Which Recent Perfect Game Was The Hardest?

Yesterday, Felix Hernandez went 27-up, 27-down against Tampa Bay, becoming just the 23rd pitcher in baseball history to throw a perfect game. Amazingly, this was the sixth perfect game in the last four years, as Felix joins the company of Mark Buehrle, Dallas Braden, Roy Halladay, Philip Humber, and Matt Cain as the newest members of the club.

Back when Cain threw his perfect game in June — striking out 14 games in the process — I looked at where that game ranked in history, and noted that it was in the conversation with Kerry Wood‘s 20 strikeout performance and no-hitters from Nolan Ryan and Sandy Koufax as one of the best games in history. While Game Score does a pretty good job of scaling relative performances, however, it doesn’t evaluate for context — park, league, opponent, etc… So, that’s what I set out to do today.

It’s simply more work than I have time for to go through every perfect game in history to evaluate the quality of the opponent, the park the game was played in, and the run environment of the day, but with the help of Jeff Zimmerman, I was able to look at the most recent six. Thanks to the fantastic custom leaderboard feature here on FanGraphs, it is not that difficult to compile the statistics of a specific group of players, like for instance the batters who were matched up against a guy who threw a perfect game. From there, we can look at the relative strength of the batters that each pitcher had to face.

For the recent six, here are the opponents they faced on their road to perfection – you can click through the links to see the custom leaderboards with each player’s performance from that year.

7/23/09: Buehrle vs 2009 Rays.

Buehrle got the Rays at home on a Thursday afternoon getaway game, but Tampa Bay still ran out most of their regulars, just subbing in the backup catcher in a day-game-after-night situation. This was a good year for Tampa’s offense too, as the nine guys in the line-up that day combined to hit .270/.356/.465 in 2009, good for a .359 wOBA. Each member of the Crawford/Zobrist/Pena/Longoria/Bartlett group posted a wOBA over .367 that year, and Crawford was the only one who wasn’t particularly good agains left-handed pitchers. Making up Crawford’s weakness against LHBs was the presence of Gabe Kapler, who posted a .394 wOBA against LHPs that year. This was simply a staggeringly good line-up to get completely shut down.

5/9/10: Braden vs 2010 Rays.

Hey, look, it’s mostly the same group of guys that Buehrle faced. Michael Hernandez got swapped out for Dioner Navarro behind the plate and Willy Aybar was DH’ing instead of Pat Burrell, but those guys were just as ineffective as the previous incarnation. This group was a bit worse, though, as Bartlett went from having an insane year back to being a weak-bat shortstop, and Zobrist’s offense regressed a lot in 2010. This group of nine combined for a .333 wOBA in 2010, so while they were still above average offensively, they weren’t hitting like they did the year before, especially against left-handers. Also worth noting – this was a Saturday afternoon game, so again, day-game-after-night situation.

5/29/10: Halladay vs 2010 Marlins.

The Marlins threw 12 different batters at Halladay, pinch-hitting for the entire bottom third of their order in the bottom of the 9th inning. So, here, it’s not quite as simple as averaging out the entire season lines for the starting nine, since there weren’t equal opportunity, but we can still evaluate the strength of that Marlins offense without too much trouble. And, to be frank, it wasn’t great. Dan Uggla and Hanley Ramirez were having strong seasons, but both were also right-handed batters, which meant that Halladay had the platoon advantage against them. Every other batter Halladay faced that day was a below average hitter vs RHPs, ranging from okay bats like Gaby Sanchez and Chris Coghlan to automatic outs like Josh Johnson and Mike Lamb. As a group, Halladay’s opponents posted a .312 wOBA against right-handers in 2010, and while weighting the line by number of at-bats given to each player shifts that up a bit, this was still a below average offense that Halladay was facing. Unlike the two against Tampa Bay before him, though, this was a night game.

4/21/12: Humber vs 2012 Mariners.

If you were to pick a team that you’d expect to get perfecto’d over the last few years, it’d probably be the Mariners. Their offense is better this year than it was the last two, but it’s still pretty bad, with John Jaso representing the team’s only above average hitter this season, and he didn’t even start against Humber. Another afternoon game, the Mariners were also using backup shortstop — and completely useless hitter — Munenori Kawasaki at shortstop, and this was the part of the season where they were still trying to extract value from Chone Figgins, so he was hitting leadoff and playing left field. The whole group of hitters Humber faced have combined for a .286 wOBA this year, and that doesn’t improve at all against right-handers. Of the guys who started for the Mariners that day, Kyle Seager‘s .321 wOBA vs RHPs was the best in the line-up, and Ichiro Suzuki was the only other batter in the line-up to clear the .300 mark. This line-up deserved what they got.

6/13/12: Cain vs 2012 Astros.

It’s easy to dismiss Cain’s performance since it came against the Astros, who are clearly baseball’s worst team this season, but that’s not totally fair to his accomplishment. They did run out a couple of solid hitters that night, including Jed Lowrie and Jose Altuve, and the overall group of batters Cain faced have posted a .310 wOBA this year. That’s not great, but they weren’t all automatic outs either, and they actually did slightly better than that against right-handed pitching this season. It was also a night game, so Cain wasn’t facing too many back-ups — beyond the joke about everyone on Houston being a back-up on a good club, anyway — and J.A. Happ‘s terrible performance meant he only got to hit once. So, while the quality of opponent here wasn’t great, this wasn’t the total cakewalk you might think when you heard someone beat up on the 2012 Astros.

8/15/12: Hernandez vs 2012 Rays.

Hey, it’s the Rays again. They show up here for the third time, but this is a pretty different group than the first two. Upton, Zobrist, and Longoria are still around, as is the ghost of Carlos Pena, but everything else has been turned over. It was another day-game-after-night situation, so Felix got to face the likes of Sam Fuld and Elliot Johnson, but Joyce and Longoria were still in the line-up and Jeff Keppinger and Desmond Jennings both pinch-hit in the ninth inning, so this wasn’t a total scrub best. Still, the group’s .319 wOBA is inflated a bit by the single plate appearances from Keppinger and Jennings, so the first two times through the order, the Rays offense was a bit worse than that number would suggest. As a righty getting to roll through the Lobaton/Johnson/Rodriguez/Fuld quartet, nearly half of the line-up the Rays rolled out there didn’t present much of a challenge.

So, just adjusting for quality of competition, Buehrle seems to stand out as the most impressive perfect game of the bunch. He faced a legitimately good offense at full strength with no platoon issues against left-handed pitchers. In fact, if we adjust for ballpark effects, Buehrle again comes out on top, as he threw his in the hitter’s haven of US Celluar Field, while every other perfect game in the last four years has been thrown in a pretty extreme pitcher’s park.

The hardest opponent and hardest environment awards go to Buehrle, with Braden probably coming in second, and then Halladay/Felix/Cain all facing similar-ish challenges in terms of park and quality of batters faced. Humber clearly had the easiest path to perfection, facing a brutally awful line-up in a park that is just destroying offense this year, primarily by suppressing hits on balls in play.

Of course, Felix, Cain, and Halladay all racked up double digit strikeouts, doing more of the work themselves and relying less on balls getting hit right at their defenders, and going by sheer dominance, they’re the ones throwing unhittable pitches. Buehrle got 21 outs on balls in play using his 85 mph fastball and diving change-up, which just doesn’t look as overpowering as the breaking balls that Felix was busting off yesterday. But, when you consider who he was pitching against, and where he was pitching, Buehrle’s perfect game is probably the most impressive of the whole bunch.

Print This Post

Dave is the Managing Editor of FanGraphs.

57 Responses to “Which Recent Perfect Game Was The Hardest?”

You can follow any responses to this entry through the RSS 2.0 feed.
  1. Uh Oh Cordero says:

    IMO, Humber’s is the hardest just because he’s Phil Humber.

    +112 Vote -1 Vote +1

  2. Dr. Internet says:

    The ones that aren’t against the Rays.

    Vote -1 Vote +1

  3. Person says:

    Perhaps we should also take into account that Buehrle needed possibly the greatest defensive play in perfect game history to pull of his feat. I still think he had the most impressive one for the other things mentioned.

    Vote -1 Vote +1

  4. John says:

    Agree with commenter who said Humber’s was the hardest. He had to overcome being awful himself.

    Vote -1 Vote +1

  5. Baltar says:

    Nothing against your post, Dave, which was very good, but I have a pet peeve with perfecto’s (as well as no-hitters and shutouts) being credited entirely to the pitcher. To me, it is analagous to crediting the win, loss or save solely to the pitcher.
    There were 9 defensive players on the field at all times when those perfectos occurred. Granted, the pitcher deserves most of the credit, but the catcher especially and the 7 other guys deserve considerable credit, even a guy who may not even have touched the ball (if he hadn’t been there, there would have been a gaping hole that could have been exploited).
    The Seattle Mariners played a perfect defensive game against the Tampa Bay Rays yesterday afternoon. Congratulations to the team!

    Vote -1 Vote +1

    • tim says:

      I agree with you completely. Thats why there’s always 1 or 2 amazing TOP 10 defensive plays that happens during the game in which is shown over and over and over again. Like in the Cain game with the RF coming out of nowhere with the flying catch. Or with the Mets No-No when the LF ran into the wall busting his sholder and being out for the rest of the year. Those guys will always be remembered. (there wasn’t any of those with King Felix)

      Vote -1 Vote +1

    • brendan says:

      this made me think of the perfect game that jonathan sanchez _didnt_ get because of an error by juan uribe. not 100% pitching, indeed.

      Vote -1 Vote +1

  6. Well-Beered Englishman says:

    Because of his weaker stuff, plus the park and opponent, Buerhle’s perfect game is definitely the most impressive of the whole bunch in terms of “wow, can’t believe he did that.” Maybe there’s a bit of luck in bringing so many groundball outs, but also a bit of sheer ballsy awesomeness. “Hey, Felix, I’m nowhere near as effective a pitcher as you are, but I can still throw a perfect game.”

    Vote -1 Vote +1

    • Steve says:

      Didn’t Buehle have a no hitter (or 2?) before that?

      The most shocking was obviously Braden and then Galarraga’s**

      Vote -1 Vote +1

      • macseries says:

        yep, buehrle no-hit the rangers on 4/18/07 and faced the minimum. he walked sammy sosa but picked him off at first.

        Vote -1 Vote +1

  7. Evan says:

    Already we’ve forgotten about Galarraga’s perfect game. At a glance the Indians had a team wOBA of .312 for 2010 though, so I doubt could be considered more difficult.

    +6 Vote -1 Vote +1

    • Zonis says:

      I agree. That was a perfect game, despite what the record books say. It was that completely blown call.

      Vote -1 Vote +1

  8. nitro2831 says:

    I don’t think wOBA is the proper tool here, in terms of perfect game difficulty. It would be more meaningful to look just at OBP, because avoiding baserunners is all you need to do to get a perfecto.

    +19 Vote -1 Vote +1

    • Marc says:

      I agree. Furthermore, you shouldn’t calculate lineup OBP by combining statistics for the year. You should estimate true talent OBP for each player against pitchers of the same handedness as the perfecto pitcher then weight each player in the line-up equally. Technically, you should calculate the geometric mean of (1-OBP) over all at-bats.

      Vote -1 Vote +1

    • Dave Cameron says:

      The problem is that pitchers don’t set out trying to throw a perfect game, nor do they approach hitters simply with the goal of not letting them get on base. If we just evaluated them by OBP, you’d end up significantly underestimating the difficulty of a pitcher facing a guy like Alfonso Soriano, who might not get on base a lot but is certainly not a guy you can just throw fastballs down the middle to without fear.

      There’s a difference between the challenge for a pitcher to face a bunch of.290 OBP guys with no power and a line-up full of aggressive sluggers.

      Vote -1 Vote +1

      • TKDC says:

        Do you think wOBA is better than OPS as a general tool for comparing perfect games? If you had to pick one.

        Vote -1 Vote +1

  9. J6takish says:

    In regards to the Galarraga game, the lineup the Indians had that day had 322 wOBA against righties. When you factor in the general awfulness of Armando Galarraga (338 wOBA against lefties, 328 against righties that year) with the fact that he had the corpses of Carlos guillen, Johnny Damon and Magglio behind him. He got through that game in something like 88 pitches. He threw a ridiculously efficient perfect game against a decent righty hitting team with awful defenders playing behind him. I think it’s very impressive, especially when you consider that he really wasn’t a good pitcher at all

    +11 Vote -1 Vote +1

  10. ponderous says:

    What’s really amazing about Buehrle is how close he came to having 2 perfect games. His first no-hitter was against a not terrible Rangers lineup at the Cell….

    Vote -1 Vote +1

    • macseries says:

      i recently had an argument with a friend about whether there could be any other “the catch” than willie mays. obviously as a sox fan and a person who recognizes the malleability of language (and who also knows dwight clark’s catch as “the catch”), i came down on the side of there being more than one “the catch.” it nearly came to internet-blows.

      another thing: with d. wise’s return to the white sox, the grounds crew has returned the epigram “the catch” above billy pierce in left center. and hawk and wimpy talked about the catch every time wise came up to bat in toronto.

      Vote -1 Vote +1

  11. byron says:

    I’m not particularly impressed with a pitching performance that lets 21/27 batters put the ball in play, whatever the defense manages to do.

    -16 Vote -1 Vote +1

  12. Kevin R says:

    But Felix had to rely on the Mariners scoring a run (cf. Pedro), which seems like the hardest feat of all.

    +22 Vote -1 Vote +1

  13. The Ancient Mariner says:

    Not as thorough, but for broader context, Adam Darowski’s post on BtBS today is also of interest: http://www.beyondtheboxscore.com/2012/8/16/3246221/the-best-lineups-to-be-perfect-gamed

    Vote -1 Vote +1

  14. Mike Atwater Witt’s perfect game in Arlington was the best, 10ks, 4 fly balls, 1-0 victory against Charlie Hough, and Texas pinch hit for the last 2 batters (jerks).

    I realize that I went beyond “recent” and that Witt’s numbers and the Game Score may be objectively inferior, but what’s the point of the Internet if you can’t just make shit up?

    Vote -1 Vote +1

  15. J says:

    Why all the day games (4 out of 6)?
    Is it just the result of giving regulars the day off?

    Vote -1 Vote +1

  16. Rob says:

    Since you brought up Kerry Wood’s 20 K game, this was the opposing lineup by wOBA:

    Craig Biggio – .406
    Derek Bell – .373
    Jeff Bagwell – .422
    Jack Howell – .372
    Moises Alou – .416
    Dave Clark – .252
    Ricky Gutierrez – .301
    Brad Ausmus – .318
    Shane Reynolds – .180
    (PH) Bill Spiers – .341

    Kinda makes Wood’s game a little more impressive, no?

    Vote -1 Vote +1

  17. De Selby says:

    I wonder who had the worst defense playing behind him. Without knowing anything about the others I’m going to go ahead and guess it was Buehrle.. Take a look at some of the guys playing behind him. Carlos Quentin in left. Podsednik in center (!). (a very creaky) Jermaine Dye in right. Jayson Nix at second. JOSH FIELDS MAKING HIS 5TH CAREER START AT FIRST BASE.

    I’m not sure which is more unlikely, that there were no errors (or “misplays”) made or that Buehrle didn’t give up any hits.

    Vote -1 Vote +1

    • Dave says:

      Actually Dewayne Wise subbed in for quentin later in the game and switched with podsednik in center, considering he robbed a home run later in the game to preserve not only the perfect game but the shutout nonetheless i believe that should exonerate any questions about the defense behind him that game

      Vote -1 Vote +1

      • De Selby says:

        I am aware that Dewayne Wise played center field in the top of the ninth inning. I don’t understand how that affects the fact that the quality of the defenders behind Buehrle in the game was sub par.

        Vote -1 Vote +1

      • macseries says:

        only in the outfield. gordon beckham is a fine 2B, ditto alexei ramirez at short.

        Vote -1 Vote +1

  18. DD says:

    Dave, why not use a stat like game score and combine that with wOBA to get the nature of dominance the pitcher showed against their opponent. Buehrle’s GS was a 93, and Halladay’s was a 98, the later may be closer to being the better outing.

    Vote -1 Vote +1

    • craigws says:

      because the article wasn’t about dominance, but it was about which line-up was the most difficult to beat?

      Vote -1 Vote +1

  19. coltholt says:

    I’m sure that this could be said about any number of similar performances, but I would love to see a similar analysis of the near perfect game by Travis Wood back in 2010. Wood’s line (although not a complete game with the Reds losing in 11…can’t really fault him for the lack of offense while facing Halladay) was 9 innings with 1 hit (only baserunner) and 8 k’s against the Phillies lineup featuring Howard, Werth, Rollins, and Victorino in a hitter friendly park at Citizens Bank.


    Vote -1 Vote +1

  20. Philip Christy says:

    Dallas Braden’s perfect game was on a Sunday, and was a day game after a day game.

    Vote -1 Vote +1

  21. themiddle54 says:

    Buehrle only had 6 K. His D had to convert 21 outs. Regardless of opposition faced, we have to allow for BABIP/defensive efficiency I think, and consider a perfect game a team effort with typically at least 60% of the outs coming with some input from the defense, who must also be perfect.

    What’s great about Humber is that he had a K per inning AND only needed 96 pitches–he was showing good stuff and working efficiently that day. His is more impressive than Buehrle’s, IMO, for the added K and doing it in 20 fewer pitches.

    For the same reason, I consider the best perfecto in my lifetime to be David Cone. 10K and only 88 pitches. That’s marvelously efficient yet overpowering. He averaged barely over 3 pitches/batter faced and still whiffed 10. Regardless of park or opposition strength, it’s unlikely we see that combo in a perfect game again.

    Vote -1 Vote +1

  22. Don Draper says:

    When does season 6 of mad men premier?

    Vote -1 Vote +1

  23. Antonio Bananas says:

    As streaky as baseball is, maybe look at the 10 games before and 10 games after of the hitters. If a guy is in a slump and needs a mechanical tweak, he’s easier to pitch against. Then again, that could be splitting hairs that don’t need to be split.

    Vote -1 Vote +1

  24. Sandy Kazmir says:

    It would have been a lot more work to look back at the other games, but my boy Whelk had this tremendous look at the most recent perfecto:

    Whelk’s platoon split regressor is just a fantastic tool that everybody should be using. An important step that he did not take was to adjust for park, which I think should be included, but looking strictly at handedness splits he was able to derive the probability of Felix throwing that perfect game at around .00420%. I’d highly recommend it to everyone with three minutes to kill.

    Vote -1 Vote +1

  25. Don Draper says:

    Season 6 of mad men anybody?

    Vote -1 Vote +1

  26. Jaybo Shaw says:

    Good Article Dave,

    Many people eluded to this, but it would be interesting to try to include aggregate defense (factors according to the number of balls in play) and see how that changes things.

    Although, just trying to determine the best way to do that just made my head spin. (Allowing more balls in play would make it more “challenging” to get a perfect game because the chances of it becoming an non-out are obviously much higher), but of course, if this exercise were to reward pitchers that were objectively inferior (had fewer Ks), then it would probably cease to have meaning.

    This was pointless.

    Vote -1 Vote +1

  27. Matt M says:

    Interestingly, Buehrle’s perfect game wasn’t the first time he faced the minimum 27. In ’04, he gave up a lead-off single against the Indians and got the next batter to ground into a double play. He was perfect the rest of the game.

    Vote -1 Vote +1