Historical replay accuracy

[BigRed75]

Originally Posted by knockahoma

The name "Lahman" has been banned from this universe, never to be spoken aloud, along with the other forbidden word: DH.

What if you get a fictional player in the draft pool whose last name is "Lahman"?

[knockahoma]

Often a victim to stream of consciousness, I got to musing about the lineup for the 73,74,75 Braves. Did it have anything to say about that old chicken or the egg, which comes first on balls in play, pitching or defense? I believe Defense is a huge factor. You could take an average pitcher and put an elite defensive team behind him and it could make a big difference, IMO. But, sometimes I feel it's the other way around. Great pitching makes defensive numbers look good.

In 73 and in 75, the Braves finished near the bottom of the league in defensive efficiency and fielding percentage. But sandwiched between those years, in the 74 season they finished at the top, or near the top in those two categories. And I wondered, who was playing defense for the Braves? Huge changes in defensive talent? I didn't remember it that way. So, I perused the lineups, according to baseball-reference.com. '73, the bad defensive year featured a group of generally mediocre defenders. 3B Darrell Evans was the only guy at this time considered a better than average defender. In fact, several were less than average in ability by rep and numbers.

73

C Johnny oates
Paul casanova

1b Mike Lum

2b Davey Johnson

SS Marty Perez

3B Darrell Evans

LF Hank Aaron

CF Dusty Baker

RF Ralph Garr

Big changes in defensive ability in 74 when the Braves topped the league in defensive efficiency, and finished 3rd in fielding average? Not really.

74

C Johnny Oates
Vic Correll

1b Davey Johnson

2b Marty Perez

SS Craig Robinson

3B Darrell Evans

LF Hank Aaron

CF Dusty Baker

RF Ralph Garr

Adding Correll in place of Casanova at catcher seems pretty much a wash. If anything, Correll had a mediocre rep for defense. Davey Johnson had lost a lot of range at 2B, so he actually split time between 2b and 1b. Lum covered 1B when Johnson was the 2nd sacker. Was he a much better 1B than Lum the year before? No. Marty Perez, at this time in his career, was better off at 2B than as a SS. He probably was a better ranging 2B, but still nothing amazing. Craig Robinson, Perez' replacement, was simply a bad shortstop his entire career. He was gone by 75. 3B and the OF remained constant, the same defensive talent (or perhaps slightly less, if Aaron's aging legs are considered).

In 75, the Braves fall back to near last in the league in the afore-mentioned defensive categories. Here were the changes in lineup:

75


C Vic Correll
Biff Pocoroba

1b Earl Williams

2b Marty Perez

SS Larvell Blanks

3B Darrell Evans

LF Ralph Garr

CF Rowland Office

RF Dusty Baker

Poco was generally considered a pretty good defensive catcher, at least similar in ability to the vacating Oates. While Lum was still around to play 60 games at 1B, the Braves went with defensively stiff Earl Williams for more games at that position. Marty Perez remained at 2B. Larvell Blanks was not a good shortstop, but neither was the guy he replaced from the previous year, Robinson. Evans remained at 3B. Garr was shifted from RF to LF, but he had played numerous games at both positions in his earlier career. Rowland Office moved Dusty Baker to RF and became the Braves CF. Office was not considered an elite CF by any means.


Of course, there were bench players, etc., to consider. But, it is interesting how pretty much the same talent, or level of talent through a three year period can finish near last, near first, and near last. It does seem to elevate pitching into the scene.

[knockahoma]

The name "Lahman" has been banned from this universe, never to be spoken aloud, along with the other forbidden word: DH.


What if you get a fictional player in the draft pool whose last name is "Lahman"?

I think I came close, LOL. I think it was Lehman. I had him adopted and given the appropriate name change.

[thamolas]

Originally Posted by BruceM

Claiming you're the best and claiming you're accurate are two different things. To me, OOTP is the best historical, but that's because I'm not concerned with accuracy as in mimicing real baseball.

I don't play historical games just because I like to spend a lot of time tooling with archetypes, inventing my own fantasy almost-everything, and experimenting with crazy lineup/strategy schemes; but I can understand this line of thought. In a simulation, I imagine there to be a lot of "wiggle room" for a player to change in ways that he didn't in real life and for teams to make mistakes they did not in real life and/or avoid mistakes they did make. A series of weird disasters that occurred in real life may not ruin the team if they do not occur in the sim and a successful real-life franchise could likewise find itself disaster-prone in a sim.

To get back to "wiggle room", I'd like to share my own real life baseball story: As a teenager, I was a knuckleballer with such ridiculous movement on my pitches that they were practically unhittable (and I easily struck out guys who have gone on to successful MLB careers). Most of the innings I pitched yielded no hits. However, I was unable to get my control to a point where I wasn't hitting batters and throwing wild pitches in every game. I went to camps, worked on all of my mechanics with top-flight pitching coaches, etc. and nothing worked. After countless hours of practice and determination with no improvement, I gave up. However, if my pitching "career" were simulated many times, who is to say that something wouldn't "click" one (or even most) of those times, leading me to have a reasonable degree of control and allowing me to go on to a successful (simulated) MLB career?

What I'm saying is that, in pro sports, most of the benchwarmers are among the best players in the world. If one coach (or fellow player) reaches that bench player in a profound way, history can change. People like to think of their favorite players as having a static set of skills and measurable talent when, in real life, they are human beings who are constantly tinkering with their games and working hard to improve (and in some cases, to maintain the high level they have achieved and eventually failing that when Father Time inevitably takes over). Real people have niggling injuries, psychological blocks, turbulent personal lives, addictions, and occasional confidence problems (too much or too little). When you're simulating an athlete's career, it's nice to have such variation. Sometimes Joe Blow has the stunning revelation that he didn't have in real life, fixes the hitch in his swing and has a better career than Ted Williams (and fixing it for one season before reverting to his old form is certainly believable). Sometimes Pete Rose will develop an even nastier habit and his career will fade early. I would find this kind of variation believable and acceptable, though I believe that the AI managers benching great players for no discernible reason would annoy me, too.

And, yeah, if OOTP focused more on it's Fictional play, I'd be thrilled. Heh.

[actionjackson]

I think that by using Lahman, there's a bit of a statistical boomerang effect going on. I may not be able to verbalize it too well, but what the heck. If the game is basing its defensive ratings off of TC or RF, then infielders on teams with multiple strikeout pitchers or flyball pitchers are going to suffer some damage to their defensive ratings because they won't be getting as many chances as they might with the Roy Halladay (shameless plug) types who keep the ball in play and mostly on the ground. Thus, the real life (insert team name here) pitching staff's tendencies and makeup has effected the defensive ratings for the imported players on said team, which may or may not be importing that pitching staff optimally putting said team into a potential heap of trouble. Now a pitcher who puts the ball in play often, say...Carl Morton from your 1974 Braves goes out there to pitch with a defense behind him that is less effective than it was in real life, he's gonna be in trouble. I can't explain Phil Niekro, except to echo the sentiment above that maybe the game doesn't deal with knuckleballers too well.

As for Belanger, arrrgh! Why can't the AI recognize that his defense is what that pitching staff needs. Although some of those Oriole staffs were a bit fly ballish according to Baseball-Reference, which might dampen his defensive excellence depending on what year he is imported in. I don't know how difficult it is to code, but for the up the middle positions the AI should be taking a defense first position because that's where a lot of the action is. Frustrating eh?

My only suggestion would be to get a database that uses neutralized stats like the Spritze DB 2008, especially if it has the fielding adjustments for the problem I've been talking about (not sure, but I do know that the original Spritze/Garlon or Garlon/Spritze DB worked on this very problem and I did stay at a Holiday Inn Express last night ) as well as a whole bunch of other cool sounding features. I can't personally vouch for it as I have yet to use it but it sounds like it might help fill in some of the defensive gap that Lahman can't begin to deal with, as well as CS in the early days and a host of other issues that Lahman (despite its brilliance) has come up short on. Could in fact be a database issue as opposed to a game engine issue...whatever it is, hopefully the geniuses around here can figure it out so this game can hit the warp speed button.

EDIT: Just realized that Spritze DB 2008 is not neutralized, while Garlon/Spritze or Spritze/Garlon is. Sorry for the confusion and maybe Mr. Spritze or Mr. Garlon can edu-meh-cate me and perhaps others as to the differences between the two.

[RonCo]

I've often thought the the game should create its own historical database. The problem, of course, is that it would take a huge amount of effort. It would, however, bring the finctional and historical game much closer together to lobby for real improvements in the game.

[Curtis]

Originally Posted by knockahoma

Again, Diamond Mind is now including the ability to edit players and create players for that fictional market, working downward from a top-notch historical premise. I think Marcus has the lead in making the perfect blend of historical and fictional, but it appears the historical boys are expanding into fictional territory.

I played the computer Strat from 1997-2002 (1996-2001 seasons), and in all those versions you could create your own players. It was a pain in the butt, but it was possible.

QUOTE: ActionJackson — "Point is there are tons of variables within each PA and there are what...70-80 PA per game between two teams and 4,860 (approx) games in a regular season. Where are the numbers people? How many different possible outcomes are there over the course of your average baseball season? I would say plenty enough to never have the same season twice and no baseball simulation on the planet, no matter what they say, will give you the same season twice. Even if all the league wide numbers are similar, the results will be different due to the timing of scoring that occurs for each team over the course of a season."

Back in the day (see above), Strat would give excellant replayability. In my first season (12 humans and 16 computer teams) my best friend lost out on the AL wildcard by one game when the computer never brought Mike Piazza and Frank Thomas back from the DL after their injuries healed in May. To compensate for this the commissioner decided that we'd run the league three times from then on, and average the results.

Over the next five seasons my best and worst results for any given season were never more than two wins apart, and I don't remember anybody having more than a five win range. Specific players might vary a lot — I might have three different pitchers lead my team in wins, or three different batters lead in home runs, or Trevor Hoffman might save 53 in one replay and 35 in another, but the team results were amazingly consistent. And Trevor ALWAYS lead the league in saves, no matter how many or how few he ended up with.

[Garlon]

Here are some of the results of a sim I ran with OOTP10, 1871-2008.

Scouting/Coaching Disabled
Suspensions/Injuries Disabled
Position Player Fatigue High
Player Development Disabled with Talent Change Randomness set to 1
Lineup Selection Traditional
AI Evaluation options: Ratings Weight 100%, Currrent/Prev Stats Weight 0%
DH Disabled all seasons both leagues
Rookie Draft set to Dec 1st using Predefined Draft Values
Financial System Disabled
Automatically Adjust League Strategy Disabled (see below)
Retire According To History Enabled
Ratings Recalc Base 3 years
Base Ratings on Neutralized Stats
Base Fielding Ratings on Entire Career
Base Pitcher Stamina on Entire Career
Automatically Adjust LTMs Enabled
All Park Factors set to Neutral all seasons
Custom neutral weather file also used

Strategy Settings used for all seasons
Use Relievers Very Often
Use Closers Often
Pitcher Stamina Low
Pinch Hit for Pitchers Very Often
Pinch Hit for Positions Players Rarely
Defensive Subs Rarely
Stealing Very Often
Hit and Run Normal
Bunting Normal

Rotations Size
1871-1881 2-man
1882-1885 3-man
1886-1900 4-man
1901-2008 5-man

Schedule Length
1871-1875 54 games
1876-1881 82 games
1882-1885 126 games
1886-1900 140 games
1901-2008 162 games

Career BA
Hornsby .361
Cobb .354
Ruth .351
Gwynn .349
Delahanty .349
Pujols .348
Lajoie .346
Burkett .344
O'Doul .342
Suzuki .340

OPS Career
Ruth 1.160
Pujols 1.0623
T. Williams 1.059
Gehrig 1.056
McGwire 1.053
Ba. Bonds 1.027
Bagwell 1.007
Hornsby 1.005
M. Ramirez 1.002
F. Thomas .997

Stolen Bases Career
Raines 1093
W. Wilson 1003
Cobb 955
Henderson 948
Campaneris 936
lofton 861
Collins 857
Butler 816
Wagner 800
Nixon 784

HR Career
Bonds 836
McGwire 831
Killebrew 822
Ruth 785
Mays 775
T. Williams 772
Griffer 767
Aaron 751
Foxx 745
Thome 739

Career Wins (Top 40, incl. 22 300-game winners)
Bender 378
Quinn 354
Ruffing 351
Clemens 349
W. Johnson 349
Blyleven 344
T. John 344
Maddux 344
Ryan 338
C. Young 338
Simmons 330
Spahn 330
Vance 321
Adams 314
Seaver 311
Mathewson 309
Schilling 308
Sutton 308
Roberts 307
Glavine 306
Plank 305
Faber 300
Tiant 298
Pennock 297
Trout 297
Alexander 295
Smoltz 294
Reuschel 293
Haines 292
Rowe 291
Grove 290
K. Brown 289
Feller 288
Moyer 287
Reuss 286
Morgan 285
R. Johnson 282
Tanana 281
Wise 281
Hoyt 279

3000+ Strikeouts
Ryan 6089
Clemens 5318
R. Johnson 5276
Schilling 4218
Smoltz 4199
Blyleven 4073
Maddux 3809
Seaver 3807
Gibson 3578
Sutton 3577
Rijo 3556
Feller 3385
K. Brown 3371
Cone 3338
Mussina 3250
Fernandez 3243
Tanana 3219
Tiant 3175
Gooden 3162
W. Johnson 3161
Koosman 3154
Leiter 3120
Moyer 3118
Koufax 3104
Bender 3099
Lolich 3090
Vance 3079
Spahn 3065
T. John 3028
Ford 3014

MVP Awards
Cobb 11
Ruth 12
T. Williams 15

[Siv]

Garlon
So this was using your database as linked below? Cool stuff and IMO belivable results.

[Garlon]

I used the in-game DB. You can choose either real-life stats or neutralized stats. The neutralized stats are essentially the same as the 750-run environment stats you can see on baseball-reference. The neutralization process puts players across the league on an even footing in terms but it does not inflate stats. The neutralized stats set also fils in the gap seasons of players so that no players are ever missing upon initial import, and it also pro-rates any low AB seasons up to 250 so when you use 3-yr recalc the game is always sampling at least 500 AB for all players - this provides more realistic ratings.

[risp2out]

Garlon,

Did players who took time off for military service, etc. in real life also do so in your sim or did they play through. Ted Williams' stats are the reason I ask. Thanks.

[PhillieFever]

Originally Posted by risp2out

Garlon,

Did players who took time off for military service, etc. in real life also do so in your sim or did they play through. Ted Williams' stats are the reason I ask. Thanks.

I'm confident that i can speak for Garlon when I say that all of the "war players" played through the war years.

[Garlon]

They play through.

Ted Williams
Stat: Real vs Sim
HR per 550 AB: 37.18 vs 35.05
BA: .344 vs .334
OBP: .482 vs .464
SLG: .634 vs .595
OPS: 1.116 vs 1.059
Most HR in a season: 43 vs 44

Babe Ruth
Stat: Real vs Sim
HR per 550 AB: 46.76 vs 44.15
BA: .342 vs .351
OBP: .474 vs .490
SLG: .690 vs .671
OPS: 1.164 vs 1.160
Most HR in a season: 60 vs 60
50+ HR seasons: 4 vs 4

[rasnell]

Garlon: Is the 1871 quickstart file valid for OOTP X? Can you tell a little more about it. Definitely going to use your settings for next historical league. Thanks for sharing.

[knockahoma]

Ted Williams
Stat: Real vs Sim
HR per 550 AB: 37.18 vs 35.05
BA: .344 vs .334
OBP: .482 vs .464
SLG: .634 vs .595
OPS: 1.116 vs 1.059
Most HR in a season: 43 vs 44

Babe Ruth
Stat: Real vs Sim
HR per 550 AB: 46.76 vs 44.15
BA: .342 vs .351
OBP: .474 vs .490
SLG: .690 vs .671
OPS: 1.164 vs 1.160
Most HR in a season: 60 vs 60
50+ HR seasons: 4 vs 4

That's some damn impressive replay. . I see some setting changes that may help out my situation. I'll try them, or maybe this will all be moot by today's (?) release?

Out of morbid curiousity, how did my two pets-- Niekro and Belanger-- come out in this sim?

[wiggins]

Garlan, Thanks for the setup. I have a question. I play out just about every game. I'm going to start with the 60 Reds. With the settings of no injuries, a 5 man roatation, won't I loose some of the realism of actually playing the game? I would love my stats at the end of 20 years to be in line with actual numbers, but I want to imerse myself in the game. i've talked about this before. I guess I want it both ways. I'll probably shut off trades. Iin version 9 the 60 yankees kept waiving Elston Howard, no matter what I did ( I play in God mode) The AI couldn't figure out the 3 catchers (Howard, Berra, Blanchard). I/m not looking to sim 100 years just to look at the numbers. If I sim 100 years I then want to be able to play the game as it was at that time. (4 man roatations, injuries etc...) I don't know if this makes sense at all. As Ive mentioned before I'm still trying to wrap my arms around what I want and expect it to be. I've played fictional for 4 years now. I just started a Historical a few months ago, then the database got corrupted, so I'm waiting for 10.

[PhillieFever]

Originally Posted by pstrickert

IMO, the batting and pitching models work quite well for single-season replays (i.e., they produce highly credible results). The fielding model, however, is not up to par. It is no surprise, then, that (in my experience) teams that excel in both pitching and defense do not "translate" adequately into the OOTP world. Markus indicated that he'll work on the fielding model in OOTP11. In the meantime, it would be wonderful if someone created a utility to make fielding modifications easier. Don't know if that'll happen, though.

This is so true,the defensive part of the game remains a sore spot and really needs to be overhauled along with the financial model to make the game more of a complete sim.

[PhillieFever]

Originally Posted by wiggans1

Garlan, Thanks for the setup. I have a question. I play out just about every game. I'm going to start with the 60 Reds. With the settings of no injuries, a 5 man roatation, won't I loose some of the realism of actually playing the game? I would love my stats at the end of 20 years to be in line with actual numbers, but I want to imerse myself in the game. i've talked about this before. I guess I want it both ways. I'll probably shut off trades. Iin version 9 the 60 yankees kept waiving Elston Howard, no matter what I did ( I play in God mode) The AI couldn't figure out the 3 catchers (Howard, Berra, Blanchard). I/m not looking to sim 100 years just to look at the numbers. If I sim 100 years I then want to be able to play the game as it was at that time. (4 man roatations, injuries etc...) I don't know if this makes sense at all. As Ive mentioned before I'm still trying to wrap my arms around what I want and expect it to be. I've played fictional for 4 years now. I just started a Historical a few months ago, then the database got corrupted, so I'm waiting for 10.

Turn injuries on low,fatigue on high.Personally,I use the default game settings as far as strategy goes and I don't see much of a problem.I also play my games out so these are necessary adjustments to make.

[wiggins]

Philly,
Are you happy with your results? Is it possible that you could give me all your other set up options. I want to get off to the right start.

[Garlon]

rasnell - no don't use that old QS file. You can just directly start a league in 1871 with OOTPX.

knockahoma -
P. Niekro: 210-238, but also had 391 saves. Went back and forth from SP to CL during his career. 3.81 ERA, 3405 IP, 1.35 WHIP

J. Niekro: 253-280, 3.90 ERA, 4708 IP, 1.36 WHIP

Belanger: 3130 AB, .240/.310/.297, .608 OPS

wiggans1 - I think that by setting Player Fatigue to Very High that in a way simulates day to day injuries that prevent guys from playing 162 games. However, you don't get DL-length injuries without injuries enabled. You could try putting Injuries on Very Low.

pstrickert - individual player fielding ratings are now editable from within the DB. OOTP will import pre-determined fielding ratings directly from the master csv file now ehen you select Base Fielding Ratings on CAREER. It is possible to open up that file and tinker with the values. All of the values in there are calcualted based on a player's career defensive stats at his primary Infield or Outfield position compared to the league average player at that position during his career. The ratings look much better now. We're improving.