Should You Build Your Staff To Fit Your Home Park?

You play 81 games at home a year, so it seems like it might be a good idea to think about that park when you’re building your team. Then again, you play 81 games on the road, maybe it’s not a good idea to worry too much about one half of the ledger, particularly if your home park is an extreme one.

Extreme parks lead to extreme home-road splits. That part seems obvious, but it bears out in the winning percentage, too. Take a look at how teams that have called extreme parks home have faired over the last five years compared to the middle.

Teams with extreme parks have won at home and have had a harder time on the road, leading to larger home-away win differences:

	HR Park Factor	H/A W Split
Pitcher-Friendly	90	36
Neutral	100	25
Hitter-Friendly	112	38

But saying that teams like the Pirates — with a +49 at home (second-largest in baseball) and a 88 home run Park Factor — have extreme splits, at least somewhat due to their home parks, is just descriptive. We might not be able to figure out the mechanism that leads to this difference, but we should try. The Rockies (+70 h/a, 116 HRPF) would like to know if there’s hope.

Could these teams be building their staffs to fit their home parks? Could that be a bad idea?

For the purposes of comparing teams, we could look at how many balls a team allows into play — including homers, which are often removed from BIP equations. So our easy BIP is just 1-((K%+BB%)/PA). We could throw hit by pitches in there but it’s not going to move the needle much at this level. It’s a decent rough estimate.

Look at this stat, and it leaps out of the page at you that the Giants (.741) and Cardinals (.742) allow balls into play more than 80% of baseball, and they have nice home parks for that sort of thing. Both are top-half teams when it comes to sinker and two-seamer usage, too. But if we break up our groups by park factors again, you’ll see little difference in our BIP percentage.

	HR Park Factor	BIP%
Pitcher-Friendly	90	0.727
Neutral	100	0.724
Hitter-Friendly	112	0.728

We are using home run park factors to separate our parks. Maybe there’s something going on in the home run splits? Well, there is. Something. Not sure what, though.

	HR Park Factor	Home HR/FB	Away HR/FB
Pitcher-Friendly	90	8.0%	10.5%
Neutral	100	9.7%	10.3%
Hitter-Friendly	112	12.1%	9.9%

The home-away splits on home runs per fly ball rates are strange. They get even stranger if you put all of baseball into just two buckets. Teams with park factors under 100 gave up five more homers per 1000 fly balls on the road than teams with hitter-friendly home parks (10.5% to 10%). That doesn’t seem like a lot, but we’re talking about five years of fly balls, and almost 100,000 away fly balls.

Why would pitcher-friendly park staffs give up more homers per fly ball on the road than league average, which hovers around 10%?

There’s evidence of higher home run per fly ball rates for ground-ball pitchers, but our staffs don’t line up like that over the last five years at least.

	GB%	Total HR/FB	Away HR/FB
Fly Ball Staff	43.0%	10.1%	10.5%
Neutral	44.5%	10.0%	10.1%
Ground Ball Staff	46.8%	10.0%	10.5%

The pitcher-friendly park staffs had a 45.3% ground-ball rate, and the hitter-friendly park staffs had a 44.7% ground-ball rate. Could that be it? Seems like the difference is too small. We can’t confidently use ground-ball rate to say that teams in certain parks are targeting ground-ball pitchers and therefore suffer more home runs per fly ball when they leave those parks.

Maybe the Pirates have done this. They use the most sinkers in the game and have the second-best ground-ball rate over the sample. They’ve built up quite a home field advantage. But they’ve paid for it, with the worst away HR/FB in the game (12%). And it doesn’t look like this is happening as a rule.

Five years is a lot of time and maybe it isn’t. The Giants pitching staff over the last five years has had many of the same faces on it, at least. And so maybe we’re just seeing the peculiarities of a few team staffs over time.

When you start with only 30 teams, you’re always cutting your sample size down when you try to compare them, too. We only had six teams in our extreme bins, maybe we can’t say anything about even 12 teams in extreme parks.

But it’s weird how, when you compare the extremes, things come around and loop back together. If your team plays in an extreme park, they’ll do a lot better at home than they do on the road when compared to neutral park teams. Home run rates play some part in that difference, too.

Ground-ball and ball in play rates, at least, don’t seem to show that pitcher-friendly park staffs are willing to let the ball in play and hitter-friendly park staffs are trying to strike everyone out. It doesn’t look like teams build their staffs with only their home park in mind. You have to play 81 on the road, after all.

Here’s the summary of the data set I was working with. Thanks to Jeff Zimmerman for pulling some of the split data.