Historical replay accuracy

[Malleus Dei]

Originally Posted by knockahoma

And I still have to figure out how to make the AI see the value in Mark Belanger types.

Since Blade just totally sucked as a hitter - he was the Triple Crown loser in 1970 (dead last in all three categories), and had a lifetime batting average of .228 in eighteen seasons with twenty HR - his value was in his defense. Now if defense is, as some claim, as much as one third of a player's overall value, then every other candidate to play SS for the team he was on would have to both be a poor hitter (as opposed to a hideously awful one, like Belanger) and be fielding the position really badly (as in with a machete and not a glove) in order to make Belanger the best option available. An average hitter/average fielder would be a far better choice.

Belanger should be no more than a defensive replacement.

[Markus Heinsohn]

Originally Posted by malleus dei

since blade just totally sucked as a hitter - he was the triple crown loser in 1970 (dead last in all three categories), and had a lifetime batting average of .228 in eighteen seasons with twenty hr - his value was in his defense. Now if defense is, as some claim, as much as one third of a player's overall value, then every other candidate to play ss for the team he was on would have to both be a poor hitter (as opposed to a hideously awful one, like belanger) and be fielding the position really badly (as in with a machete and not a glove) in order to make belanger the best option available. An average hitter/average fielder would be a far better choice.

Belanger should be no more than a defensive replacement.

+1

[Malleus Dei]

Originally Posted by Markus Heinsohn

+1

[actionjackson]

Originally Posted by knockahoma

... And I still have to figure out how to make the AI see the value in Mark Belanger types.

I find giving it a good whack helps! Maybe drop it on its head?

Seriously though, I really hope these settings get you closer to where you want to go. I would say I hope they give you exactly what you want, but closer is good...you know baby steps and all that crap. I hope everyone who loves to play historical seasons like Styx, yourself, myself and others gets better results with these settings. I understand Styx's and others' frustration though, but I probably experience it to a somewhat lesser degree.

[Jah]

Originally Posted by Malleus Dei

Since Blade just totally sucked as a hitter - he was the Triple Crown loser in 1970 (dead last in all three categories), and had a lifetime batting average of .228 in eighteen seasons with twenty HR - his value was in his defense. Now if defense is, as some claim, as much as one third of a player's overall value, then every other candidate to play SS for the team he was on would have to both be a poor hitter (as opposed to a hideously awful one, like Belanger) and be fielding the position really badly (as in with a machete and not a glove) in order to make Belanger the best option available. An average hitter/average fielder would be a far better choice.

Belanger should be no more than a defensive replacement.

Well, Earl Weaver and the Orioles didn't think so. In an eleven-season period from 1968 thru 1978, he played in 142 games or more (at short) nine times, won 8 Gold Gloves (including 6 in a row from 73 thru 78)...and his team won 5 divisions, 3 AL pennants, and a World Series.

In all those years, I guess the Orioles never found that "far better choice" you speak of.

[pstrickert]

baseball-reference.com has some neat sabermetric fielding stats. Rtz (total zone fielding runs) is one of them. It calculates the number of runs below or above average a player was worth based on fielding plays made. Here's the Top 10 (of 421 players) in the AL in 1974:

Belanger: 17.5
Doyle: 15.2
Scott: 14.2
Nettles: 14.0
B. Robinson: 13.7
B. Bell: 13.2
D. Evans: 13.0
Duffy: 12.8
Blair: 12.6
Brohamer: 10.8

When I sim the 1974 season, I often see Denny Doyle relegated to the CAL reserve roster. He actually hit a respectable .260 in 1974. Still, the AI will use Winston Llenas at 2B. Llenas hit .261 in 1974. His Rtz at 2B was -0.4.

BAL and OAK both had great pitching and great defense in 1974. When I sim the 1974 season, though, both teams have trouble replicating their success. Why? Either because (1) defensive ratings are out of whack -- or because (2) the game engine does not adequately consider defense in play outcomes.

Rtz ratings "up the middle" for BAL and OAK in 1974:

Belanger-SS: 17.5 (#1)
Grich-2B: 5.2 (#25)
Blair-CF: 12.6 (#9)

Campaneris-SS: 6.1 (#22)
Green-2B: 8.0 (#15)
North-CF: 5.5 (#24)

. . . out of 421 players in the AL.

[actionjackson]

Originally Posted by Malleus Dei

Since Blade just totally sucked as a hitter - he was the Triple Crown loser in 1970 (dead last in all three categories), and had a lifetime batting average of .228 in eighteen seasons with twenty HR - his value was in his defense. Now if defense is, as some claim, as much as one third of a player's overall value, then every other candidate to play SS for the team he was on would have to both be a poor hitter (as opposed to a hideously awful one, like Belanger) and be fielding the position really badly (as in with a machete and not a glove) in order to make Belanger the best option available. An average hitter/average fielder would be a far better choice.

Belanger should be no more than a defensive replacement.

Gotta disagree with you Mal and I guess Markus. Why? Context. Sure, we're sitting here in 2009 looking at Mark Belanger, wondering how in the world the Orioles could've played him all those years at SS with that anemic bat, but back then SS was all about the defense (and IMHO should still be defense first, then offense) and there were a lot of horrible bat great glove shortstops around. If the Orioles pitching staffs of the Belanger era knew what was good for them, I'll bet Belanger didn't have to pay for a single meal while on the road.

Also, if one does as Garlon suggested and neutralizes the stats to a 750 run environment (or uses the neutralized stats database with neutralized parks), Mr. Belanger changes drastically:

Belanger (Career Standard Batting): .228/.300/.280

Belanger (Career Neutralized Batting): .257/.335/.314

Those numbers, combined with his wizardry in the field make Belanger my everyday guy and the OOTP X Orioles pitching staffs of the Belanger era thank me. You see, it's not just about Belanger. It's about the hit the pitching staff takes if he isn't in the lineup. I believe one of the common complaints of historical players is how the simmed Oriole pitching staffs of that era absolutely pale in comparison to their real life counterparts. I submit to you that the AI's reluctance to play Belanger is a major reason why. Now, hit me with your best shot "Surly Bastage" and genius OOTP inventor. :ducks: (No disrespect intended to either of you guys)

[Cowboys3356]

A lot of great stuff here, thanks to everyone that spends time testing this stuff so the rest of us can enjoy the results and have vastly improved leagues!

[Rondell Tate]

Remember, in asking to get Mark Belanger right, you're talking about modeling a freak. Yes, SS was a fielding-first position, but even then, Belanger was a lousy hitter (108-54 1970 Orioles, Belanger hit .218, OPS .562 OPS+ 56).
Calling him a banjo hitter underestimates how much pop you can put on a baseball when you hit it with a banjo.
And yet he played 145 games that year ... which tells me that a) Earl Weaver had nobody else available or b) Earl wasn't as smart as Earl's legend.
I'm guessing it's a bunch of a) with a bit of b):
In 1969, Belanger had his career year of .287/.351/.345 OPS .696 OPS+ 95
at age 25 (his second full season in the bigs) and you expect a young hitter to improve, so the Orioles probably didn't look for backup as a priority. Instead Belanger went backward, a long way.
In 1971, he bounced back and had another year almost identical to 1969 and at age 27 hit .266/.365/.320 for a .685 OPS and OPS+ of 97. For a great defensive SS, those are fine numbers.
But then he hit .186 in 1972 (.482 OPS, 42 OPS+) and .226 in 1973 (.564 OPS, 61 OPS+) and with one exception he never again got close to being just a poor hitter again.
I don't have time to run a sabermetric analysis of how far offensively below a replacement SS Belanger was in his bad years, but I bet he didn't save that many runs regardless of his defense.
In 1973, the AL went to the DH and having a guy hitting like a pitcher at the bottom of the order probably didn't seem like such a big deal because teams were used to restarting their offense after a vacuum in the 9 hole.
Had the Orioles not been so good in the early 70s, they probably would have looked for a new SS earlier, but having won with little to no offense from Belanger, they looked elsewhere for runs and he became a fixture, and every 2-3 years he had a season where he wasn't horrible, which undoubtedly helped carry him.
If you believe your SS is going to hit worse than many pitchers, regardless of his defense, you're going to go scouting for a new one. But Belanger went in cycles of off year/horrible year/mediocre year, and Earl probably hoped/expected Belanger would be mediocre most Aprils.
So to model this in OOTP, you need to model both Belanger correctly (including get the defense right, which is the hardest part of any baseball sim because the stats are often misleading and very tricky) but you have to model Earl Weaver's belief in his own mantra of pitching, defense and the 3-run home run. And Belanger was Exhibit A for what a smart guy Earl was, because there were like three pitchers Belanger could hit and Earl made a point of batting him second against those guys.
Most managers would not start Mark Belanger pretty much every game, and I don't believe there's any question he was a net negative the last half of his career ... from 1974 to the end of his run, his OBP topped .298 exactly once and from 1977 on he was dreadful. In 1979, when they made it back to the WS, Belanger was the defensive half of a platoon with Kiko Garcia ...
So expecting a computer to replicate such a bizarrely abnormal career ... short of hard-coding SS Belanger into the lineup ... seems like a big request.

[RMc]

Belanger stayed in the Orioles' lineup because:
1) he was a Gold Glove shortstop;
2) the Orioles were winning;
3) they had other guys who could hit, thank you very much, and
4) by all accounts, he was a great guy to have in the clubhouse.

According to B-Ref, he cost the O's 232 runs with his bat...and saved 238 with his glove. Works for me...

[jduteau]

Wonderful explanation of Belanger!

[pstrickert]

Earl Weaver isn't the only manager in that era to rely on a weak-hitting SS. Look at these stats for 1974:

AL Shortstops (1974)

minimum 81 starts at SS

Toby Harrah: .260 - 21 - 74
Frank Duffy: .233 - 8 - 48
Bucky Dent: .274 - 5 - 45
Mark Belanger: .225 - 5 - 36
Ed Brinkman: .221 - 14 - 54 (only time he hit double-digit HRs)
Freddie Patek: .225 - 3- 38
Jim Mason: .250 - 5- 37
Bert Campaneris: .290 - 2- 41
Robin Yount: .250 - 3 - 26
Dave Chalk: .252 - 5 - 31
Mario Guerrero: .246 - 0 - 23
Rick Burleson: .284 - 4 - 44

Maybe the problem is that the OOTP AI is applying modern (21st century) criteria to choose lineups. In other eras, other standards were used to measure the value of players. Unless the AI is able to adapt to different eras and styles of baseball, maybe it would be better if the AI relied on GS or PA by position to choose the starters.

[Jose Cardenal's Afro]

Originally Posted by Le Grande Orange

That's exactly the point—what does historical accuracy mean in this context? If Player A in 1974 typically batted third in the line up, played in 100 games, and hit .270, does that mean in OOTP he must typically bat third, play in 100 games, and hit .270? How far from those figures can the game diverge and still be considered historically accurate?

What I'm getting at is that any major league season is the result of an enormously interwoven tapestry of events. Once you start pulling on some of the threads, the results should not expected to be identical since you are changing the nature of the tapestry. OOTP already does this in one small way—the schedules used are the original schedules, not as they were played. This means variations in off days from how the real season unfolded, and consequently alterations to pitching rotations, which necessarily should be expected to impact pitcher usage. The effect may be small, but it's there, and the more of it that goes on the more the results may likely be different at the end. The same sort of thing would be true with a myriad of other small influences.

Basically, it comes down to this: what items in a historical league in OOTP are allowed to vary from reality, and to what degree? I don't think there'll be much consensus on this, especially once you get down to the specific details, and because much of it is subjective judgement.

The game cannot accurately represent where everyone batted. Just look at your favorite team from this year, barring injuries it doesn't match generally the batting order. There are so many variations on thinking by managers. That being said, if the players generally represent themselves accurately enough then this could be a valid historical sim.

[No Pepper]

Originally Posted by pstrickert

Maybe the problem is that the OOTP AI is applying modern (21st century) criteria to choose lineups. In other eras, other standards were used to measure the value of players. Unless the AI is able to adapt to different eras and styles of baseball, maybe it would be better if the AI relied on GS or PA by position to choose the starters.

I think to a degree, how big I don't know, you've hit this on the head. It is either a design decision and/or a carry-over from previous versions where the historical aspect of the game was less a focus.

[Rondell Tate]

Originally Posted by RMc

According to B-Ref, he cost the O's 232 runs with his bat...and saved 238 with his glove. Works for me...

Yeah, but his career numbers are pulled way up by his 3+ good years. He didn't slide up and down a little, he soared and plummeted ... and like I said, it's natural that you hope/expect he'll have a good year, or else you will go look for a replacement.
If you told Earl Weaver before a season that Belanger would hit .186 with no power, more Ks than BBs and so-so speed, he would go looking for a replacement.
And those numbers suggest that he was almost exactly average (+6 over 18 seasons is practically a flat line) despite his defense, and counting in his + hitting years.
That sounds about right to me. I'm not saying he was horrible ... just that he was a horrible hitter.

[Le Grande Orange]

Originally Posted by Jose Cardenal's Afro

The game cannot accurately represent where everyone batted.

It could, however.

Retrosheet box scores and play-by-play files record the starting lineups and batting orders. Get that data into a database of one sort or another, and OOTP then has the information with which to construct nearly 100% accurate batting orders and starting lineups for historical leagues. (Assuming, of course, one wants the game to specifically mimic reality that closely.)

[Garlon]

knockahoma-

The LTM's don't kick in until the second year of a league. So if you want a more realistic 1974 season you'd have to start in 1973 or 1972 and sim up until 1974. Then the LTM's have a good harness on controlling the league output.

The only other way to do it would be to run a simulation of 1974, then calculate all of the LTM's by hand based on the end of season stats, then start a 1974 all over again and put those LTM's in place before yo beginning simming. It'd be nice if OOTP would do an initial test run of a league to calculate year #1 LTM's for us, but we don't have that in place yet.

1974 League Totals
Stat: Real vs Sim
BA: .256 vs .255
OBP: .324 vs .323
SLG: .369 vs .367
OPS: .693 vs .690
HR per 1000 AB: 20.03 vs 18.87
BB per 1000 AB: 98.01 vs 98.27
K per 1000 AB: 147.40 vs 139.02
ERA: 3.62 vs 3.58
Runs per 27 outs: 4.13 vs 4.25
BABIP: .284 vs .281

Batting Average
Carew .339
Rose .337
Madlock .328
R. Jackson .328
Oliver .316
Stargell .315
Munson .313
Singleton .312
Morgan .310
Bench .310

OPS
R. Jackson 1.006
McCovey .969
Schmidt .958
Bench .935
Morgan .929
Stargell .921
Bonds .877
Hebner .860
Rose .858
F. Robinson .855

HR
Scmidt 39
McCovey 36
R. Jackson 35
D. Johnson 34
F. Robinson 34
Wynn 33
Bench 30
C. Johnson 27
Scott 27
Hebner 25

Strikeouts
Ryan 226
Montefusco 217
Seaver 210
Blyleven 203
Halicki 192
Matlack 191
Sutton 185
Tanana 184
Messersmith 178
Reuschel 175

ERA
Brown 2.21
Tiant 2.61
Sutton 2.65
Montefusco 2.66
Seaver 2.67
May 2.80
Halicki 2.89
Strom 2.99
Messersmith 3.00
Koosman 3.01

SB
North 98
Brock 85
Campaneris 71
Rivers 71
Cedeno 62
Bonds 61
Lopes 59
Bowa 53
Bumbry 45
Carew 42


Some Gold Gloves Overall in the Simulation
C - Berra 5, Dickey 5, Sewell 4
1B - Anson 12, B. Terry 8, G. Kelly 6
2B - Lajoie 9, Doerr 9, E. Collins 8, Carew 5
3B - Brooks Robinson 10, Gaetti 9, Kamm 7
SS - Bancroft 9, Aparicio 7, O. Smith 6
OF - Speaker 10, Mays 6

[knockahoma]

LOL,

Pstrickert. You'd better hope it's "great minds" that think alike. I just posted this on the Belanger in X thread. Totally agree. You gotta look at Belanger in context:

I think it's important to remember the talent pool at SS during Belanger's age.

When you look at him in context of the other shortstops in the AL, you begin to appreciate why his amazing defense was so valued and his offense of such little concern. There was only one regular SS who hit over .270 in the AL that year, and one or two mutants who hit over 10 homers. But there are few on this list who hit well enough to bench Belanger's stand out defense. I won't even list Belanger's competition on his own club (Frank Baker? Puhleeeeze )

Baseball-reference.com listed these guys as their team starters in 74.


Quote:

BOSTON MARIO GUERRERO .246 0hr

CAL DAVE CHALK .252 5hr

CHI BUCKY DENT .274 5R

CLE FRANK DUFFY .233 8 HR

DET ED BRINKMAN .221 14hr

KC FREDDIE PATEK .225, 3HR

MIL ROBIN YOUNT .252, 3HR

MIN DANNY THOMPSON .250, 4HR

NY JIM MASON .250, 5HR

OAK BERT CAMPANERIS .290, 2HR

TEX TOBY HARRAH .260, 21 HR

BAL MARK BELANGER .225, 5hr


--------------------------------------------------------------------------------

Remember too that back in the day a guy was pretty much his natural, God-given physique. Weight lifting was not in vogue. Quickness off-set lack of muscle and vice versa in a much more pronounced way than what we see today.

Thanks, G. I know by your efforts that the historical is a big deal to you, as well. I was hoping you could post some of the 74 individual stats by team, or at least for one division. Maybe that's too big to do on this board. Of course, I can run my own sim, but I've done that a gazillion times for the 74 season. I'm not happy with some of what I see. I was curious what your test showed.

[knockahoma]

Just for fun, when viewed in the historical context, if the Orioles had Belanger and any one of the other shortstops in the AL in 74, who would have replaced Belanger?

Mario Guerrero.

Dave Chalk. Nice try. Decent shortstop with nothing but a .250 BA to offer.

Bucky Dent. Good shortstop, .250 average. I think the Orioles figured they were covered with the gold-glover.

Frank Duffy. Good shortstop. Bad hitter. And Belanger won the 8 gold gloves.

Ed Brinkman. Hmmm, .225 BA and 9 extra home runs? Brinkman was a pretty decent shortstop, but very late in his career.

Freddie Patek. .225, fewer home runs and some good speed. Nope.

Robin Yount. .252, 3HR at the age of 18? Sure. They all saw Yount coming. If the Orioles could have swung a deal. But, for 74, Belanger was probably the better overall talent.

Danny Thompson. .250-ish, 4 HR. A tragic cancer story. Nowhere near Belanger's talent with the glove.

Jim Mason. .250-ish was a really good year for Jim, who knocked around the majors for about five years, was it? Never had the Belanger touch on defense.

Bert Campaneris, .290, speed. Ah, might have been interesting. Belanger might have lost out on this one, even with Weaver.

Toby Harrah, .260, 21HR. A mutant hitter at SS! Harrah and Belanger on the same team? Easy. Toby goes to 3B (where he played a lot in his career) and the Orioles deal with the Brooks Robinson fans, just as Decinces did two years later.

When you look at it in context, Weaver and the Orioles had a right to be proud of their shortstop. May he rest in peace.

[pstrickert]

How do the stats break down between AL and NL in 1974 (BA and ERA)? Thanks!