Strategy Settings (Garlon & Others)

[Brad K]

In two tests the top 20 HR hitters are +198 and +210. Who lost?

I started with those who hit 200 in the game and worked up until I had 20 players.

Neutral 3 year

200 - 211 HRs in game

Chet Lemon -15
Bob Horner -18
Willie Davis +18
Benito Santiago -16
Oscar Gamble +1
Dave Henderson +5
Carl Everett +1
JT Snow +15
Kevin McReynolds -7
Richie Hebner +1
Gus Zernial -21
Mike Stanley +19
Joe Pepitone -13
Don Mincher +7
Carl Furillo +15
Toby Harrah +13
Deron Johnson +36
Gary Matthews -24
Rico Carty +7
Gus Bell +5

-43


Neutral 3 year current weighted

200 - 213 HRs in game

Barry Larkin +2
Gus Bell -6
Roberto Alomar -10
Chet Lemon -14
Travis Fryman -22
Bryan Jordan +18
Dave Henderson +7
Andy Pafko -6
Amos Otis +14
Ron Kittle +31
Benito Santiago -8
Pdero Guerrero -6
Carl Everett +7
Al Oliver -9
Gary Matthews -24
Bill Skowron 0
Wally Post +1
Todd Hundley +10
Dusty Baker -30
Dante Bichette -61

-106

[Scipio Africanus]

Originally Posted by Brad K

I don't consider that with a month of avoiding hef's questions there's much intention to help.

Ditto. I was checking in here occasionally to see if his questions were answered. My takeaway from reading this thread is that the people in it actually don't know the answers and don't actually know how all this historical recalc, etc. works.

I've concluded that no one knows how it actually works, hence the constant arguments about it.

[jcard]

Originally Posted by Brad K

I don't consider that with a month of avoiding hef's questions there's much intention to help.

But no worries, right, because as you yourself said recently in a thread about having a monthly developer Q and A thread:

Originally Posted by Brad K

They show up when there's a question users can't answer. It works and is more efficient than scheduled appearances.

[Brad K]

I've had new experiences since then!!! (And I've learned from them.)

[thehef]

Originally Posted by Garlon

We do have L/R data since 1901 but I need to figure out how to get them into the format we need to use them in the game. The current file in the game maybe starts around 1920 or so, I do not remember, but I know we looked into this and realized we had the opportunity to improve this. This is on my list of things for the next game.

Am I interpreting this correctly in that you are saying that real L/R splits are included for ratings beginning in 1920 (or thereabouts)? So if you start a game before that and choose real splits, you'll get randomly generated ones until you reach 1920 (or thereabouts)?

Originally Posted by Garlon

As for Mantle, I posted that he hit 58 HR per 550 AB and Maris hit 57 HR per 550 AB in 1961. Mantle gets a favorable adjustment for batting switch but it is not to the resolution that I am going in to look at exactly how many times he batted R at home vs on the road. I said he receives the overall combined Yankee Stadium HR adjustment for 1961.

There is also the fact that OOTP then creates its own estimate for the editor based on a standard number of AB vs RHP and LHP for batters. There are no issues there.

So the part about "OOTP then creates its own estimate for the editor based on a standard number of AB vs RHP and LHP for batters," do you mean that if Player A had 400 AB's against righties and 200 AB's against lefties, he would be treated the same - from a "stats for the editor" perspective as a guy with 500 AB's vs righties and 15 AB's vs lefties? Or?

Originally Posted by Garlon

The issue earlier was that Maris does not approach 61 HR when using Real stats and I said that Markus has two adjustments built into the game when using real stats so that HR totals do not get too extreme. One adjustment is HR power in general and another for his own adjustment for the home ballpark. I do not know what process is used there but I brought it up in testing that I felt HR were a bit too reduced for the best HR hitters when using real stats. For the neutralized stats the game does not make those adjustments though. I tried a 1961 with neutralized stats and Maris and Mantle both had over 60 HR when using the default import settings, settings which I think is too high and weaken too many players in the league which causes even more HR to be distributed to the top players. The adjustments made to HR power when using Real stats counteract the default import settings.

My takeway from this (the entire paragraph, not just the bold part) is that with 1-year recalc/replay settings:
- Real Stats overly reduce the likelihood that Maris would hit very close to 61 HR's in order to make it very unlikely that he'd hit way too many (like 70 or more)
- Neutralized Stats would give Maris a higher likelihood of ending up in the neighborhood of 61 HR's at the expense, however, of HR's for other players on the lower end of the spectrum.

Correct?

Re the part in bold above, I'm thinking that your conclusion really depends upon - and it may very well - how many more HR's over 60 they hit, and - especially in Mantle's case - if he had more AB's in your sim that he did in real life. Because we already know that Maris & Mantle homered at about the same rate in '61.

[Garlon]

I did address the questions and asked for more clarification on them. I do not know what is meant when you say the effect will be because it is not specific. The league totals modifiers keep the league playing to historical rates regardless of your configuration. I made some suggestions as well. You can play the game as you like because it is a sandbox.

Regarding those HR totals you reported what is your conclusion? It looks to me that the average per player is like +/-10 HR for their entire career which is only a couple per season.

Regarding the L/R before 1920 I suspect the game the game will probably provide its own values before that.

When you have the L/R totals in historical stats the game should be creating those L/R rating from those rates. It then combines them into an average amount of AB vs RHP and AB vs RHP to give you the total in the editor, but in the engine when playing games it will be using those vs RHB or vs LHB ratings for results.

My personal opinion is that the 300/50 import setting for batters is too much for single season. There are probably about 7-8 players per team that reach 300 in a season, so that means your entire bench and reserve roster is adjusted. Same for pitchers because 40/10 means 40*4 = 160 IP for pitchers and there will be fewer than 5 pitchers per team that reach 160 IP in a season, so many of the starts when using real lineups and transactions will be made by adjusted pitchers. Regardless, you may still like the results.

As for not knowing anything about recalc, I designed the league totals modifiers system with Markus around OOTP 7 and have continued to expand it since then when possible. I also built the neutralized stats, the quadratic park factors, the neutralized park factors, and all of the new fielding ratings for historical players. I updated a ton of missing statistical data in the real stats file too.

[Garlon]

Regarding these HR results you need to be careful on how you are interpreting them. Mantle and Maris combined for 115 HR in 1104 AB in 1961.

If they had the exact same AB we should expect them to combine for exactly 115 HR in 3.9% of the trials. That is the binomial probability of that occurring. There would then be about a 48% chance that they would end up with less than 115 and a 48% chance they would end up with more than 115 for the season. That is how probability works and the game engine uses probability.

[Garlon]

When setting up a historical game there are many options.

Real or Neutralized stats
Fielding Imports: 3 options
Pitcher Stamina: 3 options
Use L/R: Yes or No , 2 options
Adjust Pitchers and Hitters many options
Coaching System: 2 options
Storylines: 2 options
Personality Ratings: 2 options
Player Fatigue: 5 options
Disable Player Development 2 options
Relievers: 6 options
Stamina: 5 options
PH for Pitchers:5 options
PH for Position Players: 5 options
Defensive Substitutions: 5 options
Stealing: 5 options
Hit and Run: 5 options
Bunting: 5 options
Infield Shifts: 7 options
Recalc: 6 options
Closers: 5 options

So with these options there are 283,500,000,000 possible configurations.

[Brad K]

Originally Posted by Brad K

Since you're still lacking an answer despite multiple requests, let me attempt an answer based on my understanding of what has been posted, observations of ratings, and testing.

Assuming one real recalc and no TCR of development, a players rating is adjusted by the information on park factors in the game's era_ballparks file. However this adjustment is largely irrelevant when players are playing in their historical ballparks.

For example (and ignoring the possible effects of the "in a modern neutral environment" statement in the editor concerning raw ratings), if a player hit 40 HRs but his home ball park was hitter friendly to the point he was helped by 4 HRs then his rating is 36. However since in the game his home park is his historic one, he performance will be boosted so his effective rating is 40 HRs.

However if his 40 HR year was an outlier for him - say he never hit more than 22 in another year - his rating is adjusted further to make it extremely likely he won't exceed 40 and extremely likely he won't get close to 40. Another player who hit 35 and has multiple mid 30ish HR years in his career is likely to hit more HRs than the 40 HR player since the 35 HR player will not have his rating reduced as much or probably at all.

Middle range performers even perhaps up to the level of the consistent 30ish home run player get additional performance as there is a transfer of output to them from low AB players. High performance low AB players have their performance reduced as the low sample size makes their raw rating unreliable. Although it could be considered low performance low AB players is also unreliable due to low sample size and thus the output should be transferred to them, it is not. Instead is it sent to players who are not low AB players even though their sample size is large enough that their performance is reliable.

If three or five year recalc is used the 40 HR player's rating for the year will be even lower than the downward adjusted single year rating while the consistent mid 30ish HR player will have ratings similar to single year ratings.

With players playing in their historical parks and with historical lineups, the main effect of the adjustment of the ratings with the park factors file is to change the distribution of player's performance among the various parks. Overall performance is not changed by the park factors. Any change in the per park performance is so low as to not be discernable amid the normal randomness.

In summary, peak HR performance is reduced in players best years. Performance of low AB players is reduced. In both cases the reductions are transferred to players with AB totals that are greater than the low AB threshold. As a group, players above the low AB threshold will perform better (except those having an extraordinarily good HR year compared with their other years),

Except for HR performances like the peak years of Roger Maris and Davey Johnson, casual examination is unlikely to reveal the deviations from historical performance. And people who play "historical what it", perhaps the most popular form of historical play, will find the performance differences lost in the noise caused by players playing for different teams, missed time due to non historic injuries or time not missed due to historic injuries, development settings, TCR, and the other things used to make the game more interesting to a human GM. But it's still there.

OK, I know there are four questions, but as can be seen from the amount of material needed to explain just this one, the others are going to have to wait until my typing fingers get a rest!

In the meantime, amplifications and corrections to the above are welcome.

I wrote the previous based on my understanding of how the game worked based on comments from Garlon and the developers. Here is the testing to back it up.

real stats optimized for single year replay but run 1951 - 2007


Top 20 real one year

Career HRs in game

Barry Bonds -42
Hank Aaron -71
Ken Griffey Jr active player
Willie Mays -37
Reggie Jackson +22
Frank Robinson -2
Eddie Murray +61
Mickey Mantle +6
Sammy Sosa active player
Rafael Palmeiro -39
Harmon Killebrew -55
Stan Musial +39
Mark McGwire -72
Ted Williams -11
Ernie Banks -11
Eddie Mathews -11
Dave Winfield +34
Jeff Bagwell +45
Cal Ripkin Jr +59
Gary Sheffield active player
Jim Thome active player
Carl Yastrzemski +32
Willie McCovey -40
Mike Schmidt +28

Overall -65

Top ten -118

Next ten +53


Career in game HRs 202 - 214

Dave Henderson +3
Don Mincher +3
Sid Gordon +1
Brady Anderson +7
Kevin McReynolds +6
Todd Hundley -3
Jay Bell +10
Benito Santiago +11
JT Snow +18
Wally Joyner +3
Bill Freehan +7
Joe Pepitone -11
Wally Post 0
Cesar Cedeno +12
Gus Zernial -25
Toby Harrah +17
Ron Fairly -3
Bob Watson +29
Dick Stuart -15
Bill White +12

+82


HRs in game 1002 - 1030 career ABs excluding pitchers

First is number hit in game, second is difference

Floyd Rayford 34 -4
Allie Clark 30 -2
Junior Kennedy 5 -1
Armando Rios 32 -4
Ken Wood 27 -7
Bryan Little 4 +1
Syd O'Brien 23 -1
Jeff Stone 7 -4
Tom Egan 23 +1
Dave Marshall 13 -3
Wendell Magee 12 -12
Tommy Glaviano 24 0
Scott Thompson 8 +3
Erv Disal 21 -3
Pete Daley 20 +2
Cal Neeman 34 +4
Kevin Orie 15 -7
Wayne Krenchicki 14 -1
Pancho Herrera 35 +4
Onix Concepcion 6 +3

-31

418 real life HRs

387 game HRs

7.4% decrease

[thehef]

Originally Posted by Garlon

When setting up a historical game there are many options.

Real or Neutralized stats
Fielding Imports: 3 options
Pitcher Stamina: 3 options
Use L/R: Yes or No , 2 options
Adjust Pitchers and Hitters many options
Coaching System: 2 options
Storylines: 2 options
Personality Ratings: 2 options
Player Fatigue: 5 options
Disable Player Development 2 options
Relievers: 6 options
Stamina: 5 options
PH for Pitchers:5 options
PH for Position Players: 5 options
Defensive Substitutions: 5 options
Stealing: 5 options
Hit and Run: 5 options
Bunting: 5 options
Infield Shifts: 7 options
Recalc: 6 options
Closers: 5 options

So with these options there are 283,500,000,000 possible configurations.

Not only that, but for some of these options - like Stealing, Hit & Run, Shifts... - you might changed these at certain points during your sim. As if that 283.5b number wasn't already big enough!

So, with so many options, it's understandable that we'd want to get pretty clear on the ones that aren't as obvious as to what exactly they mean

[thehef]

Originally Posted by Garlon

When you have the L/R totals in historical stats the game should be creating those L/R rating from those rates. It then combines them into an average amount of AB vs RHP and AB vs RHP to give you the total in the editor, but in the engine when playing games it will be using those vs RHB or vs LHB ratings for results.

I sort of understand... So how what does the import & editor do in these examples?

a) 5-year recalc, player has plenty of AB vs RHP (say 400+ per year) but very few each season (say < 10 per year)? Do the adjust/weaken settings come into play only for AB vs LHP?
b) 1-year recalc, player has 400 AB and just 5 AB vs LHP. Same question.


Also, I'm still wondering about both of these questions:

Originally Posted by thehef

On the player editor panel, right side, underneath offensive ratings, we see Resulting Stats (in a neutral, modern Major League environment). In looking at the stats for some 1973 Dodger players, this is clearly as advertised (batting avg's & HR's are higher than in real life, presumably reflecting the fact that Dodger Stadium was quite the pitchers' park back then).

However, what I don't get:

1) I'm playing 1-year recalc, real stats. Why then am I not looking at real stats (converted to 550 AB's)?

2) If I want to do a what-if scenario of, say, 1-year recalc/replay settings, but the "what if" = averaging Bill Buckner's, Bill Russell's, and Tommy John's 1974 & 1976 stats and using those for 1975 to see how LA would do with those guys being healthy in '75, how would I do that? I can easily come up with the avg real stats from '74 & '76, but how am I to know how to convert those to "Resulting Stats, in a neutral, modern Major League environment?" (I suppose I could create 1974 and 1976 sim games for the sole purpose of getting the "resulting stats" for those two years, and then average those, but that's a fair amount of work...)

Am I missing something here?

Why, for a sim based on real life (as opposed to newts), would I not instead see actual real-life stats (with the only difference being that they would be reflected as per 550 AB's)?

If the answer is that these are real-life stats that adjusted for park factors, then the next question would be: How different would these stats be if they were also neutralized instead of "sorta" real-stats?

And as for the second question... I had re-stated it here:

Originally Posted by thehef

So, getting back to my example of playing with 1-year recalc but wanting to do a "what if Buckner, Russell, and John were healthy on the 1975 Dodgers" and my plan was to take an average of their real life 1974 and 1976 stats, near as I can tell there are only two ways to do this:

1) Create test games of the 1974 and 1976 seasons and see what OOTP's Resulting Stats... are for the 3 players in question and then average those, and input them into my 1975 game, or...
2) Average the real 1974 & 1976 stats for those 3 players and then find some way of converting them to match whatever OOTP is doing to come up with "Resulting Stats..."

Correct? Or is there some other less-arduous way?

How is a gamer to use these EDITABLE stats fields, given that a) we likely only have REAL stats, and b) don't know the formula to convert them to park-adjusted (or whatever they are)?


And...

Originally Posted by Garlon

I strongly suggest using neutralized statistics for career leagues and suggest using 5yr recalc with it. For replays if you like your results with real stats then use that for your games.

I understand the first recommendation, and the second sentence pretty much goes without saying. However, what is the recommendation if one wants the most accurate replay possible, using 1-year recalc, no development, in the following examples:

- Real transactions, but no real injuries or real lineups: Use real stats or neutralized? What should Adjust/weaken settings be?
- No real transactions, players importing to actual teams (thus, you will have many players playing in their real-life parks, but plenty who won't). Use real stats or neutralized? What should Adjust/weaken settings be?
- No real transactions, and no imports to actual teams. Use real stats or neutralized? What should Adjust/weaken settings be?

[pstrickert]

Correct me if I'm wrong, but I thought Markus toned down HRs a bit for the Steroid Era. If so, Bonds, Mc Gwire, and Griffey Jr (unfortunately) would be impacted.

[Brad K]

Originally Posted by Garlon

I did address the questions and asked for more clarification on them. I do not know what is meant when you say the effect will be because it is not specific.

It was an open ended question where you could talk about any effect you wanted to.

[thehef]

Hey Garlon, I'm hoping you can reply to the questions, etc., in post #111 above (https://forums.ootpdevelopments.com/...&postcount=111)

Thanks in advance!

[Brad K]

Originally Posted by Garlon

Regarding these HR results you need to be careful on how you are interpreting them. Mantle and Maris combined for 115 HR in 1104 AB in 1961.

If they had the exact same AB we should expect them to combine for exactly 115 HR in 3.9% of the trials. That is the binomial probability of that occurring. There would then be about a 48% chance that they would end up with less than 115 and a 48% chance they would end up with more than 115 for the season. That is how probability works and the game engine uses probability.

I don't know of anyone who uses the Maris/Mantle combined total as a measure of how accurate the game is. Throw enough players into the mix and the game will produce the right combined number.

[Garlon]

thehef,

Regarding the questions about the AB vs RHP and AB vs LHP I do not know how Markus handles those in terms of sampling.

As for why you will not see real stats converted to 550 AB in the editor, Markus makes some adjustments to HR for some players when using real stats and there may be other slight adjustments or rounding in there too. The game also assumes some balance of AB vs RHB and vs LHB and sums those to generate the resulting stats that you get in the editor. There are no actual issues with the game converting real or neutralized stats to the ratings. The difference is mainly the additional HR adjustment that you get with the real stats that you do not get when using neutralized statistics.

For your 1975 Dodgers I suggest simply typing in the sum of their stats from 1974-1976 and OOTP will convert it to 550 AB. Do not worry about doing the park adjustments since the adjustment are not that significant.

As for the 1yr recalc questions, I like using neutralize stats. As for the import settings for 1yr I think you can go down to about 150 /50 for batters if you are using real lineups and injuries but you ay want to go with a larger if those are not enabled. I suggest looking at baseball reference and sort all of the players in a season by AB and see if there are players between 150-299 AB who you think will be overpowered. If you are managing every game for a specific team you may want to set the imports to give your team the advantage. For example , if I am using the 2016 Cubs then AI will use 150 AB or lower because they had many really good contributors on that team who had over 100 AB and if I use the default 300 AB setting those players will not be as good. The same can be said of selecting the pitching imports. The 2016 Cubs had 5 SP who had more than 160 IP, so the default of 40/10 is fine, but if you are managing a team who had a pretty good 5th starter with only 120 IP, then you may want to go with 30/10 so that pitcher is not adjusted.

[Brad K]

thehef

If you want real splits you can enter the vs L and vs R stats separately and select calculate ratings. If you enter vs L then the overall ratings are the ratings for vs L, etc. You can then take those separate L & R ratings and enter them as player ratings.

The resultant stats will likely not match the players actual overall stats. OOTP calculates overall based on 75% vs R and 25% vs L and players seldom have those percentages. But it doesn't matter because the game runs on the separate L and R ratings..

[Brad K]

Garlon

I didn't realize the editor would produce correct ratings without entering stats per 550. I've always converted to per 550 before making entries. Thanks for the info. My spreadsheet thanks you also!

[thehef]

Thanks, Garlon, for the detailed reply

Originally Posted by Garlon

The game also assumes some balance of AB vs RHB and vs LHB and sums those to generate the resulting stats that you get in the editor.

Not sure what that means. Could you further explain?

Originally Posted by Garlon

There are no actual issues with the game converting real or neutralized stats to the ratings. The difference is mainly the additional HR adjustment that you get with the real stats that you do not get when using neutralized statistics.

I'm not seeing this. For examples, I'm seeing that 1973 Ron Cey has a .264 average in the editor, but IRL he hit .245. For Steve Garvey, .327 vs .304. Davey Lopes, .296 vs .275. (And yep, this is a game where I chose real stats, 1-yr recalc.)

Originally Posted by Garlon

For your 1975 Dodgers I suggest simply typing in the sum of their stats from 1974-1976 and OOTP will convert it to 550 AB. Do not worry about doing the park adjustments since the adjustment are not that significant.

If I sum up their real stats, then whatever conversion is boosting/adjusting the numbers (in the examples of Cey, Garvey & Lopes above) won't be factored in, right?

Originally Posted by Garlon

As for the 1yr recalc questions, I like using neutralize stats.

Why? What makes them better for you than real stats?

Also, since we've pretty much determined that the "real" stats that are shown in the editor are not actually real stats, I'm still wondering then what the difference is between "real" stats and the newts. Can you please explain?

[thehef]

Originally Posted by Brad K

thehef

If you want real splits you can enter the vs L and vs R stats separately and select calculate ratings. If you enter vs L then the overall ratings are the ratings for vs L, etc. You can then take those separate L & R ratings and enter them as player ratings.

The resultant stats will likely not match the players actual overall stats. OOTP calculates overall based on 75% vs R and 25% vs L and players seldom have those percentages. But it doesn't matter because the game runs on the separate L and R ratings..

Well, this part is still a mystery to me: Not only the questions about real stats that aren't real in the editor, but also about how L/R splits factor in. You mentioned that "you can enter the vs L and vs R stats separately and select calculate ratings," but I'm not seeing that it can be done. Rather, you only could enter totals (not splits). IOW, I don't see where you could enter them separately. And even if it could be done, we wouldn't know how to adjust them so they'd become the "real" stats that aren't real (see my Cey, Garvey, Lopes examples in the post above). Hopefully that makes sense...