Annual post on inside the park homers

[CBL-Commish]

I'm guessing I already know the answer to this, but I ask it at least once with every new release of OOTP, since V2:

Any chance inside the park homers will work correctly? Since the park design and effect appear (from comments in this thread) to be largely unchanged, I'd guess not, but I'll hold out hope.

If you don't know what I'm talking about, I mean the problem that inside the park homers are hard coded to some tiny percentage of overall homers no matter what the ballpark looks like or what the park effects are. If you have a park that's 500 ft all around and has a triple factor of 198, you'll get the same number of inside-the-parkers as in a stadium that's 325 all around with a triple factor of 0.1. And that number will always be a handful per league per season.

Also, there's no consideration of the type of hitter/runner. In real life, with a huge park, fast line drive hitters would hit many more homers than Ryan Howard types. But not in OOTP - a 500 ft fence won't mean Carl Crawford hits a lot of inside the parkers, it'll mean Ryan Howard hits a lot of 550-ft home runs. And it'll mean that Carl Crawford will hit a lot of doubles and triples in real-life situations where he could round the bases before a fielder even picked up the ball.

This isn't a big deal for typical MLB stadiums in modern baseball. But if your stadiums start getting a few very big dimensions and/or you're playing deadball era baseball inside-the-park homers will deviate significantly from reality.

[ctorg]

I have bumped into this problem as well. I had a stadium in one league with extreme dimensions (one I've referred to elsewhere in posts about VORP calculation). It's something like 450 feet at the lines and just goes out from there, with 40 foot walls all around. Basically, it's pretty much impossible to hit homeruns there in real life. However, I didn't set the ballpark factor all the way down to zero or anything because I wanted to try to account for inside-the-park homers. Unfortunately, as you said, it resulted in a lot of 500+ foot homers over the walls. Perhaps an "ITPHR" factor could be added after doubles and triples factors.

[swampdragon]

Originally Posted by ctorg

I have bumped into this problem as well. I had a stadium in one league with extreme dimensions (one I've referred to elsewhere in posts about VORP calculation). It's something like 450 feet at the lines and just goes out from there, with 40 foot walls all around. Basically, it's pretty much impossible to hit homeruns there in real life. However, I didn't set the ballpark factor all the way down to zero or anything because I wanted to try to account for inside-the-park homers. Unfortunately, as you said, it resulted in a lot of 500+ foot homers over the walls. Perhaps an "ITPHR" factor could be added after doubles and triples factors.

But unless you just wanted to base it on triples, you'd have to do an enormous amount of research to get the factor right for historical leagues. Even after the research, there's still a sample size problem except for a small number of parks during the deadball era.

[ctorg]

Originally Posted by swampdragon

But unless you just wanted to base it on triples, you'd have to do an enormous amount of research to get the factor right for historical leagues. Even after the research, there's still a sample size problem except for a small number of parks during the deadball era.

So for historical leagues, you could just leave them all at 1.000. This would really be something useful for fictional leagues.

[akw4572]

Any historical information that says more inside the park homers were hit in the polo grounds, than ebbetts, or forbes??? I don't know the answer, and I'm not sure the info is available and/or correct.

[Mr. Capo]

Originally Posted by swampdragon

But unless you just wanted to base it on triples, ...

That would be better than nothing. It seems like a formula that relied heavily on triples would end up pretty realistic.

[swampdragon]

Originally Posted by Mr. Capo

That would be better than nothing. It seems like a formula that relied heavily on triples would end up pretty realistic.

But you'd have to be careful. I don't want Babe Ruth hitting a bunch of inside the park homers. You would have to base a batter's hitting them on triples as well, so Sam Crawford would get more than Gavvy Cravath. You're better off leaving it as is than making the wrong changes.

[CBL-Commish]

Originally Posted by akw4572

Any historical information that says more inside the park homers were hit in the polo grounds, than ebbetts, or forbes??? I don't know the answer, and I'm not sure the info is available and/or correct.

The Tattersoll-McConnell Home Run Log is now available to SABR members on their web site. I don't have my login info here at work, but when I get home I'll check it out. It may have inside-the-park homer data, which could be used for historical park effects and ITPH frequency over time.

This link has a list of leaders in inside the park homers. Just glancing at Cobb, Tommy Leach, and Sam Crawford, it looks like many players in the deadball era hit half or more of their homers inside the park. So OOTP gets something like 30%-50% of their homers wrong by assuming they left the field.

[CBL-Commish]

Originally Posted by swampdragon

But you'd have to be careful. I don't want Babe Ruth hitting a bunch of inside the park homers. You would have to base a batter's hitting them on triples as well, so Sam Crawford would get more than Gavvy Cravath. You're better off leaving it as is than making the wrong changes.

Dola, Ruth did hit 10 inside the parkers, at least according to this source.

But yea, the game should definitely use player speed as one of the determining factors in the calculation. Although that it should probably be a sliding scale where speed is less important for really long fence distances. It probably doesn't matter if Billy Sunday or Harold Baines is running if the ball rolls to the base of the 610 sign.

[Simmo13]

Perhaps we should also have deadball PBP which allows for "bounce" homers which counted until the twenties in both leagues I think.

[Curtis]

Originally Posted by CBL-Commish

The Tattersoll-McConnell Home Run Log is now available to SABR members on their web site. I don't have my login info here at work, but when I get home I'll check it out. It may have inside-the-park homer data, which could be used for historical park effects and ITPH frequency over time.

This link has a list of leaders in inside the park homers. Just glancing at Cobb, Tommy Leach, and Sam Crawford, it looks like many players in the deadball era hit half or more of their homers inside the park. So OOTP gets something like 30%-50% of their homers wrong by assuming they left the field.

Originally Posted by Simmo13

Perhaps we should also have deadball PBP which allows for "bounce" homers which counted until the twenties in both leagues I think.

My understanding was that during the deadball era, which in this case includes 1871-1920, about 45% of all home runs never left the park. This includes an era in which many (or most) parks had either no fences or incomplete ones.

Simmo's post reminds me that the definition of what constitutes a homer — as well as what constitutes 'inside the park' — can change with time and further complicate the calculation. I'm led to believe, though, that the number of ITP homers should rise as the number hit out falls, thereby acting as a mitigating factor. ctorg, I'm wondering if this could be happening in your extreme park, but the play-by-play isn't reflecting it due to its own hardcoding?

[legendsport]

Originally Posted by CBL-Commish

The Tattersoll-McConnell Home Run Log is now available to SABR members on their web site. I don't have my login info here at work, but when I get home I'll check it out. It may have inside-the-park homer data, which could be used for historical park effects and ITPH frequency over time.

This link has a list of leaders in inside the park homers. Just glancing at Cobb, Tommy Leach, and Sam Crawford, it looks like many players in the deadball era hit half or more of their homers inside the park. So OOTP gets something like 30%-50% of their homers wrong by assuming they left the field.

I'm a SABR member - and the HR log does indicate if a HR was ITP. Here's Ty Cobb's 1909 HR log:

Date		HR	Pitcher		Description
06/20/1909	1	Rube Waddell	Top 5; 2 outs; 1 Runners On; Inside-the-Park HR; Day Game; Sportsman's Park III
07/15/1909	2	Charlie Smith	Btm 1; 2 Runners On; Inside-the-Park HR; Day Game; Bennett Park
 		3	Charlie Smith	Btm 5; 1 Runners On; Inside-the-Park HR; Day Game; Bennett Park
07/19/1909	4	Pete Wilson	Btm 5; 1 Runners On; Inside-the-Park HR; Day Game; Bennett Park
07/29/1909	5	Slow Joe Doyle	Top 7; Bases Empty; Inside-the-Park HR; Day Game; Hilltop Park
08/06/1909	6	Harry Krause	Top 2; Bases Empty; Inside-the-Park HR; Day Game; Shibe Park
08/27/1909	7	Jack Chesbro	Btm 4; 2 Runners On; Inside-the-Park HR; Back-to-back HR (1); Day Game; Bennett Park
09/01/1909	8	Eddie Cicotte	Btm 1; 1 Runners On; Inside-the-Park HR; Day Game; Bennett Park
09/13/1909	9	Chuck Rose	Btm 8; 2 Runners On; Inside-the-Park HR; Day Game; Bennett Park

So, ITP factors could be calculated. If someone in SABR asked David Vincent (keeper of the log) nicely, he might be able to provide the raw numbers. I'll try posting on the SABR-L and see if he responds.

[CBL-Commish]

Originally Posted by legendsport

I'm a SABR member - and the HR log does indicate if a HR was ITP. Here's Ty Cobb's 1909 HR log:

Date		HR	Pitcher		Description
06/20/1909	1	Rube Waddell	Top 5; 2 outs; 1 Runners On; Inside-the-Park HR; Day Game; Sportsman's Park III
07/15/1909	2	Charlie Smith	Btm 1; 2 Runners On; Inside-the-Park HR; Day Game; Bennett Park
 		3	Charlie Smith	Btm 5; 1 Runners On; Inside-the-Park HR; Day Game; Bennett Park
07/19/1909	4	Pete Wilson	Btm 5; 1 Runners On; Inside-the-Park HR; Day Game; Bennett Park
07/29/1909	5	Slow Joe Doyle	Top 7; Bases Empty; Inside-the-Park HR; Day Game; Hilltop Park
08/06/1909	6	Harry Krause	Top 2; Bases Empty; Inside-the-Park HR; Day Game; Shibe Park
08/27/1909	7	Jack Chesbro	Btm 4; 2 Runners On; Inside-the-Park HR; Back-to-back HR (1); Day Game; Bennett Park
09/01/1909	8	Eddie Cicotte	Btm 1; 1 Runners On; Inside-the-Park HR; Day Game; Bennett Park
09/13/1909	9	Chuck Rose	Btm 8; 2 Runners On; Inside-the-Park HR; Day Game; Bennett Park

So, ITP factors could be calculated. If someone in SABR asked David Vincent (keeper of the log) nicely, he might be able to provide the raw numbers. I'll try posting on the SABR-L and see if he responds.

That's great, exactly what I was going to look up. Please let us know if Mr. Vincent gets back to you. It would be wonderful to have real data for Markus to use - at least for a patch or a future version.

[ctorg]

Originally Posted by Curtis

My understanding was that during the deadball era, which in this case includes 1871-1920, about 45% of all home runs never left the park. This includes an era in which many (or most) parks had either no fences or incomplete ones.

Simmo's post reminds me that the definition of what constitutes a homer — as well as what constitutes 'inside the park' — can change with time and further complicate the calculation. I'm led to believe, though, that the number of ITP homers should rise as the number hit out falls, thereby acting as a mitigating factor. ctorg, I'm wondering if this could be happening in your extreme park, but the play-by-play isn't reflecting it due to its own hardcoding?

Well, if it is, then the way VORP is calculated needs to change to reflect that. Isn't a park's homerun factor meant to include inside-of-the-park homeruns when it is calculated historically?

[legendsport]

Originally Posted by CBL-Commish

That's great, exactly what I was going to look up. Please let us know if Mr. Vincent gets back to you. It would be wonderful to have real data for Markus to use - at least for a patch or a future version.

Will do.

Originally Posted by ctorg

Well, if it is, then the way VORP is calculated needs to change to reflect that. Isn't a park's homerun factor meant to include inside-of-the-park homeruns when it is calculated historically?

I'd bet that the historical park factors in OOTP do not differentiate between the two types of homers. Any published factors you'd find in say, Total Baseball, are for all homers in a park, whether they're the land-in-the-bleachers variety or inside-the-park.

The data does exist to "split" the factor and give a true indication of how parks played, but since this is really only a big deal in the deadball era, I wonder if Markus would even be interested in putting something like that in the game.

[CBL-Commish]

Originally Posted by legendsport

The data does exist to "split" the factor and give a true indication of how parks played, but since this is really only a big deal in the deadball era, I wonder if Markus would even be interested in putting something like that in the game.

It's also a pretty big deal in the Continental League, where we have at least a half dozen parks with one or more fences at least 460 feet away.

In other words it might be pretty useful to fictional players, too.

[legendsport]

Originally Posted by CBL-Commish

It's also a pretty big deal in the Continental League, where we have at least a half dozen parks with one or more fences at least 460 feet away.

In other words it might be pretty useful to fictional players, too.

Good point. I agree that something like this would be very cool to have. From the player side, getting an ITP HR would have to depend on gap power and speed more so than strictly HR power - since sluggers (generally) tend to not be the fleetest of fellows.

Hopefully it can be squeezed in as a patch or in a future version, as you said.