Predicting perfection

Bob Levey

In the aftermath of Yu Darvish's second near miss with perfection against the Houston Astros this season, we look at whether there is a particular batting profile that predicates a team to being on the losing end of a perfect game.

This week produced yet another dominating outing from Texas Rangers starter Yu Darvish against the Houston Astros in the form of an eight inning, one-hit performance, which saw him take a perfect game into the sixth inning. Between this performance and his season-opening, almost-perfect masterpiece, you can't help but feel that Darvish has cracked the code to complete immobilization of the Astros lineup. When you also consider it was about a year ago when Matt Cain threw his perfect game against the Astros, you sense that there might be something innate within the Astros lineup that puts them in the unenviable position of perfect game fodder.

It wasn't too long ago -- okay, fine, it was last week -- where we discussed the slight differences between a pitcher pitching himself into the history books with a perfect game and the almost perfect one-hit, complete game. From that analysis, we learned that, not shockingly, strikeout rate played a commanding role, with ground ball rates playing a larger role than fly ball rates in determining perfect game nirvana. However, the focus in that analysis was on the pitcher -- with the results of April 2nd and August 12th Darvish outings against the Astros, let's look again at what makes a perfect game perfect, but this time, focus on hitting results. What tendencies does a team display that might predispose them to being on the wrong end of a perfect game?

Let's start with some data, all of which are courtesy of Baseball Reference. For this exercise, we will query the data available for all perfect games since 1988, as well as some team specific and MLB-wide averages for a handful of hitting statistics. 1988 is the cutoff simply due to the fact that some of the data we are interested in were not recorded before this season or are incomplete. So with that caveat, what stats will we look at, and why?

Broadly, a perfect game is obviously the result of no hits or walks allowed and as we previously mentioned susceptible to striking out a lot as well as hitting into more ground ball outs than fly ball outs; what kind of lineup would be at risk of satisfying these criteria?

After some rumination, I came up with seven stats: batting average on balls in play (BABIP), walk rate (BB%), strikeout rate (K%), ground ball to fly ball ratio (GB/FB), home run to fly ball rate (HR/FB), balls in play rate (IP%), and first pitch swinging rate (1stSw%). The hypothesis is as follows:

Teams that strikeout a lot (high K%), don't walk a lot ( low BB%), tend to be impatient or aggressive hitters (high 1stSw%) and don't put the ball in play as often as their opponents (low IP%) will be more susceptible to perfect games. Add to it the idea that when they do put the ball in play, it doesn't become a hit or home run (low BABIP and HR%) and is more frequently a ground ball than a fly ball (low GB/FB), and we have our rubric for perfect game susceptibility.

Now, let's move on to see how this hypothesis holds with the data. The first table presented are the aforementioned stats for each of the teams that have suffered perfect game infamy since 1988:

Year Team BABIP BB% K% GB/FB HR/FB IP% 1stSw%
1988 Dodgers 0.281 7.2 15 0.85 7.4 73 39
1991 Dodgers 0.287 9.5 15.5 0.93 5.5 71 32
1994 Angels 0.294 9.1 16.1 0.89 8.2 71 29
1998 Twins 0.297 7.3 14.6 0.95 5.4 74 31
1999 Expos 0.292 7.1 15.3 0.86 7.8 73 34
2004 Braves 0.309 9.3 18.3 0.87 8.3 68 32
2009 Rays 0.303 10.3 19.8 0.68 9 66 28
2010 Rays 0.309 10.7 20.6 0.73 7.7 65 29
2010 Marlins 0.293 8.3 22.3 0.83 7.7 65 25
2012 Rays 0.284 9.4 21.7 0.76 9 65 30
2012 Astros 0.288 7.7 22.7 0.86 7.9 65 28
2012 Mariners 0.276 7.7 20.8 0.76 7.4 68 28

... with the MLB average for each of those stats for that year.

Year BABIP_m BB%_m K%_m GB/FB_m HR/FB_m IP%_m 1stSw%_m
1988 0.282 8.2 14.7 0.80 8.2 74 33
1991 0.285 8.7 15.2 0.80 6.2 72 30
1994 0.300 8.9 15.9 0.85 7.9 71 31
1998 0.300 8.7 16.9 0.80 7.8 70 31
1999 0.302 9.4 16.4 0.78 8.4 70 29
2004 0.297 8.6 16.9 0.80 8.1 70 28
2009 0.299 8.9 18 0.78 7.8 69 26
2010 0.297 8.5 18.5 0.80 7.3 69 26
2012 0.297 8 19.8 0.84 8.1 68 27

So our hypothesis looks something like this:

BABIP < BABIP_m

BB% < BB%_m

K% > K%_m

GB/FB > GB/FB_m

HR/FB < HR/FB_m

IP% < IP%_m

1stSw% > 1stSw_m

Comparing table one to table two, these are our results for each of the 12 perfect games of each of the stats, using the italicized comparisons:

BABIP
BB%
K%
GB/FB
HR/FB
IP%
1stSw%
8 of 12 (.67) 6 of 12 (.50) 10 of 12 (.83) 8 of 12 (.67) 6 of 12 (.50) 8 of 12 (.67) 10 of 12 (.83)

Here, we see that eight out of 12 perfect games involved a team with a BABIP lower than league average, six out of 12 with a walk rate lower than league average, and so on. Statistical testing showed no significant differences between the perfect game team rates and the MLB averages for that season, but with this third table, we see that a number of our stats do help explain some of what a team does (or doesn't) do that makes them perfect game fodder -- in fact we see five of our seven stats occur more frequently than not in perfect games. We also see two -- BB% and HR/FB -- that are essentially coin flips, so let's remove those from further consideration. We now have five stats that pass the eyeball test when it comes to possibly being culprits in perfect game infamy. Let's take those stats and apply them to 2013 data and see if there is something about the Astros -- or any other team for that matter -- that might make them more likely to be on the wrong end of history.

Let's create our own homespun perfect game susceptibility scoring system using the frequencies from the previous table and with the help of 2013 MLB averages (as of August 13), create a perfect game scorecard. For these five stats -- BABIP, K%, GB/FB%, IP%, 1stSw% -- we will assign points to each team who satisfy the aforementioned italicized criteria for a statistic, with our point scale coming from the third table frequencies:

Points Stat
0.83 K%, 1stSw%
0.67 BABIP, GB/FB, IP%

The maximum points possible in this measure of ineptitude is 3.67 (0.83 + 0.83 + 0.67 + 0.67 + 0.67) -- from this, we sum up each team's total, divide by 3.67, and multiply by 100 to get a percentage. Here's the verdict:

Team K% 1stSw IP% BABIP GB/FB Perfect Game Score
Percent
Brewers 0.83 0.83 0.67 0.67 3 81.7
Reds 0.83 0.67 0.67 0.67 2.84 77.4
Cardinals 0.83 0.83 0.67 2.33 63.5
Giants 0.83 0.83 0.67 2.33 63.5
Rockies 0.83 0.83 0.67 2.33 63.5
Athletics 0.83 0.67 0.67 2.17 59.1
Mets 0.83 0.67 0.67 2.17 59.1
Rangers 0.83 0.67 0.67 2.17 59.1
Nationals 0.83 0.67 0.67 2.17 59.1
Phillies 0.83 0.67 0.67 2.17 59.1
Astros 0.83 0.67 0.67 2.17 59.1
Padres 0.83 0.67 0.67 2.17 59.1
Tigers 0.83 0.83 1.66 45.2
Blue Jays 0.83 0.67 1.5 40.9
White Sox 0.83 0.67 1.5 40.9
Orioles 0.83 0.67 1.5 40.9
Yankees 0.83 0.67 1.5 40.9
Braves 0.83 0.67 1.5 40.9
Royals 0.83 0.67 1.5 40.9
Diamondbacks 0.83 0.67 1.5 40.9
Angels 0.83 0.67 1.5 40.9
Twins 0.67 0.67 1.34 36.5
Mariners 0.67 0.67 1.34 36.5
Pirates 0.67 0.67 1.34 36.5
Marlins 0.67 0.67 1.34 36.5
Rays 0.83 0.83 22.6
Dodgers 0.83 0.83 22.6
Red Sox 0.67 0.67 18.3
Indians 0.67 0.67 18.3
Cubs 0.67 0.67 18.3

Using perfect game results of the previous 25 years to set our criteria and some quick math, we see that with the stats selected, the Milwaukee Brewers are the team most susceptible to being perfect gamed. Beyond their top ranking, we are left with some surprising stats. There are a few teams at the top you wouldn't expect to see ranked so high in a measure of offensive impotence and some teams at the bottom of the list whose lineups don't conjure memories of the 1927 Yankees. What these results possibly infer is that the potential for a perfect game rests more squarely on the fprowess of the pitcher and his ability to command a game more so than the exacerbation of an offense's ineptitude.

For once, the Astros can honestly say, it's not us, it's Yu.

. . .

All statistics courtesy of Baseball Reference.

Stuart Wallace is a writer at Beyond The Box Score. You can follow him on Twitter at @TClippardsSpecs.

More from Beyond the Box Score:

Btbs-twitter-insert_medium

X
Log In Sign Up

forgot?
Log In Sign Up

Forgot password?

We'll email you a reset link.

If you signed up using a 3rd party account like Facebook or Twitter, please login with it instead.

Forgot password?

Try another email?

Almost done,

Join Beyond the Box Score

You must be a member of Beyond the Box Score to participate.

We have our own Community Guidelines at Beyond the Box Score. You should read them.

Join Beyond the Box Score

You must be a member of Beyond the Box Score to participate.

We have our own Community Guidelines at Beyond the Box Score. You should read them.

Spinner.vc97ec6e

Authenticating

Great!

Choose an available username to complete sign up.

In order to provide our users with a better overall experience, we ask for more information from Facebook when using it to login so that we can learn more about our audience and provide you with the best possible experience. We do not store specific user data and the sharing of it is not required to login with Facebook.

tracking_pixel_9351_tracker