For those who suspect the game secretly protects the HR record...

[uruguru]

What is realistic for me is Mantle having more power than Maris. I see Maris' incredible 61-homer season as what it was, a fluke for his career.. an aberration. I have no problem with Maris occasionally hitting 61-63 home runs in a 1961 sim but anything higher than that feels weird.

On the other hand, I see Mantle's 54 homers that year as a good season for him, but closer to typical for his peak. I would have no problem with Mantle regularly hitting more homers than Maris in a 1961 season sim, and hitting 61-63 homers as well. In fact, I would be more comfortable seeing Mantle do this more than Maris.

I would wager that the pitchers for 1961 would agree with me that Mantle had more power than Maris. Let's not forget that Maris, despite hitting 61 homers and winning the MVP, had the luxury of having Mantle hitting behind him in the lineup. How many intentional walks did the future 61-homer MVP actually get that year? ZERO.. so there was no point in the season when a pitcher would have rather faced Mantle than Maris.

The problem lies in how much variability do you want in a sim? Too much and you get unrealistic performances. Too little and it's too predictable. These are all subjective measures, imo, so there's no way for the OOTP dev to develop a solution that makes everyone feel right.

The HR record in particular is so iconic that something always feels "wrong" when someone like Dick Stuart hits 65 homers in a season. It breaks immersion and reminds us that we are just playing a computer sim. And since it's on a computer, we can then conclude that this "wrongness" is a problem with coding. In the end, I suspect there is no way to tone down the variability without breaking the model and railroading player performances as they get near the end of the season. I don't think anyone wants that.

But OOTP lets us edit player ratings so, in the end, it's a problem that each individual player can solve for themselves if they feel passionately enough about it. There's no reason, for example, why BradK can't develop what he thinks are more realistic ratings for Maris in 1961 and use them in his games, or share them with the community for anyone else who wants them.

[David Watts]

Originally Posted by HRBaker

Different peoples' expectations are the major issue here I think. Everyone has a different idea how a historical replay should occur. Some want "almost-exact" results and others want "believable" results. I believe OOTP as it currently stands is "believable". I see myself in the second group. An "exact" replay has already happened historically. I'm looking for variation.

This discussion also brings up how different fantasy scenarios are. Since there are no historical expections to meet, the experience is more like viewing the MLB for the first time. "Anything can happen."

(2 cents)

I think you would find that the vast majority of historical players have zero interest in "almost-exact" results. Brad isn't asking for exact results. Thehef, isn't asking for exact results. I play straight historical , but the majority of my time is spent playing random debut. Exact results don't even factor in when it comes to random debut play, as it's basically fictional play with real players using the historical season as an outline for results.

[quillenl]

Originally Posted by David Watts

I play straight historical , but the majority of my time is spent playing random debut. Exact results don't even factor in when it comes to random debut play, as it's basically fictional play with real players using the historical season as an outline for results.

3 year Recalc Random Debut. This is the way.

[David Watts]

Originally Posted by quillenl

3 year Recalc Random Debut. This is the way.

I'm using 5 year, but I've nothing against 3 year. Chose 5 year with hope that if I get any of the war year guys, they won't get crushed

[jcard]

Originally Posted by Brad K

No theory. Mantle was better vs LHP than vs RHP and Maris was terrible against LHP.

Incorrect. The batter platoon splits in OOTP will comprise three factors:

1) Batter Split Ratings
2) Park Splits
3) Pitcher Splits

If the average pitcher has better ratings against same-handed batters than opposite-handed batters (including switch-hitters, of course) this will impact batter splits.

Note that the editor does not take batter handedness into account. Players with the same ratings will have the same projected batting stats, indicating that the projections do not include the effect (based on batting handedness) of the typical pitcher split.

Interestingly, the projected pitching stats DO take into account batter handedness, though not the batter ratings splits. Instead, the assumed proportions in the projection model of LHBs and RHBs faced changes based on pitcher handedness, and correspondingly different weights are applied to each of the pitcher ratings split.

Anyway, my comment about handedness that you summarily- not to mention, wrongly- dismissed actually weighs in on LOWERING SHBs ratings relative to LHBs and RHBs (and, to a lesser degree, LHBs relative to RHBs) to compensate for the omission of presumed pitcher ratings splits in the batter projections.

[Syd Thrift]

Yeah, even at the time the park was considered a big part of what gave Maris that last boost (even though 1 year park factors weren't actually super great). But yeah, there are a whole, whole bunch of other factors and if you really and truly want to see how often Maris breaks that record in 1961, you have to go back and do replays. There just plain isn't a way to easily math this out from the ratings. That's not really how the game works.

[jcard]

Originally Posted by Syd Thrift

Yeah, even at the time the park was considered a big part of what gave Maris that last boost (even though 1 year park factors weren't actually super great). But yeah, there are a whole, whole bunch of other factors and if you really and truly want to see how often Maris breaks that record in 1961, you have to go back and do replays. There just plain isn't a way to easily math this out from the ratings. That's not really how the game works.

I agree. Personally, I have seen the quote from a developer about projected stats and results corresponding, and think that they should have been clearer about particular factors (beyond the noted seasonal modifiers) outside of the projection model- particularly when the question at hand involves comparing two particular players and extreme right-tail performances. I think that the comment may have been taken too literally, but I am no expert (I just cannot get invested in a historical mode that is based on broad-brush algorithms; not a criticism, as I do not think OOTP ever intended to be or advertised itself as offering a curated historical mode).

[jcard]

Originally Posted by Syd Thrift

Yeah, even at the time the park was considered a big part of what gave Maris that last boost (even though 1 year park factors weren't actually super great). But yeah, there are a whole, whole bunch of other factors and if you really and truly want to see how often Maris breaks that record in 1961, you have to go back and do replays. There just plain isn't a way to easily math this out from the ratings. That's not really how the game works.

Syd- If you have not read “Baseball’s Best Sluggers” by Michael E. Schell, I think you might really enjoy it. I can elaborate on this if you wish. (Note- if you search for it, make sure the title is as given; he had previously written an earlier, less robust book with the similar title “Baseball’s Best Hitters”.)

[locuspc]

Originally Posted by MathBandit

Maris *is* rated exactly as his performance dictates. His performance does *not* dictate hitting 61 HRs, though.

This is the crux of it. It's a philosophical disagreement as to how to impute ability from performance, about how much "regression to the mean" to expect. It's fair to disagree with the philosophical choice they've made, but it's not fair to accuse them of malice because of it.

[Brad K]

Originally Posted by David Watts

I think you would find that the vast majority of historical players have zero interest in "almost-exact" results. Brad isn't asking for exact results. Thehef, isn't asking for exact results. I play straight historical , but the majority of my time is spent playing random debut. Exact results don't even factor in when it comes to random debut play, as it's basically fictional play with real players using the historical season as an outline for results.

I'm not asking for exact results. I'm asking for plausible results.

Maris never approaching 61 in dozens of my saves isn't plausible. My current Pirates Moneyball 1951 - 2008 where the season record is 53 isn't plausible. In five year of playing OOTP I've never had any player hit 61 in a year or 714 in a career. That's not plausible. Neither is me having 6 .380+ season BAs 51 - 08 in Pirates Moneyball.

OOTP is great on overall game performance. It hasn't acknowledged that randomness alone will not produce plausible individual performances.

[Brad K]

Going back to what David posted in #29, Mantle is rated 49 on for a season he hit 54 and Maris is rated 45 on a season he hit 61.

Park factors have been mentioned. Well, for the 45 to really mean 60 for someone playing half his games in Yankee Stadium would require a LHB HR park factor of 166.

[Garlon]

In 1961 Maris had 61 HR in 590 AB.

OOTP sets everything to per 550 AB in the editor.

(61/590)*550 = 56

However, in the editor he has 45 HR per 550 AB when using real stats because OOTP adds an adjustment.

If you take a look at his neutralized stats he has 56 HR in 590 AB because Yankee Stadium in 1961 had a good HR factor for LHB that season (actually 3 seasons).

(56/590)*550 = 52

In the editor he gets 52 HR. There is no additional adjustment that you get when you use real stats. I had to beg Markus to give us this a few years ago when using neutralized stats.

So even though his neutralized stats are 56 HR in 590 AB and his real stats are 61 HR in 590 AB, he will import with a higher HR rating when using neutralized stats because you do not get the additional adjustment you get with real stats when you play the game.


Maris real HR: 275

Maris neutralized HR: 260


Maris real BA: .260

Maris neutralized BA: .261


Maris real OPS: .822

Maris neutralized OPS: .818

The neutralized stats are actually fairly subtle adjustments for players, but the biggest advantage is that you get better imports for players.


Ty Cobb 1912 real BA: .409 (226/553)
Ty Cobb 1912 neutralized BA: .410 (227/553)

Ty Cobb real stats 1912 BA in the editor: .376

Ty Cobb neutralized stats 1912 BA in the editor: .411


Maybe consider trying neutralized stats for your games.

[Brad K]

Hi Garlon. I'm glad you joined the thread.

My impression is that with N stats the game's park factors are irrelevant, that a player will perform the same in any park. Is this correct?

[Brad K]

I was expecting to make a joke that Yankee Stadiums high LH HR park factor was bootstrapping because Mantle and Maris it a lot of HRs there. Surprise! Mantle was 24 at home 30 on the road and Maris was 30 and home and 31 on the road.

[Garlon]

When I made the neutralized stats I took out the effect of the park factors so that we get a better idea of how the player would perform in a neutral park in a given season. When you play OOTP the park factors you are using in any season will still apply to the outcomes. So, you have Maris importing at 56 HR per 590 AB but playing in a Yankee Stadium for half the season in a favorable environment.


Consider a player who hit 50 HR in 500 AB playing half of their games in a stadium with a 1.20 HR factor in a league with 8 teams. The entire league will have an average HR factor of 1.000.

For this given player their home factor is 1.20, and their road factor is (8 - 1.20)/7 = 0.971

We combine these and divide by 2. (1.20 + 0.971)/2 = 1.085

This batter had a 8.5% advantage over the season in terms of HR. We take the 50 HR and divide by 1.085 and get 46 HR instead of 50.

Most players do not end up with a significant adjustment.

Clemente Real:
H: 3000
2B: 440
3B: 166
HR: 240
BA: .317
OBP: .359
SLG: .475
OPS: .834

Clemente Neutralized:
H: 2966
2B: 428
3B: 124
HR: 288
BA: .314
OBP: .356
SLG: .477
OPS: .833

[OutS|der]

Originally Posted by Garlon

In 1961 Maris had 61 HR in 590 AB.

OOTP sets everything to per 550 AB in the editor.

(61/590)*550 = 56

However, in the editor he has 45 HR per 550 AB when using real stats because OOTP adds an adjustment.

If you take a look at his neutralized stats he has 56 HR in 590 AB because Yankee Stadium in 1961 had a good HR factor for LHB that season (actually 3 seasons).

(56/590)*550 = 52

In the editor he gets 52 HR. There is no additional adjustment that you get when you use real stats. I had to beg Markus to give us this a few years ago when using neutralized stats.

So even though his neutralized stats are 56 HR in 590 AB and his real stats are 61 HR in 590 AB, he will import with a higher HR rating when using neutralized stats because you do not get the additional adjustment you get with real stats when you play the game.


Maris real HR: 275

Maris neutralized HR: 260


Maris real BA: .260

Maris neutralized BA: .261


Maris real OPS: .822

Maris neutralized OPS: .818

The neutralized stats are actually fairly subtle adjustments for players, but the biggest advantage is that you get better imports for players.


Ty Cobb 1912 real BA: .409 (226/553)
Ty Cobb 1912 neutralized BA: .410 (227/553)

Ty Cobb real stats 1912 BA in the editor: .376

Ty Cobb neutralized stats 1912 BA in the editor: .411


Maybe consider trying neutralized stats for your games.

Where has it been stated what totals the editor uses to get it's "neutralized stats"

Does it use the current year? I' always wondered

[Garlon]

When using neutralized stats you can see both their real stats and neutralized stats. There are additional files in the DB with the neutralized batting and pitching statistics for all seasons.

[uruguru]

Originally Posted by Brad K

I was expecting to make a joke that Yankee Stadiums high LH HR park factor was bootstrapping because Mantle and Maris it a lot of HRs there. Surprise! Mantle was 24 at home 30 on the road and Maris was 30 and home and 31 on the road.

Wasn't #61 to that short right field porch?

[Brad K]

I think I get it. Neutralization is the numbers a player's real life career would have produced in a neutral park. If in OOTP you put him in his historic park his rated performance matches historical, if in another park it doesn't.

So this leaves the question about what are real stats because to me yours are as real as it can get.

[Brad K]

Originally Posted by uruguru

Wasn't #61 to that short right field porch?

No idea. I wasn't aware of baseball at the time it happened and don't recall ever seeing a video of it.