Creating Historical Managers: How to Approach (Mod WIP)

[G-Force]

So as I get myself back into OOTP and prepare to kick off my first new dynasty/league in several years, I've decided to at least look into creating managers for the game. I've created a ton of rosters, between High Heat, OOTP, even MVP (I still have the scars from that one), but I felt very comfortable in how I went about those. Stats are right there for the taking, you just have to decide how to utilize them in your roster system.

This will be different. I go into it knowing there's probably only so close to accurate I can get these. But we're going to give it a shot and see how it goes.

I'm putting this in "general" just because I think more people will see it and your feedback could be very helpful here. But if mods feel it should be moved into the mod forum, by all means.

I'm writing this from my office at work, so I don't have the game in front of me, but I pulled up a screengrab of a manager profile. So let's take a look at the strategy categories:

Stealing
Aggressive Baserunning
Hit & Run
Sac Bunts
Starting Pitcher hook
Relief Pitcher hook
Favor L/R matchups
Pinch hitting
Use infield shifts

Some of these should be pretty easy to work out. Others FAR more difficult.

One thing I think is that most managers have a minimal impact and honestly go pretty much by a book of sorts. What that means to me is that the overwhelming majority of settings should be in the below average-average-above average points (between one click left of center and one click right of center).

It could be argued that old time managers in particular should have two settings/profiles: one that's era appropriate, and one that evaluates them relative to their era so they can be utilized cross era if desired. We'll cross that bridge eventually (hopefully, unless this completely flames out!)

My impetus for this is a new dynasty I'm running called the Alltime Baseball League, which involves the best players from throughout history. So I wanted to incorporate real managers to go along with it.

It's a 16-team league. I'm going to start off with a list of 32 managers from history and if that goes well, then I'll expand this project to include more.

We'll see how it goes!

GH

[JaBurns]

Good luck I will be interested in what you come up with and add my if when it is appropriate or else wise if you don’t mind; take my comments with a grain of salt.

[pstrickert]

MLHFFYP (user name) is the person who created the strategy profiles. He’s more familiar than anyone with the settings. But he’s been absent from the forums since 2015, according to his user profile. I remember he said that some settings are much more sensitive than others. So, in order to see any difference at all (in some areas), he had to go to extremes. At the same time, other settings needed just a small nudge to make a noticeable difference.

Markus recommended using these profiles with caution. Playing out games manually is fine, I guess. But simming will often produce less than ideal results.

[G-Force]

Originally Posted by pstrickert

MLHFFYP (user name) is the person who created the strategy profiles. He’s more familiar than anyone with the settings. But he’s been absent from the forums since 2015, according to his user profile. I remember he said that some settings are much more sensitive than others. So, in order to see any difference at all (in some areas), he had to go to extremes. At the same time, other settings needed just a small nudge to make a noticeable difference.

Markus recommended using these profiles with caution. Playing out games manually is fine, I guess. But simming will often produce less than ideal results.

Interesting and noted. Will definitely test before implementing anything.

GH

[The Game]

are you creating fictional or real MGR profiles? If real i would like to see Joe Maddon.

[G-Force]

So the first manager I decided to look at was Bobby Valentine, since I know him well having watched him manager my Mets.

One thing that came clear quickly is that so much of how a manager operates depends on personnel, and the manager and GM aren't always aligned there. So depending on the team, a manager may need to operate differently than his ideal. Which makes really setting a "tendency" difficult.

Let's look at Valentine's desire to steal bases.

When Valentine managed the Texas Rangers from 1986-1991 (I didn't use partial season data because it was skewed), he utilized the stolen base often, particularly in his first three seasons.

TEX SB Att/League Average
1986 1.2
1987 1.07
1988 1.19
1989 .91
1990 1
1991 .96

Now, is this shift a product of a change in attitude or personnel?

He then managed the Mets, where we'll evaluate him from his full seasons of 1997 to 2002.

Mets SB Att vs NL Avg
1997 .9
1998 .73
1999 1.21
2000 .76
2001 .83
2002 .93

So now you've got a clear downward trend, largely due to the fact that those Mets teams were ploddingly slow with guys like Piazza, Ventura, Zeile and Alfonzo. The 1999 spike? Rickey Henderson and Roger Cedeno in the lineup (Cedeno stole 66 bases).

His one year with the Red Sox, they were a .9.

I won't know until I do a bunch of managers and see how this method plays out. But with the average of .97 over his managerial career, a figure that holds even if I pull out the highest and lowest as outliers, I would be inclined to classify that as average. Particularly given pstrickert's post above, we'll want to stay in the middle where there's any doubt I think. I venture that when the setting is "average" the personnel really dictates the strategy, which is at it should be.

So...are all the figures going to end up like that? For Valentine, his teams' attempted sacrifices over the course of his career worked out to a .99 (definitely average) as did his teams' age relative to league. Taking the extra base/baserunning aggressiveness? A 1.02.

But...use of pinch hitters...1.25!

I was excited to see this one because I was sure of it having watched Bobby V, who loved his platoons and R/L matchups. So seeing the data demonstrate that pleases me. His use of pinch hitters and R/L matchups would be something above average, though how much so depends on the game sensitivity mentioned in the prior post.

Sorry this is long, just figured I'd walk through what I was looking at and go from there.

Up next, bullpen management and I'll look at Dick Howser.

GH

[G-Force]

Originally Posted by The Game

are you creating fictional or real MGR profiles? If real i would like to see Joe Maddon.

I'll look at Maddon tonight or tomorrow.

GH

[pstrickert]

Bob McGraw, the MLHFFYP referenced above, used the book Evaluating Baseball’s Managers (by Chris Jaffe) as a source. But Bob also did a ton of testing.

[G-Force]

So I ran Dick Howser. Quick rundown after the previous longwinded post

Note, 1=league average. Anything in that ballpark should, for the moment, be considered average (i.e. from .95-1.05). Also, note NONE of this has been tested yet. I'm just discussing the framework of coming up with ratings here. So I wouldn't go actually using these yet.

Stealing Attempts: 1.12 (Above average)
Sacrifices: .82 (Below average - perhaps 2 clicks pending testing)
Pinch hitting: .81 (Below average - perhaps 2 clicks pending testing)
Aggressive baserunning: 1.03 (Average)
Young vs Veteran: 1.05 (Average)
L/R Matchups: .99 (Average)

I'll do Joe Maddon next. And yes, I also know some ratings aren't discussed yet, as I haven't gotten into methodology for evaluating them yet.

GH

[G-Force]

Maddon's a great guy to look at for this because the perception is that he's very hands on. As such, he should probably yield more "non-average" ratings than many others. The numbers bear this out.

Yet it also demonstrates a caution of employing other factors into the mix.

The numbers themselves (again, for our initial purposes we'll consider anything between .95 and 1.05 "average").

Stealing: 1.2
Sac bunts: 1.03
Pinch hitting: 1.35
Aggressive baserunning: 1.04
L/R matchups: 1.06

So the numbers certainly support the perception of Maddon as an involved, hands-on manager. That said, we need to be a bit careful.

Look at stealing...From 2006-2012 Maddon employed the stolen base a TON. His SBA were from 36% to 71% above the AL average. Having Carl Crawford stealing 50 bags a season certainly influenced that.

But...from 2013-2017, Maddon's teams - his last couple years with the Rays and his 3 with the Cubs - have attempted to steal BELOW the league average 4 of 5 years (2015, his first year with the Cubs was the exception).

So did Maddon's philosophy change or was it just a matter of personnel, not having a threat like Crawford to deploy.

I'm inclined to think it's a personnel issue and that Maddon, if he had his way, would steal plenty. In 2011, the year after Crawford left the Rays, the Rays still attempted 37% more steals than the league average and ended up with 7 players in double digit steals. This is further supported by the fact that Maddon's teams in his 12 seasons managing have taken the extra base at or above the league average 10 out of 12 seasons, which supports the notion of an aggressive approach on the basepaths.

I won't know how many clicks above to use until I get into testing things a bit more, but it seems safe to say Maddon will be at or above average pretty across the board on the offensive side of things.

GH

[G-Force]

A quick note...there were 26 fictional managers created for my alltime league, so I was curious as to the ranges of their strategy preferences.

Pstrickert mentioned earlier in the thread that Markus had said the settings could have great impact if overdone. That of course was said a couple of years ago. While it still may hold true, it seems clear that the game engine wants to keep things from being too extreme in terms of strategy.

In analyzing the fictional manager settings:

CATEGORY	Avg	Median	Mode
SB	     0.58	0.5	-1
BR	     0.58	1	1
H&R	     0.23	0	-1
R&H	     0.15	0	-1
SAC	     0.19	0	-1
BH	     0.15	0	-1
SQUEEZE	0.15	      0	-1
PA	                0.35	      0.5	1
IW	                0.27	      0	-1
HR	                0.15	      0	0
GL	                0.58	      1	1
INF IN	         0.54	0	0
COR IN	         0.81	1	1
SHIFT	         0.92	1	3
PR	     1.12	1	3
PH POS     	1.04	       1	1
PH PIT	         0.88	1	-1
HOOK SP	-0.54	-1	-1
HOOK RP	-0.5	     -0.5	-1
CLOSER	              1	2	2
L/R	                 1.08	1	1

So what's clear here is that the settings should most frequently be middle or one click either way, based on the game engine. As such, when I plot this out I'm going to limit the extremes significantly.

GH

[G-Force]

I grabbed Jaffe's book, Evaluating Baseball's Managers. Great work but REALLY disappointed that he only gives the data for a few managers on each stat. For example, while he notes in the back of the book he collected the data on 54 managers for hit & run data, he only gives the figures for 10. I don't get why you would do that...

Oh well. Will use what I can.

GH

[pstrickert]

Jaffe's brother posts on the OOTP forums sometimes.

[G-Force]

I've now done 11 managers through my first sheet, and it's reinforcing my belief that so many of what we call manager tendencies are, at least frequently, just adapting to personnel, changes in the game itself or simple development of a manager over time.

Looking at Earl Weaver, he changed dramatically, which makes evaluating him on the whole somewhat difficult.

Let's look at bunting. Again, all these figures are rates relative to the league, 1 being average.

Weaver's figures from 1969-1975 are 1.38, 1.47, 1.3, 1.14, 1.5, 1.37 and 1.18. Clearly, Weaver was a strong proponent of the sacrifice.

Now let's look at him from 1976-1981: .84, .85, .66, .66, .74, 92

Weaver suddenly became less enamored with the sacrifice bunt. The Earl Weaver we so often hear about is focused on his love affair with pitching and the 3-run homer. But this attitude developed with time.

A quote from Weaver on this: "Weaver preached his prescription for winning: “pitching and three-run homers.” He explained, “Pitching keeps you in the games. Home runs win the game.” That formula developed with experience; in his first seasons with the Orioles he called for more sacrifice bunts than the average AL manager, but he came to believe that “your most precious possessions on offense are your 27 outs,” and “if you play for one run, that’s all you’ll get.”

Oddly enough, that quote appears in a book about the 1970 Orioles - a year in which Weaver sacrificed 47% more than the average AL team.

But that 1975 line for Weaver is around the time for a few other changes to his tendencies.

Stolen Bases:
1969-1974: .95, 1.03, .88, 1.03, 1.47, 1.22
1975-1982: .88, .99, .85, .8, .8, .94, .71, .55

Pinch hitting:
1969-1974: .72, .76, .64, .94, .97, .77
1975-1982: 1.07, .99, 1.09, 1.25, 1.15, 1.38, 1.38, 1.5

Platoon rate:
1969-1975: .93, .91, .93, .98, .95, .95, .8
1976-1982: .98, 1.09, 1.12, 1.09, 1.26, 1.12, 1.12

So the appearance seems to be that Weaver came to value those 27 outs far more over time. Less risks on the bases, more use of pinch hitters and platoons seeking the more favorable batter/pitcher matchup. Which is logical, a simple matter of evolution of a philosophy as he gained more experience.

Of the 11 managers I've done, nobody has a change in philosophy this clear across multiple categories.

Forgive the rambling, but I'm enjoying researching this project and like talking baseball

GH

[pstrickert]

Good work, GF. Keep it coming. I remember your rosters from long ago. You’re part of OOTP legend and lore.

[G-Force]

Originally Posted by pstrickert

Good work, GF. Keep it coming. I remember your rosters from long ago. You’re part of OOTP legend and lore.

Well thanks, I appreciate that! Glad to be back at it!

GH

[G-Force]

So with 11 managers run through, I listed all of their individual seasons in an effort to start a framework to convert their decision ratios, for lack of a better term, into OOTP ratings. Attached is what I have for the 11 managers right now, who range from the well known (Joe Maddon, Earl Weaver, Bobby Valentine) to a few lesser known skippers (Joe Altobelli, Lum Harris).

The VS Book column is just the addition of pluses and minuses to see who deviates most from the "average" rating.

More guys to run through, and then I'll try to start working on bullpen management ratings, which are really, especially to the modern baseball fan, probably the area most debated in how a manager handles his team.

GH

[JaBurns]

Ralph Houk looks like he did nothing according to the book. Actually, if I am interpreting the numbers correctly, he looks the leading contrarian of all time.

[BaseballMan]

Will you have to create the managers with every new league or can they be imported or copied?
Rating managers might work for what if historical leagues but i don't think it would be worth the time for leagues using historical transactions and real rosters as those teams should reflect the success of the real teams which should reflect the success of the historical manager.

I use real rosters and historical managers in name only.
Harry Wright is usually at the top in wins for the 19th century. Then Connie Mack and John McGraw are at the top with Joe McCarthy winning a lot of championships.

[G-Force]

Originally Posted by JaBurns

Ralph Houk looks like he did nothing according to the book. Actually, if I am interpreting the numbers correctly, he looks the leading contrarian of all time.

Sure seems that way don't it? I'll be curious to read a bit more about Houk in the books and see some more detailed analysis of him.

GH