Home Field Bug still Present in 4.0c

Peter Bergstrom · 04-03-2002, 09:38 AM

Home teams are still losing >50% of all home with the new patch. Try simming a few seasons; it's all too evident.

Statistically, the home team should prevail about 55% of the time. Not in OOTP4, however. Seems like a simple matter to fix, odd that it wasn't! <img src="confused.gif" border="0">

[ 04-03-2002: Message edited by: Peter Bergstrom ]

Hammer755 · 04-03-2002, 09:57 AM

[quote]Originally posted by Peter Bergstrom:
Home teams are still losing >50% of all home with the new patch. Try simming a few seasons; it's all too evident.

Statistically, the home team should prevail about 55% of the time. Not in OOTP4, however. Seems like a simple matter to fix, odd that it wasn't!

[ 04-03-2002: Message edited by: Peter Bergstrom ]<hr></blockquote>

Peter, I've seen this vary largely from league to league. I just did some quick research on a couple of online leagues that I'm in. A modern-day, fictional league, after 3 complete seasons, yielded a home-field record of 1831-1865, which is a 0.495 W%. However, in another league, a dead-ball league with real players, home teams went 2701-2668 in 4+ seasons, which is a 0.503 W%. So, I imagine a league can have a pretty large swing from one end of the spectrum to another. It's tough to generalize by saying home teams always have a losing record without a ton of collected data, which is the case with pretty much any trend or stat included in OOTP.

Hammer755 · 04-03-2002, 10:39 AM

[quote]Originally posted by blynch10:

10 teams with 55% or more winning % at home.
<hr></blockquote>

This data set has a total Home W% of 0.515, but as I said before, one season cannot prove or disprove any theory.

blynch10 · 04-03-2002, 10:45 AM

I was not trying to prove or disprove the theory on better winning % at home or not. Just showing that last year the huge swing between % for the teams. There are too many variables at work to say that teams will always win more at home. If true the Reds would have to do better then 26 and 53 at home.

The numbers are from the real MLB 2001 season standings.

[ 04-03-2002: Message edited by: blynch10 ]

[ 04-03-2002: Message edited by: blynch10 ]

Peter Bergstrom · 04-03-2002, 11:20 AM

[quote]Peter, I've seen this vary largely from league to league. I just did some quick research on a couple of online leagues that I'm in. A modern-day, fictional league, after 3 complete seasons, yielded a home-field record of 1831-1865, which is a 0.495 W%. However, in another league, a dead-ball league with real players, home teams went 2701-2668 in 4+ seasons, which is a 0.503 W%. So, I imagine a league can have a pretty large swing from one end of the spectrum to another. It's tough to generalize by saying home teams always have a losing record without a ton of collected data, which is the case with pretty much any trend or stat included in OOTP. <hr></blockquote>

Sorry, Hammer, that's way too low. Historically, the Home Field Advantage hovers around 55%. In fact last year it was 57 and 53 for the NL and AL respectively. It's a bug, albeit a minor one.

Scott Vibert · 04-03-2002, 11:44 AM

This isn't a bug, nor is it a design decision. The game doesn't give any artificial "home field advantage" to a team in this game. As a result the homefield winning percentage should tend to be around 50% or maybe a bit above or below in OOTP, each team that plays in this game will have as good a chance at beating another team at home as it does on the road. (I really wouldn't want to see some artificial homefield advantage built in).

I don't see this as a problem, and its not even an annoyance to me.

Hammer755 · 04-03-2002, 11:58 AM

[quote]Originally posted by Peter Bergstrom:


Sorry, Hammer, that's way too low. Historically, the Home Field Advantage hovers around 55%. In fact last year it was 57 and 53 for the NL and AL respectively. It's a bug, albeit a minor one.<hr></blockquote>

[Whiny Al Gore voice]I have to disagree with your numbers.[/Whiny Al Gore voice]

My brief research shows in 2001, the NL home W% was .517 and the AL W% was .528, which combines for a .522 W%. Those numbers were down across the board, as in 2000 it was NL-.543, AL-.537, MLB-.540.

[quote]Originally posted by Peter Bergstrom:It's a bug, albeit a minor one<hr></blockquote>

I wouldn't necessarily say it's a bug, but a design choice. A bug is when an aspect of the game does not behave as it was designed to do. Now, you or I may disagree with the design decision, but that doesn't make it a bug.

Hammer755 · 04-03-2002, 12:01 PM

[quote]Originally posted by ScottVib:
This isn't a bug, nor is it a design decision. <hr></blockquote>

I agree Scott, I didn't mean that the game's W% are artificially tweaked one way or another. By design decision, I was refering to the decision not to include this facet.

Peter Bergstrom · 04-03-2002, 12:35 PM

[quote] wouldn't necessarily say it's a bug, but a design choice. A bug is when an aspect of the game does not behave as it was designed to do. Now, you or I may disagree with the design decision, but that doesn't make it a bug. <hr></blockquote>

Then, with all due respect, I think it's a ill-judged design choice. There is a built-in home field advantage in all major sports and baseball is no exception. This is reality. Though not a game killer, to omit this factor is to render the game a bit less realistic. In my opinion.

All you have to do is raise the pitcher's ratings a bit at home and depress them a bit on the road. This isn't rocket science.

Scott Vibert · 04-03-2002, 12:43 PM

[quote]Originally posted by Peter Bergstrom:

All you have to do is raise the pitcher's ratings a bit at home and depress them a bit on the road. This isn't rocket science.<hr></blockquote>

And what is the reasoning for the pitcher being worse on the road then they are at home? Is the pitcher not the same individual? So why should he pitch worse on the road? Think of the skewed results you'll get from pitchers who have the misfortune of starting frequently on the road.

It's a good thing Nomo didn't react by your reasoning in his first start last year at Camden Yards, when he threw his no hitter.

Steve Kuffrey · 04-03-2002, 12:51 PM

I feel that making "across the board" ratings changes as you as saying should be done is not realistic. Some players play better on the road than at home.....

To make this adjustment no matter how slight I feel is not a good design model. To allow the players and teams to perform as they will whether at home or on the road to me is a better model, and this is the design that we have chosen.

Statistics may or may not bear out the home field advantage issue, but I would be curious how many sports simulation actually weigh this difference into their formulas? I know that Action PC Football give this as an "option". That may be something to consider in the future. But I feel to force the issue would not be right.

IngredientX · 04-03-2002, 01:00 PM

[quote]Originally posted by blynch10:
Also can anyone tell me why what is typed in the message is not formated the same once displayed int the forum.....All my nice columns are squished.
<hr></blockquote>

It's because the font used in the text box used to compose messages is Courier, which is a fixed-width font (all characters are the same width, which makes columns easy to create). But the actual message board uses a variable-width font. In this font, the letter m is wider than the letter i.

m
i

In Courier, these letters are the same width as each other, and every other character.

Hope this answers the easier question in your post.

Peter Bergstrom · 04-03-2002, 01:02 PM

[quote]And what is the reasoning for the pitcher being worse on the road then they are at home? Is the pitcher not the same individual? So why should he pitch worse on the road? Think of the skewed results you'll get from pitchers who have the misfortune of starting frequently on the road.
<hr></blockquote>

Crowd supoprt? Stadium familiarity? More sex? I dunno, Scott. That's the way it is. And pro football and basketball are even more extreme in this regard. Not factoring in HF advantage in these two sports would be hallucinatory. Why do you think bookies give away points when teams are away? In football, HFA is worth a field goal plus a fraction.

And the results won't get skewed because, over time, pitchers will split their home and away innings evenly. Don't worry.

Basically we're all here because we like sports and we like numbers. It's just a question of getting the nembers right. I just think it's remarkable that a young German guy could come up with such a convincing replication of our national game.

Peter Bergstrom · 04-03-2002, 01:10 PM

From a website:

[quote]Abstract: The World Series of baseball in the United States is a best-of-seven series. Therefore, the first team to win four games is the victor. To try to make the series fair, the first two games are played at one team's home park, the next three games are played at the other team's park (just two may be played if one team wins four in a row), and the final two games (if needed) are scheduled at the first team's park. It is therefore possible that one team will play four games in its home park while the other team may only play three games at home.

It is generally assumed that there is some advantage to playing at home. Each baseball park is different, so there may be some advantage to playing on a familiar field. The hometown fans provide emotional support for the team, and so on. Is there an advantage to playing World Series games on your home field?

The dataset contains information on the year, the two teams competing in the world series, whether the American League (AL) or National League (NL) team won the game, and the AL and NL winning at home or away percentage.

For all the games given, the teams won at home 55.5% of the time and away games were won 45% of the time. Using these probabilities and assuming that the chance of winning a home game is constant and that the games are independent, one can find the probabilities of winning in exactly 4, 5, 6 or 7 games as well as some other basic probability questions.

The 2-way ANOVA by League and Home/Away shows significance for both factors. While the interaction term is not significant at the .05 level, the interaction plot suggests that the home field advantage may be greater for American league teams than for National League teams. <hr></blockquote>

Scott Vibert · 04-03-2002, 02:23 PM

[quote]Originally posted by Peter Bergstrom:

And the results won't get skewed because, over time, pitchers will split their home and away innings evenly. Don't worry.

<hr></blockquote>

I still see no justification for increasing the home pitchers ratings vs. the road pitchers ratings, I just did a quick study from a tour of the stats from ESPN.

I took one team (Boston) and used the three year averages (1999-2001) of each pitcher on the squad the results are presented here:
* 3 year splits missing last years stats presented

[code]
Name Hm Av Hm ERA Hm IP Rd Av Rd ERA Rd IP
Arrojo .295 5.48 226.2 .243 2.31 190.0
Banks* .000 0.00 4.0 .192 1.35 6.2
Burkett .297 5.30 231.0 .253 3.43 270.0
Castillo .257 4.61 125.0 .227 3.31 149.2
Crawford .288 6.21 29.0 .237 2.50 36.0
Fossum .211 4.01 24.2 .312 5.95 19.2
Garces* .221 3.86 35.0 .217 3.94 32.0
Hermanson .274 3.95 305.2 .277 3.98 301.0
Lowe .254 2.70 146.2 .241 3.09 145.2
Martinez .196 2.16 270.1 .182 1.85 276.2
Oliver .290 5.48 213.2 .303 2.70 244.2
Urbina* .243 3.25 36.0 .217 4.11 30.2
Wakefield .275 4.58 232.0 .248 5.00 236.0
Wallace* .279 4.84 22.1 .192 2.25 28.0
</pre><hr></blockquote>

Before you complain "well Fenway is a hitters park" notice that the pitcher who've spent significant amounts of the time shown not on the Sox (Burkett, Castillo, Hermansn, Oliver, Urbina, Wallace) show no evidence of being better at home either.

So it doesn't look like your method holds up here, at least not for this admittedly small sample. You'll also notice some pitchers (Arrojo, Burkett, Castillo, Oliver) do indeed have a big enough split between times pitching at home and times pitching on the road to skew the statistical results.

In my opinion altering the pitchers ratings because of their status as the home or visitor is not accurate and should not be done.

BleacherBum · 04-03-2002, 03:27 PM

There's no question that in real life baseball, there is an advantage to playing at home. The question here is how do you quantify this advantage and incorporate it into a simulation model ?

A suggestion was made to just "bump up the ratings" for the home team players. Simple enough, but what are the reasons? Consider the factors that impact real life home and road performances, and then determine if they can actually be modeled in a game.

1) Home crowd - does the support of the crowd improve performance? or does it add pressure? and how is it different for the road team? are they adversely affected? or is performance also heightened in the face of adversity? Plus, over a 162 game season how significant is any effect on a day to day basis?

2) Comfort and Familiarity. Knowing your stadium must be nice, but what does it mean to the game engine? Do you make fewer errors because you know your field? Do you hit better because you are comfortable with the backdrop? Are you more focused because you can go through the same pre-game routine. Perhaps these types of things warrant a performance bonus, but is this equal for everyone?

These first 2 are psychological factors. If you believe they belong in the game, then what other psychological factors also merit inclusion? Are we really expecting a game that models human behavior?

Now, consider some more tangible factors:

3) Stadium. Consider factors such as playing surface, dimensions, foul territory, altitude, weather. These all have an impact on elements of the game. But none of them automatically give an advantage to the home team. The advantage comes in tailoring your team to benefit from your home park. So, the engine could certainly account for stadium factors, and if it does that's great. But to get a home benefit, you have to build a team with these factors in mind. (e.g. lefty power in Yankee stadium, pitching and speed on a large turf park, etc.) So realizing this advantage falls on the GM, not on the game engine. Do we as GM's, and do the OOTP computer GMs consider the home stadium at all when building a team? This is an important difference from real life.

4) Last at bats. This hasn't been mentioned yet, but this is an important strategical advantage. If you don't buy this, then I ask which side you choose in a pickup ballgame? But this doesn't warrant any ratings adjustments. The field manger must be skillful enough to benefit from last at-bats. So in OOTP, the AI of the field manager must be scrutinized. Perhaps in OOTP last at bat is not an advantage at all. Another potential difference from real life.

So, when looking at the real life stats that support a 55% home winning percentage, realize that 1) this has a psychological element which may not have a place in this game, and 2) requires some management intelligence that just might not be incorporated into the OOTP model.

Peter Bergstrom · 04-03-2002, 03:53 PM

For what it's worth, Diamond Mind Baseball models Home Field Advantage.

Face it, men, we need it!

Scott Vibert · 04-03-2002, 04:31 PM

[quote]Originally posted by Peter Bergstrom:
For what it's worth, Diamond Mind Baseball models Home Field Advantage.

Face it, men, we need it!<hr></blockquote>

For what? The game produces realistic enough stats, and win loss records? Why introduce an artificial, and potentially results skewing (for individual players) rating for the sake of two fairly insignificant win - loss splits? To me this really isn't necessary.

And even if it was necessary how do you measure and introduce this rating. (Adjusting the pitchers doesn't seem appropriate) Last year Boston, Toronto, Baltimore, Cleveland, Seattle, Anaheim, Atlanta, Houston, Milwaukee, Cincinnati, Los Angeles and San Diego all did better on the road, than they did at home. Does this mean these teams need negative adjustments to their "home field advantage" factor just to offset the gains they showed on the road.

Heck Atlanta made the playoffs without even winning half of their home games, so just how significant is the home field advantage in the major leagues, if you can make the playoffs without winning the majority of their own home games.

In the end all the stats balance out and are fairly good without this artificial factor.

Edit - This reminds me of what my mother used to say: If Diamond Mind Baseball jumped off a bridge would you?

[ 04-03-2002: Message edited by: ScottVib ]

Steve Kuffrey · 04-03-2002, 04:32 PM

[quote]Originally posted by Peter Bergstrom:
For what it's worth, Diamond Mind Baseball models Home Field Advantage.

Face it, men, we need it!<hr></blockquote>

Doesn't really prove anything.....

bigphesta · 04-03-2002, 05:04 PM

[quote]Originally posted by Peter Bergstrom:


Then, with all due respect, I think it's a ill-judged design choice. There is a built-in home field advantage in all major sports and baseball is no exception. This is reality.

Crowd supoprt? Stadium familiarity? More sex? I dunno, Scott. That's the way it is. And pro football and basketball are even more extreme in this regard. Not factoring in HF advantage in these two sports would be hallucinatory. Why do you think bookies give away points when teams are away? In football, HFA is worth a field goal plus a fraction.

<hr></blockquote>

Sorry, but you could not be more wrong when comparing MLB to the NFL. Actually, in the NFL the home field advantage has all but dissappeared in all but a few select places (ie Kansas City, Green Bay, Denver, San Francisco) Many teams, struggle to win at home yet dominate on the road. This is due to pressure by fans at home who boo and curse players when they do not perform. They don't have to deal with that on the road so they play more loosely. Look at this year's playoffs, New England goes to the Superbowl playing through Pittsburgh (also a historically "home" team) Chicago looses to Philly at home. Two seasons ago, Baltimore plays ALL road games and makes the Superbowl. So even in the NFL, most teams have just as much of a chance to win on the road as at home. And as for the bookies doing the spread concerning home field advantage is just plain wrong... at least for the NFL. And I feel as an expert on the subject considering there has only been ONE week in the last three years where I didn't have money on every team who played that week... so I follow this VERY closely.

04-03-2002, 09:38 AM	#1
Peter Bergstrom Minors (Rookie Ball) Join Date: Mar 2002 Posts: 27	Home Field Bug still Present in 4.0c Home teams are still losing >50% of all home with the new patch. Try simming a few seasons; it's all too evident. Statistically, the home team should prevail about 55% of the time. Not in OOTP4, however. Seems like a simple matter to fix, odd that it wasn't! <img src="confused.gif" border="0"> [ 04-03-2002: Message edited by: Peter Bergstrom ]</p>

04-03-2002, 09:57 AM	#2
Hammer755 Hall Of Famer Join Date: Dec 2001 Location: Houston, TX Posts: 2,348	[quote]Originally posted by Peter Bergstrom: <strong>Home teams are still losing >50% of all home with the new patch. Try simming a few seasons; it's all too evident. Statistically, the home team should prevail about 55% of the time. Not in OOTP4, however. Seems like a simple matter to fix, odd that it wasn't! [ 04-03-2002: Message edited by: Peter Bergstrom ]</strong><hr></blockquote> Peter, I've seen this vary largely from league to league. I just did some quick research on a couple of online leagues that I'm in. A modern-day, fictional league, after 3 complete seasons, yielded a home-field record of 1831-1865, which is a 0.495 W%. However, in another league, a dead-ball league with real players, home teams went 2701-2668 in 4+ seasons, which is a 0.503 W%. So, I imagine a league can have a pretty large swing from one end of the spectrum to another. It's tough to generalize by saying home teams always have a losing record without a ton of collected data, which is the case with pretty much any trend or stat included in OOTP.

04-03-2002, 10:39 AM	#3
Hammer755 Hall Of Famer Join Date: Dec 2001 Location: Houston, TX Posts: 2,348	[quote]Originally posted by blynch10: <strong> 10 teams with 55% or more winning % at home. </strong><hr></blockquote> This data set has a total Home W% of 0.515, but as I said before, one season cannot prove or disprove any theory.

04-03-2002, 10:45 AM	#4
blynch10 Minors (Triple A) Join Date: Jan 2002 Posts: 273	I was not trying to prove or disprove the theory on better winning % at home or not. Just showing that last year the huge swing between % for the teams. There are too many variables at work to say that teams will always win more at home. If true the Reds would have to do better then 26 and 53 at home. The numbers are from the real MLB 2001 season standings. [ 04-03-2002: Message edited by: blynch10 ] [ 04-03-2002: Message edited by: blynch10 ]</p>

04-03-2002, 11:58 AM	#7
Hammer755 Hall Of Famer Join Date: Dec 2001 Location: Houston, TX Posts: 2,348	[quote]Originally posted by Peter Bergstrom: <strong> Sorry, Hammer, that's way too low. Historically, the Home Field Advantage hovers around 55%. In fact last year it was 57 and 53 for the NL and AL respectively. It's a bug, albeit a minor one.</strong><hr></blockquote> [Whiny Al Gore voice]I have to disagree with your numbers.[/Whiny Al Gore voice] My brief research shows in 2001, the NL home W% was .517 and the AL W% was .528, which combines for a .522 W%. Those numbers were down across the board, as in 2000 it was NL-.543, AL-.537, MLB-.540. [quote]Originally posted by Peter Bergstrom:<strong>It's a bug, albeit a minor one</strong><hr></blockquote> I wouldn't necessarily say it's a bug, but a design choice. A bug is when an aspect of the game does not behave as it was designed to do. Now, you or I may disagree with the design decision, but that doesn't make it a bug.

04-03-2002, 11:20 AM	#5
Peter Bergstrom Minors (Rookie Ball) Join Date: Mar 2002 Posts: 27	[quote]Peter, I've seen this vary largely from league to league. I just did some quick research on a couple of online leagues that I'm in. A modern-day, fictional league, after 3 complete seasons, yielded a home-field record of 1831-1865, which is a 0.495 W%. However, in another league, a dead-ball league with real players, home teams went 2701-2668 in 4+ seasons, which is a 0.503 W%. So, I imagine a league can have a pretty large swing from one end of the spectrum to another. It's tough to generalize by saying home teams always have a losing record without a ton of collected data, which is the case with pretty much any trend or stat included in OOTP. <hr></blockquote> Sorry, Hammer, that's way too low. Historically, the Home Field Advantage hovers around 55%. In fact last year it was 57 and 53 for the NL and AL respectively. It's a bug, albeit a minor one.

04-03-2002, 11:44 AM	#6
Scott Vibert Hall Of Famer Join Date: Dec 2001 Location: My Computer Posts: 8,249	This isn't a bug, nor is it a design decision. The game doesn't give any artificial "home field advantage" to a team in this game. As a result the homefield winning percentage should tend to be around 50% or maybe a bit above or below in OOTP, each team that plays in this game will have as good a chance at beating another team at home as it does on the road. (I really wouldn't want to see some artificial homefield advantage built in). I don't see this as a problem, and its not even an annoyance to me.

04-03-2002, 12:01 PM	#8
Hammer755 Hall Of Famer Join Date: Dec 2001 Location: Houston, TX Posts: 2,348	[quote]Originally posted by ScottVib: <strong>This isn't a bug, nor is it a design decision. </strong><hr></blockquote> I agree Scott, I didn't mean that the game's W% are artificially tweaked one way or another. By design decision, I was refering to the decision not to include this facet.

04-03-2002, 12:35 PM	#9
Peter Bergstrom Minors (Rookie Ball) Join Date: Mar 2002 Posts: 27	[quote] wouldn't necessarily say it's a bug, but a design choice. A bug is when an aspect of the game does not behave as it was designed to do. Now, you or I may disagree with the design decision, but that doesn't make it a bug. <hr></blockquote> Then, with all due respect, I think it's a ill-judged design choice. There is a built-in home field advantage in all major sports and baseball is no exception. This is reality. Though not a game killer, to omit this factor is to render the game a bit less realistic. In my opinion. All you have to do is raise the pitcher's ratings a bit at home and depress them a bit on the road. This isn't rocket science.

04-03-2002, 12:43 PM	#10
Scott Vibert Hall Of Famer Join Date: Dec 2001 Location: My Computer Posts: 8,249	[quote]Originally posted by Peter Bergstrom: <strong> All you have to do is raise the pitcher's ratings a bit at home and depress them a bit on the road. This isn't rocket science.</strong><hr></blockquote> And what is the reasoning for the pitcher being worse on the road then they are at home? Is the pitcher not the same individual? So why should he pitch worse on the road? Think of the skewed results you'll get from pitchers who have the misfortune of starting frequently on the road. It's a good thing Nomo didn't react by your reasoning in his first start last year at Camden Yards, when he threw his no hitter.

04-03-2002, 12:51 PM	#11
Steve Kuffrey Administrator Join Date: Jul 1999 Location: S.E. TN - Georgia born and raised Posts: 17,036	I feel that making "across the board" ratings changes as you as saying should be done is not realistic. Some players play better on the road than at home..... To make this adjustment no matter how slight I feel is not a good design model. To allow the players and teams to perform as they will whether at home or on the road to me is a better model, and this is the design that we have chosen. Statistics may or may not bear out the home field advantage issue, but I would be curious how many sports simulation actually weigh this difference into their formulas? I know that Action PC Football give this as an "option". That may be something to consider in the future. But I feel to force the issue would not be right. __________________ Steve Kuffrey DABS Atlanta Braves - 2008 Eastern Division Champ *DBLC Atlanta Braves - 2011, 2014 East Division Champ, 2012, 2013 NL Wildcard Baseball Maelstrom-Montreal Expos-2013 Tourney winner, 2014 WC Team Sparky's League - Tampa Bay D'Rays Epicenter Baseball League - Astros 2014 The CBL Rewind - Phillies '95

04-03-2002, 01:00 PM	#12
IngredientX Minors (Double A) Join Date: Mar 2002 Location: New York Posts: 117	[quote]Originally posted by blynch10: <strong>Also can anyone tell me why what is typed in the message is not formated the same once displayed int the forum.....All my nice columns are squished. </strong><hr></blockquote> It's because the font used in the text box used to compose messages is Courier, which is a fixed-width font (all characters are the same width, which makes columns easy to create). But the actual message board uses a variable-width font. In this font, the letter m is wider than the letter i. m i In Courier, these letters are the same width as each other, and every other character. Hope this answers the easier question in your post. __________________ "Schopenhauer was right, wouldn't you say? 'Life without pain has no meaning.' Gentlemen, I wish to give your lives meaning."

04-03-2002, 01:02 PM	#13
Peter Bergstrom Minors (Rookie Ball) Join Date: Mar 2002 Posts: 27	[quote]And what is the reasoning for the pitcher being worse on the road then they are at home? Is the pitcher not the same individual? So why should he pitch worse on the road? Think of the skewed results you'll get from pitchers who have the misfortune of starting frequently on the road. <hr></blockquote> Crowd supoprt? Stadium familiarity? More sex? I dunno, Scott. That's the way it is. And pro football and basketball are even more extreme in this regard. Not factoring in HF advantage in these two sports would be hallucinatory. Why do you think bookies give away points when teams are away? In football, HFA is worth a field goal plus a fraction. And the results won't get skewed because, over time, pitchers will split their home and away innings evenly. Don't worry. Basically we're all here because we like sports and we like numbers. It's just a question of getting the nembers right. I just think it's remarkable that a young German guy could come up with such a convincing replication of our national game.

04-03-2002, 01:10 PM	#14
Peter Bergstrom Minors (Rookie Ball) Join Date: Mar 2002 Posts: 27	From a website: [quote]Abstract: The World Series of baseball in the United States is a best-of-seven series. Therefore, the first team to win four games is the victor. To try to make the series fair, the first two games are played at one team's home park, the next three games are played at the other team's park (just two may be played if one team wins four in a row), and the final two games (if needed) are scheduled at the first team's park. It is therefore possible that one team will play four games in its home park while the other team may only play three games at home. It is generally assumed that there is some advantage to playing at home. Each baseball park is different, so there may be some advantage to playing on a familiar field. The hometown fans provide emotional support for the team, and so on. Is there an advantage to playing World Series games on your home field? The dataset contains information on the year, the two teams competing in the world series, whether the American League (AL) or National League (NL) team won the game, and the AL and NL winning at home or away percentage. For all the games given, the teams won at home 55.5% of the time and away games were won 45% of the time. Using these probabilities and assuming that the chance of winning a home game is constant and that the games are independent, one can find the probabilities of winning in exactly 4, 5, 6 or 7 games as well as some other basic probability questions. The 2-way ANOVA by League and Home/Away shows significance for both factors. While the interaction term is not significant at the .05 level, the interaction plot suggests that the home field advantage may be greater for American league teams than for National League teams. <hr></blockquote>

04-03-2002, 02:23 PM	#15
Scott Vibert Hall Of Famer Join Date: Dec 2001 Location: My Computer Posts: 8,249	[quote]Originally posted by Peter Bergstrom: <strong> And the results won't get skewed because, over time, pitchers will split their home and away innings evenly. Don't worry. </strong><hr></blockquote> I still see no justification for increasing the home pitchers ratings vs. the road pitchers ratings, I just did a quick study from a tour of the stats from ESPN. I took one team (Boston) and used the three year averages (1999-2001) of each pitcher on the squad the results are presented here: * 3 year splits missing last years stats presented [code] Name Hm Av Hm ERA Hm IP Rd Av Rd ERA Rd IP Arrojo .295 5.48 226.2 .243 2.31 190.0 Banks* .000 0.00 4.0 .192 1.35 6.2 Burkett .297 5.30 231.0 .253 3.43 270.0 Castillo .257 4.61 125.0 .227 3.31 149.2 Crawford .288 6.21 29.0 .237 2.50 36.0 Fossum .211 4.01 24.2 .312 5.95 19.2 Garces* .221 3.86 35.0 .217 3.94 32.0 Hermanson .274 3.95 305.2 .277 3.98 301.0 Lowe .254 2.70 146.2 .241 3.09 145.2 Martinez .196 2.16 270.1 .182 1.85 276.2 Oliver .290 5.48 213.2 .303 2.70 244.2 Urbina* .243 3.25 36.0 .217 4.11 30.2 Wakefield .275 4.58 232.0 .248 5.00 236.0 Wallace* .279 4.84 22.1 .192 2.25 28.0 </pre><hr></blockquote> Before you complain "well Fenway is a hitters park" notice that the pitcher who've spent significant amounts of the time shown not on the Sox (Burkett, Castillo, Hermansn, Oliver, Urbina, Wallace) show no evidence of being better at home either. So it doesn't look like your method holds up here, at least not for this admittedly small sample. You'll also notice some pitchers (Arrojo, Burkett, Castillo, Oliver) do indeed have a big enough split between times pitching at home and times pitching on the road to skew the statistical results. In my opinion altering the pitchers ratings because of their status as the home or visitor is not accurate and should not be done.

04-03-2002, 03:27 PM	#16
BleacherBum All Star Reserve Join Date: Mar 2002 Posts: 672	There's no question that in real life baseball, there is an advantage to playing at home. The question here is how do you quantify this advantage and incorporate it into a simulation model ? A suggestion was made to just "bump up the ratings" for the home team players. Simple enough, but what are the reasons? Consider the factors that impact real life home and road performances, and then determine if they can actually be modeled in a game. 1) Home crowd - does the support of the crowd improve performance? or does it add pressure? and how is it different for the road team? are they adversely affected? or is performance also heightened in the face of adversity? Plus, over a 162 game season how significant is any effect on a day to day basis? 2) Comfort and Familiarity. Knowing your stadium must be nice, but what does it mean to the game engine? Do you make fewer errors because you know your field? Do you hit better because you are comfortable with the backdrop? Are you more focused because you can go through the same pre-game routine. Perhaps these types of things warrant a performance bonus, but is this equal for everyone? These first 2 are psychological factors. If you believe they belong in the game, then what other psychological factors also merit inclusion? Are we really expecting a game that models human behavior? Now, consider some more tangible factors: 3) Stadium. Consider factors such as playing surface, dimensions, foul territory, altitude, weather. These all have an impact on elements of the game. But none of them automatically give an advantage to the home team. The advantage comes in tailoring your team to benefit from your home park. So, the engine could certainly account for stadium factors, and if it does that's great. But to get a home benefit, you have to build a team with these factors in mind. (e.g. lefty power in Yankee stadium, pitching and speed on a large turf park, etc.) So realizing this advantage falls on the GM, not on the game engine. Do we as GM's, and do the OOTP computer GMs consider the home stadium at all when building a team? This is an important difference from real life. 4) Last at bats. This hasn't been mentioned yet, but this is an important strategical advantage. If you don't buy this, then I ask which side you choose in a pickup ballgame? But this doesn't warrant any ratings adjustments. The field manger must be skillful enough to benefit from last at-bats. So in OOTP, the AI of the field manager must be scrutinized. Perhaps in OOTP last at bat is not an advantage at all. Another potential difference from real life. So, when looking at the real life stats that support a 55% home winning percentage, realize that 1) this has a psychological element which may not have a place in this game, and 2) requires some management intelligence that just might not be incorporated into the OOTP model. __________________ Right Field Sucks!

04-03-2002, 03:53 PM	#17
Peter Bergstrom Minors (Rookie Ball) Join Date: Mar 2002 Posts: 27	For what it's worth, Diamond Mind Baseball models Home Field Advantage. Face it, men, we need it!

04-03-2002, 04:31 PM	#18
Scott Vibert Hall Of Famer Join Date: Dec 2001 Location: My Computer Posts: 8,249	[quote]Originally posted by Peter Bergstrom: <strong>For what it's worth, Diamond Mind Baseball models Home Field Advantage. Face it, men, we need it!</strong><hr></blockquote> For what? The game produces realistic enough stats, and win loss records? Why introduce an artificial, and potentially results skewing (for individual players) rating for the sake of two fairly insignificant win - loss splits? To me this really isn't necessary. And even if it was necessary how do you measure and introduce this rating. (Adjusting the pitchers doesn't seem appropriate) Last year Boston, Toronto, Baltimore, Cleveland, Seattle, Anaheim, Atlanta, Houston, Milwaukee, Cincinnati, Los Angeles and San Diego all did better on the road, than they did at home. Does this mean these teams need negative adjustments to their "home field advantage" factor just to offset the gains they showed on the road. Heck Atlanta made the playoffs without even winning half of their home games, so just how significant is the home field advantage in the major leagues, if you can make the playoffs without winning the majority of their own home games. In the end all the stats balance out and are fairly good without this artificial factor. Edit - This reminds me of what my mother used to say: If Diamond Mind Baseball jumped off a bridge would you? [ 04-03-2002: Message edited by: ScottVib ]</p>

04-03-2002, 04:32 PM	#19
Steve Kuffrey Administrator Join Date: Jul 1999 Location: S.E. TN - Georgia born and raised Posts: 17,036	[quote]Originally posted by Peter Bergstrom: <strong>For what it's worth, Diamond Mind Baseball models Home Field Advantage. Face it, men, we need it!</strong><hr></blockquote> Doesn't really prove anything..... __________________ Steve Kuffrey DABS Atlanta Braves - 2008 Eastern Division Champ *DBLC Atlanta Braves - 2011, 2014 East Division Champ, 2012, 2013 NL Wildcard Baseball Maelstrom-Montreal Expos-2013 Tourney winner, 2014 WC Team Sparky's League - Tampa Bay D'Rays Epicenter Baseball League - Astros 2014 The CBL Rewind - Phillies '95

04-03-2002, 05:04 PM	#20
bigphesta Banned Join Date: Dec 2001 Location: Des Moines, Iowa Posts: 1,366	[quote]Originally posted by Peter Bergstrom: <strong> Then, with all due respect, I think it's a ill-judged design choice. There is a built-in home field advantage in all major sports and baseball is no exception. This is reality. Crowd supoprt? Stadium familiarity? More sex? I dunno, Scott. That's the way it is. And pro football and basketball are even more extreme in this regard. Not factoring in HF advantage in these two sports would be hallucinatory. Why do you think bookies give away points when teams are away? In football, HFA is worth a field goal plus a fraction. </strong><hr></blockquote> Sorry, but you could not be more wrong when comparing MLB to the NFL. Actually, in the NFL the home field advantage has all but dissappeared in all but a few select places (ie Kansas City, Green Bay, Denver, San Francisco) Many teams, struggle to win at home yet dominate on the road. This is due to pressure by fans at home who boo and curse players when they do not perform. They don't have to deal with that on the road so they play more loosely. Look at this year's playoffs, New England goes to the Superbowl playing through Pittsburgh (also a historically "home" team) Chicago looses to Philly at home. Two seasons ago, Baltimore plays ALL road games and makes the Superbowl. So even in the NFL, most teams have just as much of a chance to win on the road as at home. And as for the bookies doing the spread concerning home field advantage is just plain wrong... at least for the NFL. And I feel as an expert on the subject considering there has only been ONE week in the last three years where I didn't have money on every team who played that week... so I follow this VERY closely.