Baseball’s Anti-TTO Ballpark

I can recall a handful of baseball stats that have just blown me away upon first viewing. Most recently, I was floored by advancements in pitch-framing research, and now I can’t think about any catcher without looking up how well he does or doesn’t receive. Obviously, PITCHf/x was sort of world-changing right away, and the same goes for the glimpses we’ve had of HITf/x. Years and years ago, I thought we solved almost everything with general batted-ball data, and I also remember opening a book and seeing batting averages and slugging percentages against specific pitch types. And there was an article I read in the Hardball Times, talking about various park factors. Some of them have been obvious for a long time. But it wasn’t until that day that I realized parks can affect outcomes like strikeouts and bases on balls.

There’s every reason for that to make sense. Every single ballpark is different, so in a way, every single ballpark’s version of baseball is different. The baseball will look different to the batter, and how the batter sees the ball is sort of one of the game’s fundamental components. The batter’s decision and swing lead to everything else. But what this isn’t is intuitive, or easily explained. People don’t pay much attention to these park factors, because they’re weird and ultimately not that important. Yet they exist and ought to be acknowledged, and one park in particular is extreme in a number of ways.

PNC Park probably doesn’t get enough attention for being strange. This is presumably because it gets most of its attention for being spectacular. When people put together rankings of ballpark experiences around MLB, PNC is usually at or near the top of the list, for its beauty and sight lines and goods available for purchase. Less attention is paid to how PNC plays, as a stadium. It’s known to be somewhat pitcher-friendly, but it was never considered on the level of an old Petco or an old Safeco or AT&T. There hasn’t been a growing demand for PNC to bring in its fences.

It’s worth noting that PNC is tough on homers. By our own park factors, only the Giants and Marlins play in more dinger-suppressing home environments. Breaking it down by handedness, PNC is the seventh-toughest place on left-handed dingers, and the very toughest place on right-handed dingers. It’s a narrow edge, but in this way, PNC is the league’s most extreme something. Righties might not have a harder time going deep.

But PNC goes further than that. It’s easy to understand why a park might inflate or suppress homer totals. It isn’t so easy to understand why PNC reduces walks more than anywhere else, or why it also reduces strikeouts more than anywhere else. These are difficult concepts to grasp, but the data is the data and the absence of clear explanations doesn’t invalidate what we see in the numbers.

Over the last five years, Pirates batters rank 30th in home walk rate over road walk rate. They rank 29th in home strikeout rate over road strikeout rate. Pirates pitchers rank 27th in home walk rate over road walk rate, and they rank 26th in home strikeout rate over road strikeout rate. These are our FanGraphs park factors, and you can play around. Sort the SO and BB columns and you find the Pirates toward one end. Keep in mind, also, that those park factors have been halved.

You’re familiar with the concept of the three true outcomes. They’re the three FIP components, those being walks, strikeouts, and home runs. Let’s come up with an overall combined TTO park factor, by simply summing the factors for the three components. That’s what you’ll find in the sortable table below.

Team HR SO BB TTO
Pirates 90 95 96 281
Giants 89 99 99 287
Marlins 89 99 101 289
Twins 93 98 99 290
Cardinals 92 99 100 291
Royals 94 97 100 291
Nationals 99 97 96 292
Angels 95 100 97 292
Athletics 92 99 101 292
Red Sox 97 99 98 294
Tigers 99 96 99 294
Dodgers 99 100 97 296
Indians 97 101 98 296
Rays 96 101 99 296
Mariners 97 102 99 298
Braves 97 102 100 299
Diamondbacks 103 99 98 300
Padres 98 102 100 300
Cubs 102 100 102 304
Mets 103 102 100 305
Phillies 105 101 101 307
Rangers 107 98 102 307
Blue Jays 107 102 99 308
Astros 104 103 101 308
Orioles 110 99 100 309
Rockies 113 96 101 310
Yankees 110 100 101 311
Reds 112 102 100 314
Brewers 110 102 102 314
White Sox 112 102 107 321

Toward one extreme, there’s U.S. Cellular Field. A known bandbox, it inflates homers, strikeouts, and walks. At the other end is PNC, reducing all three pretty heavily. It’s not simply in order of pitcher-friendliness, because pitchers want more strikeouts, and this table also isn’t weighted, but it is sort of in order of balls in play. In three different ways, PNC encourages hitters to hit the ball and hit the ball fair, in front of the fences. The park is a full two standard deviations from the TTO mean.

The end result of everything is simple. Put it all together and PNC is pitcher-friendly, a little bit. The game is built on runs, and in Pittsburgh, there will be fewer runs. That’s not unlike saying that Matt Garza, Ervin Santana, and Ubaldo Jimenez are all probably roughly three-win pitchers. That’s true, and it gets to the heart of the matter, but what’s really interesting are the details and the processes. The things that make the free-agent starters different. The reasons why PNC is what it is. We know that the average plate appearance isn’t changed very much by whatever it is that’s causing PNC to reduce walks and whiffs. But we also know, beyond most reasonable doubts, that PNC does reduce walks and whiffs, along with home runs, and that’s fascinating. Why does PNC lead to a greater rate of balls put in play? How does this affect the rest of the game? Is there a way for the Pirates to leverage this? What does this mean for player evaluation? In what way is PNC’s pitcher-batter interaction unique?

I’m not sure the average person appreciates that every single ballpark is incredibly complicated, and little things you wouldn’t think would make a difference can make a difference in the actual gameplay. There are obvious ballpark traits and less obvious ballpark traits, with less obvious consequences, but there are so many consequences and the ones in PNC make it particularly unusual. There’s something going on there — somethings — and there’s science to be done. As long as there’s science, baseball will never grow stale.



Print This Post



Jeff made Lookout Landing a thing, but he does not still write there about the Mariners. He does write here, sometimes about the Mariners, but usually not.


Sort by:   newest | oldest | most voted
MCHC
Guest
MCHC
2 years 6 months ago

Judging by the teams at the top, I assume Barry Bonds just used up all the HR and BB of those two teams for the next few decades.

bob
Guest
bob
2 years 6 months ago

The first thing I can think of is that pitchers are more willing to pitch to contact in a park like PNC because they know how hard it is to get really hurt there by the long ball. In more hitter friendly parks I am sure the pitchers are prone to nibble more knowing a lazy fly ball might end up in the sets.

I am also curious to know just how much the quality of a team’s pitchers and lineup has on the park factor itself. Is there a breakdown? I am sure the park isn’t solely responsible for the results. How much does a bad pitching staff play into it or a very good hitting lineup?

Brad Johnson
Member
Member
2 years 6 months ago

That was my first thought as well, but why don’t we see Oakland or San Diego higher on the list due to their reputation?

Eyeballing the list, it looks like there’s a rough correlation between HR factor and the others. But for all stadiums besides PIT and WAS, the factors for SO and BB are barely divergent from average. Perhaps those specific teams are employing a targeted, team-wide approach while some of the other stadiums have only a guy or two pitching differently? Or maybe it is mostly tied up with stadium effects like backdrop rather than pitcher strategy.

Pitnick
Guest
Pitnick
2 years 6 months ago

“employing a targeted, team-wide approach”

This sounds plausible, as Pittsburgh, anecdotally, seems big on team-wide strategy and has a very involved coaching staff, preaching grounders, using aggressive shifts, etc.

JCA
Guest
JCA
2 years 5 months ago

McCatty in Washington is a big believer in pitching to contact. I could see wanting more contact in Washington than in Philly or Atlanta, which would influence the home / road ratio (you also see this for the Marlins and the reverse for the Phillies).

shthar
Guest
shthar
2 years 6 months ago

96 to 107 is not much of a spread.

95 to 103 is even less.

I’m not seeing anything here worth studying.

Brad Johnson
Member
Member
2 years 6 months ago

Spreads of 8 and 11 percent are not small over a full 81 game season. Perhaps I’ll do the math later to see just how many BB and K that is, now is not the time :)

shthar
Guest
shthar
2 years 6 months ago

But the spread is only that big comparing 2 of the parks, the two extremes. Most of them, it’s not even 5%.

It’s like 3.2 beer.

RC
Guest
RC
2 years 5 months ago

THe vast majority of baseball players live in an 8-10 point spread. 8-10 points is the difference between a guy who hits .260 and a guy who hits .350.

bdhudson
Member
Member
bdhudson
2 years 6 months ago

God forbid we study the minor differences that can affect the outcomes of games!

Dirty Larry
Guest
Dirty Larry
2 years 6 months ago

Great post here. From where I sit at one of the top pitcher-friendly parks, my view is that you need to get into the head of the pitcher to understand the lack of TTO at PNC Park. Pitchers feel the outfield is fair to roomy, so attacking the zone early in the count is an easy thing to do. Challenge the hitter with your best stuff with confidence and good things will happen. You g guys pitching in Philadelphia or Cincinnati are in a different mindset. Baltimore is tough as well, especially with no pitcher killing a spot in the batting order for you. So you are aggressive early, rather than nibbling…thus more likely to get ahead of the hitter. That means more strikeouts and fewer walks. A same of 10+ seasons is quite valid in this regard. Another tho g to consider is the proximity of the backstop to home plate; the closer in the wall behind the plate is, the more the pit her feels he is on top of the hitter. It is a subtle but important thing for pitchers. PNC, Nats Park, Target Field are all set up this way and help the perspective of the man on the mound.

Go Nats
Guest
Go Nats
2 years 6 months ago

Yes but this sees fewer strike outs and fewer Walks, so what you say is not explaining anything useful to this paper.

BaseballGuy
Guest
BaseballGuy
2 years 6 months ago

This seems kinda silly, really looking hard for something interesting when it’s not there. Wouldn’t it be a lot simpler to say that the Pirates (whether by design or not) have a slightly higher than normal number of patient hitters who walk and strike out a lot, and a lot of big-armed pitchers who walk and strike out a lot of hitters? Why bring the ballpark into it at all?

Brandon Firstname
Member
2 years 6 months ago

Because that’s not how park factors work. They are based on home/road splits for both the home and away teams, so overall player skill is (mostly) factored out.

uniqueusername
Member
uniqueusername
2 years 6 months ago

Road numbers are slightly weighted to the Division they play in. 46-47% Of the games compromise in Division.

Baseball Splits (Twitter)
Guest
Baseball Splits (Twitter)
2 years 6 months ago

It seems silly to call something “silly” when you don’t understand what it means :)

Boris Chinchilla
Guest
Boris Chinchilla
2 years 6 months ago

Your avatar is silly

Kris
Guest
Kris
2 years 6 months ago

Well, besides that not being how park factors work, the conclusions you draw from this info (if that were how it worked) should be the opposite anyway. That Pirate hitters *don’t* walk or strike out a lot, and neither do their pitchers.

But in reality, Pirate hitters and pitchers walk a lot less and strike out a lot less *when they play at PNC* compared to when they play at other parks. Why is that? This is the actual interesting question the article is wondering about.

Andy
Guest
Andy
2 years 6 months ago

I suppose the Rockies have just been among the best hitters in baseball for their entire existence. Their park obviously has nothing to do with it.

nickolai
Guest
nickolai
2 years 5 months ago

Dood, did you even read the article? Try again…

Nick
Guest
Nick
2 years 6 months ago

Chicago White Sox/Adam Dunn is bad joke!

Steve B
Guest
Steve B
2 years 6 months ago

How much of the differences have to do with which teams played in the ballpark? If the home team hasn’t hit well over the last 5 years, wouldn’t that be a major factor? Also, playing in a weak/strong division… etc.

Baseball Splits (Twitter)
Guest
Baseball Splits (Twitter)
2 years 6 months ago

No. Park factors compare how a team hits at home with how they hit on the road, so if a ballpark has a reputation of being a “pitcher’s park”, it means that team’s offense performs better on the road than they do at home.

Bob II
Guest
Bob II
2 years 6 months ago

What bob said. Pitchers in parks like Coors and Great American Launching Pad will likely have a conscious game plan to avoid the strike zone, and pitches out of the zone lead to more walks and more strikeouts.

Dr.Rockzo
Member
Dr.Rockzo
2 years 6 months ago

This doesn’t hold as valid given the other information available. GABP has roughly the same k/bb rates as ATT despite a dramatically higher HR rate? If the idea is that HR factor serves as a function to affect k/bb rate, then shouldn’t those parks with varying HR factors also have divergent k/bb factors?

Part of the issue is sample size as a size of 30 really isn’t very informative, but the trend in the parks themselves do not support the notion that HR factors affect the k/bb factors of the park because of style of pitching.

The Foils
Member
The Foils
2 years 6 months ago

So many armchair analysts who either (a) don’t know what park factors are or (b) did not actually read the article.

Mr Punch
Guest
Mr Punch
2 years 6 months ago

There are a lot more parks with little foul ground than there used to be in the days of all-purpose stadiums, but this remains a factor that varies widely. I’m surprised by the lack of evident impact (evident to me, anyway) on SO/BB.

Dave
Guest
Dave
2 years 6 months ago

Could the Pirates spread be because over the years they’ve generally had a pitching staff that was more of a pitch to contact pitching staff? Last season was really the first year in which the Pirates actually had 3 guys who could strike out 7 or more during a game in a long time.

Go Nats
Guest
Go Nats
2 years 6 months ago

If the park factors are true, then the park is turning the Pirates pitchers into pitch to contact pitchers that somehow give up fewer homers than one would expect from such pitchers. On the road they are not pitch to contact pitchers and they give up more homers.

RC
Guest
RC
2 years 6 months ago

In addition to the impact these numbers should have on player development and roster construction, they should also greatly influence managerial decisions. For example, if you play in a park that yields fewer homers and walks, then the hit and run would be a more viable strategy. After all, the pitcher is going to come right after the hitter. He is not going to be nibbling very often.

john
Guest
john
2 years 6 months ago

maybe foul territory? If the foul territory in PNC is bigger than most parks, foul balls in other stadiums may become outs, which would reduce the chance of getting a strikeout or a walk in that at-bat

Baseball Splits (Twitter)
Guest
Baseball Splits (Twitter)
2 years 6 months ago

I believe Oakland has the biggest foul territory though (by far), and their K & BB rates were right at average.

bradsbeard
Guest
bradsbeard
2 years 6 months ago

Is there a park factor published somewhere for BABIP? Not really sure how or if it would be useful to compare with the TTO park factors, but it seems like it should be somehow.

Joseph Ruth
Guest
Joseph Ruth
2 years 6 months ago

Interesting but more info would be nice. For example, do you have a link for the Hardball Times article?

Personally, I think tactics are hugely more important than minor differences in park conditions for things like walks and strikeouts but that’s just my own opinion. Some of the points brought up in the comments seemed pretty pertinent as well.

Why don’t they just asked players (and x-players) like Greg Maddux, Randy Johnson, and Jamie Moyer what they think instead of merely adding up some numbers and hoping they make sense? For players like these, this type of info is professional life an death and I’m sure they know it or have a feel for it. The idea would be to take their general notions and THEN start data diving.

Go Nats
Guest
Go Nats
2 years 6 months ago

so you think that baseball players and managers play much differently in home games than in road games because they have radically different road and home strategies that have nothing to do with the ball park?

uh why?

Why would a team only hit and run in home games if their was no advantage to hitting and running at home? Just to be colorful?

RC
Guest
RC
2 years 5 months ago

People do things that aren’t logical or beneficial all the time.

joser
Guest
joser
2 years 6 months ago

This Hardball Times article, probably.
http://www.hardballtimes.com/main/article/home-run-park-factor-a-new-approach/
Note the date and author, and then go look at the credits at the top of ESPN’s home run tracker page.

Spit Ball
Guest
Spit Ball
2 years 6 months ago

As a 30 year fan of the team that plays in Fenway, park factors don’t seem to go far enough. I’ve been thinking about park effects since Bill James was predicting Tim Naehring’s breakout in the 89 abstract.

joe
Guest
joe
2 years 6 months ago

Park Chan Ho's Beard
Guest
Park Chan Ho's Beard
2 years 6 months ago

Thanks for the input, joe

coldseat
Guest
coldseat
2 years 6 months ago

Whitey Herzog had all this figured out in the 80’s. Reading his books and interviews teases all this stuff out.

Matt
Guest
Matt
2 years 6 months ago

Is it possible that outside factors like elevation, sunlight, humidity, wind, rain, etc. factor into this?

I remember that one of the reason that pre-humidor Coors field increased offensive statistics was because the thin air placed less resistance on the baseball causing less movement. Less movement may translate to more pitches over the plate which would reduce both BB and K rates.

I don’t see clear geographic relevance here – but perhaps measuring the air density, and similar factors, at each ballpark may lead to your causing factor.

Gyre
Guest
Gyre
2 years 6 months ago

Is it April 1st already? Wow, time to watch games again!

isavage30
Guest
isavage30
2 years 6 months ago

It makes sense that the park would impact Ks and BBs. If you’re playing in a part where it’s hard to hit the ball out, it would make sense as a pitcher to challenge hitters and reduce walks, and for hitters to cut down on their swing and try to put the ball in play. It’s probably difficult as a player (hitter moreso than a pitcher) to have a different approach based on the stadium you’re playing in, but ideally you would adjust your game to the park.

Jim Garman
Guest
2 years 5 months ago

As to employing team wide hitting strategies. The very first thing Davey Johnson did when he took over from Riggleman was to institute a team wide hitting approach. That combined with the new stadium bears some study. Thanks for starting this discussion.

Plucky
Guest
Plucky
2 years 5 months ago

Has anyone gone and quantified park effects at the PITCHf/x level? i.e., comparing all pitchers homve-vs-road averages on pitch movement to say things like “in park X, curveballs move .25 inches more on everage?”

Matt
Guest
Matt
2 years 5 months ago

Pointing towards one exception doesn’t nullify a trend. The fact of the matter is that there is some correlation between HR and K/BB park factors.
The stronger correlation is between HR and BB (0.49), which makes sense if you think that pitchers might be less afraid to pitch in the zone if the park suppresses homers. The correlation between HR and K is less strong (0.30), but jumps significantly (to 0.46) if you exclude Coors, which has the biggest gap, probably because of the effect of altitude on pitch break.

Josh
Guest
Josh
2 years 5 months ago

On a similar note to Dave and Steve B, isn’t it also a factor how certain teams play? For example, the Twins are known as a ‘pitch-to-contact’ team, they don’t hit many homers and their pitchers chased records last year for fewest k’s in a season. I just think you can’t equate how a ballpark plays with TTO because how good the team that plays there is, especially at generating TTO, matters.

Jon
Guest
2 years 5 months ago

You guys are over thinking this. PNC is hard to hit home runs in. Now consider the ramifications. The teams won’t employ a one base at a time conservative strategy. They are going to play smallball. This means teams are going to bunt more often. Bunts are typically not hard to lay down, especially for a National League team. Each bunt in essence shortens the game by an out because the offense is just giving the defense an out. Teams will employ the hit and run more. This means the batter is asked to swing when they may not have. Typically, these plays aren’t called on three ball or two strike counts. If the batter is asked to force contact a strikeout or walk wont happen. With less home runs, the bases are empty less often, which means there are more runner on third less than two situations. The batter usually has a different approach during these situations; he generally isn’t looking for a walk, but to drive in the baserunner.

Baseball Splits (Twitter)
Guest
Baseball Splits (Twitter)
2 years 5 months ago

So, why is Marlins Park right at average with K+BB? According to the PF, it was even harder to hit a HR there.

Jon C
Member
Jon C
2 years 5 months ago

Because there are always outliers. It’s disingenuous to point out a single outlier and say “this proves you wrong”. The simple fact is TTO corresponds almost directly with home run rate.

Jon C
Member
Jon C
2 years 5 months ago

Because there are always outliers. It’s disingenuous to point out a single outlier and say “this proves you wrong”. The simple fact is TTO corresponds almost directly with home run rate.

The TTO total is loosely three times that of the home run rate in the vast majority of the parks. This isn’t coincidence.

Jon
Guest
2 years 5 months ago

Also less home runs will lead to closer run differentials. This means more close games and more close games leads to more smallball play as well. This chart isn’t all that surprising.

Forrest Gumption
Member
Forrest Gumption
2 years 5 months ago

Isn’t Adam Dunn a TTO hitter no matter what team he plays for? If he was on the Pirates they wouldn’t be #1 on your list.

A TTO hitter is not going to not be a TTO hitter because of the park. I don’t understand this piece.

wpDiscuz