the fact the rules of baseball back then - as this fun site shows (https://www.baseball-almanac.com/rulechng.shtml) - were so different from our conception of the game. So much so, MLB - unlike bRef and OOTP - doesn't even recognize the NA as a major league. So, if I can live with player abilities coming from MLB-credited underhand pitching and 9 balls, I can live with a few odd stat ratios from NeL'ers in the game. That said, some allowance will need to be made from missing stats including ones you mentioned like pitcher HR/9 or batter K's, etc. where the stats are incomplete, can't just be left missing and so will need stand-ins. But maybe the game does that some for the 19th century players as it is and there's a process for it already. Thinking aloud a bit myself and being practical about the order of future events, if the Seamhead stats are needed if only for the history tab, they'll need to be gotten either way. All the more reason to let them run, informing the players before turning to the work of MLEs. Esp. as one considers that Spritze's MLE's at the moment are based off the current in-game NeL stats and would also need to be updated to reflect current research/numbers. Anyway, I understand the appeal an MLE approach has, even if I take the other side for now. Well, I'm back from work and time to check the 1903-2000 sim that ran while I was out. Color line dropped - this time 5-year recalc instead of my default 3-, and DHs enabled for an additional offensive slot in both leagues...

05-03-2022, 02:01 PM	#13
sprague All Star Starter Join Date: Aug 2011 Posts: 1,947	Thanks to all of you for joining in on this discussion. It is important to wind up with a true accurate sim of baseball history. That the negro values are high by themselves are not a problem, if the only league running was the negro league. That is why it is fine players in 1871 have high values, there is no one else they are matched up to. This becomes an issue when one type of league combines with players from another league used as the over all baseline- in this case MLB I have created a few of these myself over the years. Where do i have Pop Lloyd compared to Honus Wagner, or Satchel Paige to Lefty Grove. I have used a few different stat equivalent outlets there to help guide the values I have used. It would be like if I just took straight hockey stats for the Soviet Leagues of the 1970's and used the wrong equivalents. How good of a defenseman was Vladimir Lutchenko if he played in the NHL for example, and you need a variety of stats (from russia, international, canada cups, then the right tweek of it all to produce the honest result) The negro players will require something similar. The negro players should be good, but if the top 15 batters turn out to be negro players then its out of whack for the reality of the day. So that is the challenge, how to get the AI to create equivalents for them that will showcase their skills, let them be the type of good or great player in an MLB environment yet not overtake what would be the likely historical reality. Josh Gibson should lead in home runs, but Jimmy Foxx or Hank Greenberg should wind up 2 3 for example, not 9 or 10. You get my drift

05-03-2022, 02:40 PM	#14
LansdowneSt Global Moderator Join Date: Feb 2004 Location: From Duxbury, Mass residing Baltimore Posts: 7,167	Yup, agree with all that, sprague. After things settle with 23's release, I'd like to learn more from the development team on if it treats players that have "MLB" stats but still reside on the MiLB csv file differently when creating the ratings. I think we all agree the NeL real stats are in desperate need of upgrade, but even I'll put that to the side and look at... say, Heliodoro Diaz. In my recent sim (recalc every year, 5-year span, retire to history), Helio won the 1928 Cy Young for the Red Sox going 24-8, 3.04 ERA, 302 IP, 132K, 3.5 BB/9 and 3.9 K/9 but that same year's OOTP "real life stats" have him at 3-14, 4.56 ERA, 140 IP, 100K, 3.5 BB/9 and 6.4 K/9. His surrounding years aren't spectacular either. That was league average for that NeL season. I'm not sure if I used bRef's Stathead and looked up a comparable pitcher in the AL or NL for those real stats, I'd get someone that ought to be a CY Young winner. It's a sim, so there's a chance I would. But even anecdotally, I think something isn't being considered by the AI. By that I don't mean a league quality dampener for the NeL. I mean, if Helio pitched that line for the Browns or the Sox in real life and the AI was converting those legit MLB stats to game performance... I'm not sure he'd be a CY Young winner. So, I'm inclined to rack my brain thinking of a technical AI translation problem before moving on to it being an NeL problem. For example, the same issue could arise if we took the Pacific Coast League and changed those stats to major league. Would the AI treat the PCL stat line X differently than the MLB .csv stat line Y... even if X = Y. Know what I mean?

05-03-2022, 03:04 PM	#15
David Watts Hall Of Famer Join Date: Apr 2002 Location: Looking for a place called Leehofooks Posts: 9,597 Infractions: 0/1 (1)	Does anyone know for sure that recalc gets run on the players outside the Major League DB?

05-03-2022, 10:05 PM	#17
Hoiles All Star Reserve Join Date: Dec 2001 Location: Canada Posts: 574	Did some further sleuthing and made my way onto the thread. It's pretty evident the NeL pitchers get a huge boost to their movement and control ratings for some reason. Looking at the single-season 1927 import at defaults, (single season to avoid effects of other seasons and recalcs) Lefty Grove: Imported resulting stats (in a modern day MLB environment): 227IP, 6 HR, 64 BB, 157 SO., 3.03 FIP Real-life (pretty reliable) 262IP, 6 HR, 79 BB, 174 SO, 3.04 FIP Heliodoro Diaz: Imported resulting stats (in a modern day MLB environment): 231IP, 0 HR, 46 BB, 175 SO., 2.35 FIP "Real-life" (according to OOTP database): 172IP, 6 HR, 56 BB, 112 SO, 3.36 FIP Grove's stats are very close to his real-life ones. Diaz actually had a solid year IRL, but gets a huge boost in ratings based on the assumption that he wouldn't allow a home run in a full season. In fact, 22 out of the 64 NeL starters had a 80 movement rating, whereas 1 of the 193 MLB starters had a 75 movement rating (none 80). The NeL starters' control ratings were also noticeably higher, though to a lesser extent. The NeL hitters don't get a similar obvious boost. Looking at the league totals in the Stats and AI section, the MLB and NeL had pretty comparable league stats (in fact, MLB had slightly more offense, so Grove should have gotten a bit more of a boost). So it doesn't seem to be a normalization vs league average type of thing. I suspect we can get a better MLB/NeL balance by fiddling with the adjust/weaken settings, though that might affect the sim quality of the individual leagues themselves. Last edited by Hoiles; 05-03-2022 at 10:06 PM.

05-03-2022, 11:58 PM	#18
luckymann Hall Of Famer Join Date: Nov 2019 Posts: 13,012	So I'm just having a bit of a muck about with 23 after the patch and thought I'd bring things back a notch. I have run a sim from 1915 to 1948 with the color bar on and just focusing on NeL. 5-year recalc / double-weight / development OFF / TCR 10. With this - as with all my sims - I have set hitter "make bad" to 120/100 and pitcher "make bad" to 25/18. Here's how the leaderboards came out: Nothing obviously untoward there. Although I'd love to know how, if the random LTMs that get applied to minor leagues are in effect here, how is the game expected to replicate NeL seasons with even the remotest level of realism and accuracy? No doubt you can just enter them in manually and do an autocalc if you feel so inclined, but still... For mine, among the many questions I have, that of how recalc is applied is right at the top of the list. So the only way I can think of to at least get some idea in the absence of any official word is to look at the ratings of a few players' scouting from this sim. So here is Charlie Smith under 5-yr / double-weight. Interesting the actual report says he is a pitcher. He shows no MLB level stats in his IRL tab. It looks like he got two recalcs in '25 and '26 then nothing. Here are his in-game IRLs. And here's Gentry Jessup (who also, it seems, gets imported automatically a second time as "Joseph" in addition to "Gentry". I have chosen these two because their short careers allow us to see the range in one screenshot. He also seems to get just two OSA updates in each of his first two seasons, then nothing. Here are his IRLs as per the game. What I'd like to know here is, if dev is OFF and no recalc is being done, how is he being scouted from that point on to generate his in-game ratings? Or am I reading this report incorrectly? __________________ HISTORICAL DO-OVERS A'S RED SOX DODGERS CUSTOM SAVES ECLIPSE LEAGUE MOON SHOT LEAGUE EVERYMAN LEAGUE GULF LEAGUE USBA Last edited by luckymann; 05-04-2022 at 12:02 AM.