Skip to content

Revisiting AT&T Park’s Affect on Hitters

December 15, 2010

A few days back, Scott Ostler of the San Francisco Chronicle asked a question that’s been plaguing the minds of many Bay Area sports fans for quite some time:

“Why is it that when San Francisco and Oakland teams phone great athletes around the country to gauge their interest in coming here to play, the players fake a foreign accent and say it’s the wrong number?”

I don’t know about foreign accents, but I do know that the Giants have had plenty of issues with bringing free agents to don the orange and black.  This is most likely due to AT&T Park because, as Ostler puts it, it “(…) scares them.”  The question is, of course, “should it?”

I took a look at AT&T’s component park factors based on batter handedness back in July and found that the overall effect was essentially the same between left-handed and right-handed hitters—the park suppressed their production, but not by a whole lot.  Not enough, in my opinion, for so many free agents to spurn the Giants so quickly.  I looked at 2005-2009 rather than the entire decade in part due to the Barry Bonds effect—I thought Barry would have certainly skewed the results for lefties, and I didn’t think it necessary to include every single year of the park’s existence.

I really don’t know what I was thinking.  The Bonds Factor is an easy thing to adjust for—simply subtract Bonds’ home batting line from all left-handed hitters at AT&T Park.  And, of course, when it comes to estimating the effect of a park on hitters, the more data you use the more accurate an estimate you’ll have of how the park truly plays.  Three years is a good starting point, five years is better, and ten years and beyond is ideal.  Chances are you won’t find a large discrepancy between five-year park factors and ten-plus year park factors, but it should help give a more refined estimate.  So I thought I’d update my park factors for AT&T and see how it looks if we include all years rather than just the most recent ones.  The methodology hasn’t changed—all park factors are based on the rate of the event at AT&T Park compared to all other National League parks, regressed 10%.  I toyed with the idea of using different rates per component (i.e. 3B/2B rather than 3B/PA) but found little difference in the results, so I’m sticking with rates per plate appearance for now.  First, the right-handers:

The biggest change we see by expanding the dataset is the effect of triples—rather than being moderately favorable towards hitters, it’s a bit bigger than what I previously thought.  Everything else is essentially the same.  Now the left-handers:

No substantial changes, but we do see triples having a slightly larger effect, home runs being suppressed a bit more, and ROE moving closer towards the mean. Now here’s the interesting part: what if we include Bonds in the calculations?

Good stuff.  Barry changes things up a bit, mostly in the home run department, although we see a decrease in on-base events such as NIBB and HBP by excluding him.  It’s crazy to think that one player who accounts for about 7.6% of all plate appearances at the park could make an impact like that.

Based on the 2000-2010 run environment, right-handed hitters created 75.2 runs in 650 plate appearances (approximately a full season).  If we apply the park factors to their line, it drops down to 73.7 runs; meaning that, on average, a right-handed hitter will be expected to perform at 73.7/75.2 = 98% of their performance in a neutral environment.  For a left-hander, the effect is actually more dramatic—rather than creating 79.5 runs, he would be expected to create 75.2, retaining 95% of his park-neutral performance.  It appears that AT&T certainly does favor right-handed hitters over lefties.  Both hitters will see their numbers suppressed, but the effect on righties is relatively small and with lefties a bit more pronounced—roughly five runs of value over the course of a full season; about half a win.  Generally speaking, it looks like left-handed hitters should be a bit wary of hitting at AT&T.

Lest anyone take this little “study” as definitive proof that left-handed hitters should avoid playing at AT&T Park, I’d like to point something out.  We’re estimating the overall effect of the park; certain hitters will be affected differently than others.  This does not mean that a player like Adam Dunn, who hits moonshots with regularity, would see his home run production drop 18%.  The big-time power hitters would likely see a loss of a few home runs per season but nothing substantial.  The player’s batted ball distribution will also lend some insight as to how much they may be affected at the park- if you’re a dead pull hitter, you won’t be as affected as a guy that tends to hit home runs out to right-center.  If you’re a player that likes to line doubles into the gaps, you’re likely to wind up with a few more triples.

So, should free agent hitters be scared to sign with the Giants?  I’d say it really depends.  For a right-handed hitter, there really shouldn’t be a problem.  For a lefty, I’d say it depends on your hitting style.  If you’re a pull power hitter or a pure doubles hitter, you’ll be just fine.  If you hit home runs to right or left-center, you might want to avoid the park.

3 Comments leave one →
  1. December 15, 2010 8:09 AM

    I think that this is good information to know, I think that the reuptation of AT&T park for outweighs the actual numbers.

    The biggest reason is that people have a tendancy to remember the bad events, the 400 foot outs that don’t even make it to the warning track and don’t remember the double that they got down the line because the rightfielder was gaurding triples alley.

    That combined with the heavy air in the evenings that effects all of the westcoast ballparks where players don’t get that carry really gets in their head.

    I haven’t looked into this but do you know if thre are any problems with people not wanting to play in Dodger Stadium which has simlar effects at surpressing offense.

    • triplesalley permalink*
      December 15, 2010 10:47 AM

      “I haven’t looked into this but do you know if thre are any problems with people not wanting to play in Dodger Stadium which has simlar effects at surpressing offense.”

      Someday I’d like to build the same component PF for all NL and AL Parks- I have the data at hand already for the NL stadiums; the difficult part is finding out which parks saw a change to their dimensions and when.

  2. December 15, 2010 10:17 AM

    Awesome post.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: