|Is Marlins Park a pitcher's park?|
|Theory and Strategy - Platinum|
|Written by Todd Zola|
|Friday, 10 August 2012 10:57|
We have a pretty good idea why David Wright is smiling a lot more this season. Just look at his numbers. Is the change in dimensions of his home ballpark a factor in that? Meanwhile, down in Florida, the Miami Marlins are 25th in baseball in home runs at their park, yet their pitching staff is 21st in baseball in home ERA. What can we make of that?
In both these instances, the early returns on how Citi Field and Marlins Park are playing yield surprising results … which could offer the astute fantasy gamer an edge down the stretch.
Here's a theory: Park factors are useful only after a venue has three years of data to pull from. At least that's what has been drummed into us. This is not completely true, however. Conventionally, park factors are presented as a three-year rolling average, but some trends and generalizations can be gleaned from fewer data points, as long as the conclusions are applied in a global sense and used as an aid in making a decision, not as the primary deciding point.
The primary purpose of today's exercise is to get an early read on Marlins Park, and to determine if moving the fences in at Citi Field has had any effects. We'll also take a look at Target Field to see if our perceptions of it as being an extreme pitchers park continue to ring true.
Let's begin by looking at the home run and runs scored park indices from the past three seasons for each venue, then compare them to how the park is playing through the games of Aug. 5. We'll include only the 27 parks with three years of data, excluding Marlins Park, Citi Field and Target Field. The three-year average will be computed and compared to the present factor.
There are several versions of the park factor calculation. To compute the HR factor, the numerator is the sum of the home runs totaled and allowed by the home team, divided by the total number of at-bats for both teams; the denominator is the sum of the number of home runs totaled and allowed while the squad is on the road, divided by the total number of at-bats for both teams during these away games. This number is multiplied by 100 to yield the familiar index, with 100 being neutral. To determine the run factor, runs replace homers and games replace at-bats.
Here is the data looking at home runs going back to 2009. The final column labeled "Difference" measures how far off the present (2012) index is from the three-year average: a positive result indicates a higher-than-expected index and a negative outcome represents a lower-than-anticipated factor.
Now let's think about the park factor itself, and what it means for player performance. A player with a neutral home park should hit about as many homers at home as on the road. Let's say he hit 30 for the season, 15 home and 15 away. Using the same hitter with a home-field HR index of 110, his projected home total would then be 16.5, meaning he would have smashed 31 or 32 dingers instead of 30. If the factor were 90, his total would have been 28 or 29. In other words, 10 park factor units equal only one or two homers per season.What we'll do now is focus on the difference column and decide what that means in a big-picture sense. To do this, we must accept that park factors are best thought of as ranges and not a static number. It shouldn't be too hard to convince yourself of this; just look at the variance between the three years on almost all of the venues. If you are still reticent, draw a parallel to how you evaluate players when putting together your cheat sheets. Most employ value tiers, where players within each tier are considered the same, even though their dollar value or ranking may be a little different. All we are doing is "tiering" the ballparks.
Looking at the above data, 12 of 27 ballparks, or 44 percent, have a difference of 10 or fewer. Eighteen of the 27 ballparks (67 percent) show a difference of 20 or fewer. That means two-thirds of the teams sampled show a 30-homer hitter to still hit 27 to 33 homers in their ballparks.
Now let's look at the park factor for runs.
In this table, 13 parks fall within 10 units of the anticipated, and 21 of the 27 teams within 20 units of what is expected. The player impact is a bit different, as runs are more relevant to pitchers. In a neutral venue, a pitcher with an ERA of 3.50 over 180 innings surrenders 70 earned runs. This amounts to about three to four runs per every 10 park units. Even 10 park units are only about seven runs. You may question the word "only," but realize that over the course of a season, so many other factors go into runs allowed (which is why we spent two previous installments looking at the components of ERA) that even a difference of 20 park units is going to be just one of many contributors, not the deciding factor.
What does this all mean in terms of small sample reliability? Well, let's recall what we are trying to accomplish: deciding how dependable the early returns are for Marlins Park and Citi Field. Obviously, the certainty is not absolute. But we are at the point in the season where even the slightest edge could be the difference between winning and losing, especially in head-to-head formats. Sure, it would have been better if all the teams were within the discussed ranges, but what we have is pretty significant, especially when you consider how many other decisions are made with data that have similar noise. Your mileage may vary, but at this point, I'm comfortable looking at the data for Marlins Park and Citi Field in an effort to glean an edge down the stretch.
As of this writing, the HR index for Marlins Park is 87, and the run index is a somewhat surprising 108. The perception coming into the season was that the new facility would favor pitchers. Well that's half right; homers are indeed depressed. But overall, run scoring is above average, even on par with U.S. Cellular Field and Yankee Stadium. While the quality of opponent is more important than the park factor, I am certainly going to think twice before I automatically start Mark Buehrle, Nate Eovaldi, Wade LeBlanc or especially Ricky Nolasco just because the Marlins are at home.
The same holds true for marginal visiting pitchers, though the Triple-A lineup the Marlins currently have trumps the park index. I'll certainly trust a visiting pitcher with a strong strikeout rate facing the mediocre Marlins attack. With regard to enemy batters: Although homers are depressed, run production in general gets a boost, which buoys the RBI and runs scored totals.
The Citi Field construction workers who spent the winter moving the right-field fence closer to home plate, not to mention David Wright and Ike Davis, should be pleased. The renovation seems to have had the desired effect. Homers have risen to an index of 109 at Citi Field after averaging 84 for the ballpark's first three campaigns.
Curiously, the runs index is actually down a few points, sitting at 83 after settling around 91 previously. If this holds fast, it is the best of both worlds; the Mets' home run hitters can continue to add to their totals, but their pitchers need not be concerned that their ERAs will be negatively affected by the closer and shorter fences. To be completely frank, I'm skeptical the run index will continue to be that low. However, the reasons for my skepticism extend beyond the scope of this treatment (but are tentatively scheduled to be discussed, along with a more in-depth treatment of park factors in general, during the offseason).
I suspect Citi Field will still be pitcher-friendly, just not to the extent currently thought. As such, the Mets' staff is still fine to start at home. And if you have been benching opposing hitters, especially lefties, at Citi Field, it is time to loosen the reins. It's not exactly Coors Field, but home run hitters should no longer be intimidated by the place.
Before calling it a day, let's take a gander at Target Field, revealing what may be surprising results. Here's the data for its first two seasons, along with the current campaign:
It sure looks as though Target Field is not quite as daunting as we first thought. Reasons for the above fluctuation in HR index will be broached in the future, but the take-home lesson is that Target Field, while still not favorable for power, is not nearly as intimidating as it appeared to be in its inaugural season. Not to mention, with respect to run scoring, it has been playing almost neutral the entire time.
In my eyes, this is especially relevant in leagues in which starts and innings are at a premium. Minnesota arms are even less desirable; their wins potential was already suspect, and now it appears a good offense can inflict some damage that the park will not suppress. React accordingly.
|Last Updated on Friday, 10 August 2012 11:08|