Coors effect. - Fantasy Baseball Cafe 2015

## Coors effect.

Moderator: Baseball Moderators

### Coors effect.

Help me out here, fellas. I was reading something on the Coors field effect on hitting and although it makes perfect sense to me on paper, it makes no sense to me in reality. To sum up the calculations, there is a 20% bump a Rockie hitter gets playing at Coors, but on the road there's a -19% "hangover effect" for those same hitters. So, in short, there's only a 1% bump a player will get by playing for the Rockies. Am I correct in my calculations when I say a .280, 20 homer, and 90 RBI player in a normal hitting park will become a a whopping .283, 20, and 91 after being traded to the Rockies? I don't know about you fellas, but this seems like a bunch of matchematical mumbo-jumbo to me. What say you baseball gurus out there?
Erboes
Major League Manager

Posts: 1433
Joined: 26 Mar 2003
Home Cafe: Baseball

### Re: Coors effect.

Erboes wrote:Help me out here, fellas. I was reading something on the Coors field effect on hitting and although it makes perfect sense to me on paper, it makes no sense to me in reality. To sum up the calculations, there is a 20% bump a Rockie hitter gets playing at Coors, but on the road there's a -19% "hangover effect" for those same hitters. So, in short, there's only a 1% bump a player will get by playing for the Rockies. Am I correct in my calculations when I say a .280, 20 homer, and 90 RBI player in a normal hitting park will become a a whopping .283, 20, and 91 after being traded to the Rockies? I don't know about you fellas, but this seems like a bunch of matchematical mumbo-jumbo to me. What say you baseball gurus out there?

I'm not a mathematical or statistical whiz by any means but I would think the names Castilla, Bichette, Galaragga, Cirillo, Payton, Pr. Wilson, etc., etc. would fly in the face of the argument that Coors doesn't effect hitter's #'s dramatically.
nsulham
General Manager

Posts: 4883
Joined: 26 Dec 2002
Home Cafe: Baseball
Location: Charlottesville

Here's the full piece I was reading:

Here's the full-blown calculation:

In order to determine how much Rockies hitters' numbers are boosted by Coors, we need to know three things:

A) The conventional Coors park factor.

B) The extent to which the hangover effect hurts our hitters' overall numbers.

C) The extent to which the hangover effect affects the conventional Coors park factor.

The overall hangover-adjusted Coors park factor, then, would be A-B-C.

We know that A is approximately +20%, as Lou mentioned.

What about B? Well, according to the link that Lou gave, the hangover effect reduces hitters' road OBP and SLG from .339/.411 to .302/.363. Using OBP*SLG, we then find that that is equal to a 27% reduction in runs created in road games, which translates to a 13% reduction overall (I rounded down because most of our hitters' PA's come at home).

Now, onto the tricky one. How does the hangover effect affect the conventional park factor? Let's put it in equation form:

RSH: Runs scored at home
RAH: Runs allowed at home
RSR: Runs scored on the road
RAR: Runs allowed on the road

The conventional park factor is calculated as ((RSH+RAH)/(RSR+RAR)+1)/2. Set all four variables equal to one another to calculate the park factor of a neutral park; obviously, this is equal to 1.

Now, to account for the hangover effect. Let's first set all four variables equal to 4.61, the average number of runs scored per game in the 2003 NL. Now, the hangover effect reduces RSR by 27%, so we replace RSR with 4.61/1.27, or 3.63. Now we calculate the park factor, and come up with 1.06. That means that the hangover effect, by itself, inflates the conventional park factor by 6%. This is our value for C (in my first equation).

So we have our answer. The park factor equals A-B-C, where A=+20%, B=-13%, and C=-6%. Therefore, Coors inflates Rockies hitters' numbers by 1 percent.

Obviously, that's not an exact calculation; we don't know enough about the hangover effect (or about the Coors inflation effect) to come up with a precise answer. But it's pretty close, and it certainly goes to show that any system that does not account for the hangover effect is completely worthless when it comes to evaluating Rockies hitters.

End quote

I think it has to be a bunch of garbage too, but maybe some of the statistical guys here can shed some light on this.
Erboes
Major League Manager

Posts: 1433
Joined: 26 Mar 2003
Home Cafe: Baseball

Erboes wrote:Here's the full piece I was reading:

Here's the full-blown calculation:

In order to determine how much Rockies hitters' numbers are boosted by Coors, we need to know three things:

A) The conventional Coors park factor.

B) The extent to which the hangover effect hurts our hitters' overall numbers.

C) The extent to which the hangover effect affects the conventional Coors park factor.

The overall hangover-adjusted Coors park factor, then, would be A-B-C.

We know that A is approximately +20%, as Lou mentioned.

What about B? Well, according to the link that Lou gave, the hangover effect reduces hitters' road OBP and SLG from .339/.411 to .302/.363. Using OBP*SLG, we then find that that is equal to a 27% reduction in runs created in road games, which translates to a 13% reduction overall (I rounded down because most of our hitters' PA's come at home).

Now, onto the tricky one. How does the hangover effect affect the conventional park factor? Let's put it in equation form:

RSH: Runs scored at home
RAH: Runs allowed at home
RSR: Runs scored on the road
RAR: Runs allowed on the road

The conventional park factor is calculated as ((RSH+RAH)/(RSR+RAR)+1)/2. Set all four variables equal to one another to calculate the park factor of a neutral park; obviously, this is equal to 1.

Now, to account for the hangover effect. Let's first set all four variables equal to 4.61, the average number of runs scored per game in the 2003 NL. Now, the hangover effect reduces RSR by 27%, so we replace RSR with 4.61/1.27, or 3.63. Now we calculate the park factor, and come up with 1.06. That means that the hangover effect, by itself, inflates the conventional park factor by 6%. This is our value for C (in my first equation).

So we have our answer. The park factor equals A-B-C, where A=+20%, B=-13%, and C=-6%. Therefore, Coors inflates Rockies hitters' numbers by 1 percent.

Obviously, that's not an exact calculation; we don't know enough about the hangover effect (or about the Coors inflation effect) to come up with a precise answer. But it's pretty close, and it certainly goes to show that any system that does not account for the hangover effect is completely worthless when it comes to evaluating Rockies hitters.

End quote

I think it has to be a bunch of garbage too, but maybe some of the statistical guys here can shed some light on this.

Ugh. I don't have the patience this morning to try and read through all that.

I'm not sure what they mean by hangover effect. (Is there a link to the article missing in the above post?) I do remember reading last year that when Rockies hitters start a road trip it takes them a few games to adjust to seeing curves break again. But after a few games they start hitting curves again.

Is the claim that Rockies hitters actually hit worse in other ballparks than every other hitter? So much worse that over a season we shouldn't expect any bump in stats from hitting at Coors? That seems ... odd.
"The game has a cleanness. If you do a good job, the numbers say so. You don't have to ask anyone or play politics. You don't have to wait for the reviews." - Sandy Koufax
ramble2
General Manager

Posts: 2952
(Past Year: 5)
Joined: 27 Feb 2003
Home Cafe: Baseball
Location: Salt Lake City

Ramble2 wrote:Is the claim that Rockies hitters actually hit worse in other ballparks than every other hitter? So much worse that over a season we shouldn't expect any bump in stats from hitting at Coors?

Now that right there is a very interesting question.
Yes doctor, I am sick.
Sick of those who are spineless.
Sick of those who feel self-entitled.
Sick of those who are hypocrites.
Yes doctor, an army is forming.
Yes doctor, there will be a war.
Yes doctor, there will be blood.....
Mod in Retirement

Posts: 53856
(Past Year: 1)
Joined: 29 Apr 2003
Home Cafe: Baseball
Location: Taking Souls...

Agreed with Erboes and ramble; at first glance, this does seem odd. Time to take a closer look...
Arlo
Hall of Fame Hero

Posts: 8847
Joined: 4 Jan 2002
Home Cafe: Baseball
Location: The Cheap Seats

Yep, that's what they're saying. At first glance, it's seems ludicrious since the Rockies have led the NL in runs and average pretty much every season since their inception except for the last two (humidor?) and I don't think they've had that good of talent. I will look more closely at it when I find the time. What's the old saying? There are three types of lies: Lies. Damn lies. And statistics. I think this may apply here.
Erboes
Major League Manager

Posts: 1433
Joined: 26 Mar 2003
Home Cafe: Baseball

Ramble2 wrote:Is the claim that Rockies hitters actually hit worse in other ballparks than every other hitter? So much worse that over a season we shouldn't expect any bump in stats from hitting at Coors?

i would think the home/road splits tend to even things out but if you have daily transactions then you can get the good without the bad. example: i used jay payton as my 4th outfielder and played him for home games and select road games (since he wasnt completely terrible on the road). i got really good stats from him at home and missed all of the garbage on the road.
j_d_mcnugent
General Manager

Posts: 3766
Joined: 1 May 2003
Home Cafe: Baseball
Location: bored at work

Whhhhhew! Some interesting stuff. I'm browsing through some players' home and away splits and came upon this doozy.

Heltons at home (career):

Avg -- .378
HR -- 134
RBI's --- 453

Avg -- .294
HR -- 85
RBI's -- 287

To give you a better idea how good Helton is at home, if he played all his games at Coor's he'd average .378, with 45 homers and 117 rbi's. If he played on the road for all his games he'd be at .294 with 28 homers and 96 rbi's. My guess would be, if he were traded to a team with an average park he'd probably be a .310 hitter and about 30-35 homers and 100-110 rbi's, which leads me to believe that there is a "hangover effect", but not anywhere near close enough to negate Coor's advantages.

I think looking at the numbers before and after Coor's of a player is probably more accurate. Wilson, for example, average about .265, 30 homers, and 95 rbi's if his numbers were averaged for 600 at bats. The 600 at bats he had last season he hit in total .282, 36, and 141. His road splits were more aligned with his career averages if you pro-rate them over an entire 600 at abats: .260, 15, and 57.

Looking at Wilson, it seems like there wasn't anything below normal except for his average on the road. This may not be enough of a sample size, but I'm beginning to believe this "hangover effect" is highly inflated.

This also gets me thinking. Wilson's road numbers were pretty much his average over his career. Maybe he didn't have a great season last season, but just an average one that was highly inflated by Coor's. Just something to think about I guess.
Erboes
Major League Manager

Posts: 1433
Joined: 26 Mar 2003
Home Cafe: Baseball

I still don't know what a hangover effect is.
"The game has a cleanness. If you do a good job, the numbers say so. You don't have to ask anyone or play politics. You don't have to wait for the reviews." - Sandy Koufax
ramble2
General Manager

Posts: 2952
(Past Year: 5)
Joined: 27 Feb 2003
Home Cafe: Baseball
Location: Salt Lake City

Next