VORP issue with extreme ballpark factors - Page 2

Khrog · 01-15-2007, 05:05 AM

Looks like what the actual problem is that "critical hit"/homerun effects are somewhere in the hitting calculations. Those calculations happen a few % of the time and are increased with pitcher fatigue, etc. and the ballpark's extreme difficulty for homeruns doesn't get factored into those kinds of calculations.

The other "problem" is that with such an extreme rating that its almost impossible not to exceed expected homerun totals affecting VORP.

I'd be interested in seeing what a .100 would produce(now each HR=~10HR allowed/hit) May even get the same overall # of HRs or statistically similar.

tysok · 01-15-2007, 10:30 AM

Park factors change from year to year, occasionally by big margins. That's why formulas that normalize stats to the park use a 5 year floating average, or some other set in order to average the effects.

The game should be creating the number to use in this calculation at the end of the year instead of using what's entered as the generic factor... then this would go away.

You've entered such a low factor that you'd be expected to see almost no homeruns hit in this ballpark... that's not gonna happen. The factors are, to some extent, controlled by who's playing there as well as the amount of foul ground, depth of fences, and weather effects.... Coors field was supposed to be very homer friendly, but if the entire team had the power of Juan Pierre it wouldn't be so much... Oakland is supposedly pitcher friendly but if you took a team full of the biggest sluggers in the game in there for 81 games it wouldn't be so much.

The factor you enter as "park factors" should be a basis, what the game uses in order to play the game there. The factors used in calculations should be what really did happen there. I can see the point, with all the players and all the fields there could be in the game, it'd be a nightmare to figure park factors then figure VORP after every game... but it should be figuring the park factors (to be used in these calculations) at the end of each year (floating average if possible) instead of using those general numbers that may or may not mean anything right now.

ctorg · 01-15-2007, 02:23 PM

Quote:

Originally Posted by tysok

Park factors change from year to year, occasionally by big margins. That's why formulas that normalize stats to the park use a 5 year floating average, or some other set in order to average the effects.

The game should be creating the number to use in this calculation at the end of the year instead of using what's entered as the generic factor... then this would go away.

You've entered such a low factor that you'd be expected to see almost no homeruns hit in this ballpark... that's not gonna happen. The factors are, to some extent, controlled by who's playing there as well as the amount of foul ground, depth of fences, and weather effects.... Coors field was supposed to be very homer friendly, but if the entire team had the power of Juan Pierre it wouldn't be so much... Oakland is supposedly pitcher friendly but if you took a team full of the biggest sluggers in the game in there for 81 games it wouldn't be so much.

The factor you enter as "park factors" should be a basis, what the game uses in order to play the game there. The factors used in calculations should be what really did happen there. I can see the point, with all the players and all the fields there could be in the game, it'd be a nightmare to figure park factors then figure VORP after every game... but it should be figuring the park factors (to be used in these calculations) at the end of each year (floating average if possible) instead of using those general numbers that may or may not mean anything right now.

Yeah, this is a really extreme situation. Really, I don't think it's quite possible to have such a low ballpark factor for homers. Maybe if you built a stadium with 90 foot high walls that are 340-380 feet out (all the homers would be in the park, but there wouldn't be many because the walls aren't that far out).

Anyway, that's not going to happen in reality. My concern is whether this is happening to lesser degrees with not-so-extreme parks. I do recall that when using OOTP6 and catobase, where you could look at park factors each year, those factors were often very far off from the assigned ones (to the point that there was very low correlation between the assigned values and the actual park behavior). But that was OOTP6, which had a different engine.

The value being calculated at the end of the year based on the previous year is probably the best bet. Granted, teams with new stadiums won't be factored correctly, but it would be better than what currently exists.

tysok · 01-15-2007, 03:06 PM

Quote:

Originally Posted by ctorg

Yeah, this is a really extreme situation. Really, I don't think it's quite possible to have such a low ballpark factor for homers. Maybe if you built a stadium with 90 foot high walls that are 340-380 feet out (all the homers would be in the park, but there wouldn't be many because the walls aren't that far out).

Anyway, that's not going to happen in reality. My concern is whether this is happening to lesser degrees with not-so-extreme parks. I do recall that when using OOTP6 and catobase, where you could look at park factors each year, those factors were often very far off from the assigned ones (to the point that there was very low correlation between the assigned values and the actual park behavior). But that was OOTP6, which had a different engine.

The value being calculated at the end of the year based on the previous year is probably the best bet. Granted, teams with new stadiums won't be factored correctly, but it would be better than what currently exists.

I did the Win Shares program that would take the data dumps and figure the Win Shares, HOF tests, and the park factors... the park factors weren't bad (the worst one looks like Boston was supposed to be a HR factor of 1.065 but was really 1.541)... but of course they were never exactly what the assigned values were. It makes more sense to use the real numbers, what the park factors really were for the last few seasons, than to use the assigned value which is never correct.

ForemanFan · 01-16-2007, 09:50 AM

Quote:

Originally Posted by ctorg

Yeah, this is a really extreme situation. Really, I don't think it's quite possible to have such a low ballpark factor for homers. Maybe if you built a stadium with 90 foot high walls that are 340-380 feet out (all the homers would be in the park, but there wouldn't be many because the walls aren't that far out).

Anyway, that's not going to happen in reality. My concern is whether this is happening to lesser degrees with not-so-extreme parks. I do recall that when using OOTP6 and catobase, where you could look at park factors each year, those factors were often very far off from the assigned ones (to the point that there was very low correlation between the assigned values and the actual park behavior). But that was OOTP6, which had a different engine.

The value being calculated at the end of the year based on the previous year is probably the best bet. Granted, teams with new stadiums won't be factored correctly, but it would be better than what currently exists.

Out of curiosity, do you manage all of your games or just sim them? There's always a chance that many of those homers are indeed in-the-parkers. The park home run factor may not have as much of an impact on that type of four-baggers. Then again, looking at the stolen base total of your hitters, it seems as if you have a team full of un-Mercurial Boog Powells.

EDIT: Well, 'cept the fleet-footed Barry MacGill. Thirteen SB's to lead the team? No small ball here.

ctorg · 01-16-2007, 10:21 AM

Quote:

Originally Posted by ForemanFan

Out of curiosity, do you manage all of your games or just sim them? There's always a chance that many of those homers are indeed in-the-parkers. The park home run factor may not have as much of an impact on that type of four-baggers. Then again, looking at the stolen base total of your hitters, it seems as if you have a team full of un-Mercurial Boog Powells.

EDIT: Well, 'cept the fleet-footed Barry MacGill. Thirteen SB's to lead the team? No small ball here.

I'm doing a very quick sim to get a league history, so I'm not managing anything. In my experience (this is not the first time I've used these park factors, just the first time I've noticed the VORP issue), though, inside-the-park homers don't happen any more or less often - at least not to this degree or any degree I've noticed - in different parks. Perhaps it would be a function of the ballpark's triples factor, but I don't know. The game doesn't really "know" the size of the ballpark, just the ballpark factors, as far as I know, so it wouldn't have anything to use as a basis for there being more inside-the-parkers.

ForemanFan · 01-17-2007, 10:34 PM

Quote:

Originally Posted by ctorg

I'm doing a very quick sim to get a league history, so I'm not managing anything. In my experience (this is not the first time I've used these park factors, just the first time I've noticed the VORP issue), though, inside-the-park homers don't happen any more or less often - at least not to this degree or any degree I've noticed - in different parks. Perhaps it would be a function of the ballpark's triples factor, but I don't know. The game doesn't really "know" the size of the ballpark, just the ballpark factors, as far as I know, so it wouldn't have anything to use as a basis for there being more inside-the-parkers.

Ah, that makes perfect sense. It would be nice, however, if in a future version this could be changed to acknowledge actual dimensions rather than those durned Park Factors. Yeah, yeah, I know that this requires about two decades of coding and that I'm asking a lot, but it's just a dream that I have. . . . from time to time. . . .

Curtis · 01-17-2007, 11:04 PM

Have you tried making copies of your league and running one season each with every park at .01, .02, .04, .08, .16, .32 and .64? You could find out how much of a difference each increment makes. (Alternatively, use 1.00, .50, .12, .06, .03 and .015.)

ctorg · 01-18-2007, 07:47 AM

Quote:

Originally Posted by ForemanFan

Ah, that makes perfect sense. It would be nice, however, if in a future version this could be changed to acknowledge actual dimensions rather than those durned Park Factors. Yeah, yeah, I know that this requires about two decades of coding and that I'm asking a lot, but it's just a dream that I have. . . . from time to time. . . .

An idea I threw up a few years back that disappeared into the abyss was having a button that creates ballpark factors from dimensions you entered (also, having a button that automatically adjusts factors so that they average out to 1.0 would be nice, but more complicated).

ctorg · 01-18-2007, 07:48 AM

Quote:

Originally Posted by Curtis

Have you tried making copies of your league and running one season each with every park at .01, .02, .04, .08, .16, .32 and .64? You could find out how much of a difference each increment makes. (Alternatively, use 1.00, .50, .12, .06, .03 amd .015.)

I haven't really tried anything yet. I only get a few hour of OOTP time a week, so I don't know when I'll get to test things.

01-15-2007, 10:30 AM	#22
tysok Hall Of Famer Join Date: Aug 2003 Posts: 4,925	Park factors change from year to year, occasionally by big margins. That's why formulas that normalize stats to the park use a 5 year floating average, or some other set in order to average the effects. The game should be creating the number to use in this calculation at the end of the year instead of using what's entered as the generic factor... then this would go away. You've entered such a low factor that you'd be expected to see almost no homeruns hit in this ballpark... that's not gonna happen. The factors are, to some extent, controlled by who's playing there as well as the amount of foul ground, depth of fences, and weather effects.... Coors field was supposed to be very homer friendly, but if the entire team had the power of Juan Pierre it wouldn't be so much... Oakland is supposedly pitcher friendly but if you took a team full of the biggest sluggers in the game in there for 81 games it wouldn't be so much. The factor you enter as "park factors" should be a basis, what the game uses in order to play the game there. The factors used in calculations should be what really did happen there. I can see the point, with all the players and all the fields there could be in the game, it'd be a nightmare to figure park factors then figure VORP after every game... but it should be figuring the park factors (to be used in these calculations) at the end of each year (floating average if possible) instead of using those general numbers that may or may not mean anything right now. __________________ I don't know about you, but as for me, the question has already been answered: Should we be here? Yes! Jack Buck, September 17, 2001 It's what you learn after you know it all that counts. I firmly believe that any man's finest hour... is that moment when he has worked his heart out in a good cause and lies exhausted on the field of battle - victorious. (Vince Lombardi) I don't measure a man's success by how high he climbs but how high he bounces when he hits bottom. (George S. Patton)

01-17-2007, 11:04 PM	#28
Curtis Hall Of Famer Join Date: Jul 2006 Location: Watertown, New York Posts: 4,567	Have you tried making copies of your league and running one season each with every park at .01, .02, .04, .08, .16, .32 and .64? You could find out how much of a difference each increment makes. (Alternatively, use 1.00, .50, .12, .06, .03 and .015.) Last edited by Curtis; 01-18-2007 at 11:00 AM. Reason: can't spell 'and'

01-15-2007, 05:05 AM	#21
Khrog Minors (Double A) Join Date: Aug 2006 Posts: 131	Looks like what the actual problem is that "critical hit"/homerun effects are somewhere in the hitting calculations. Those calculations happen a few % of the time and are increased with pitcher fatigue, etc. and the ballpark's extreme difficulty for homeruns doesn't get factored into those kinds of calculations. The other "problem" is that with such an extreme rating that its almost impossible not to exceed expected homerun totals affecting VORP. I'd be interested in seeing what a .100 would produce(now each HR=~10HR allowed/hit) May even get the same overall # of HRs or statistically similar.