Thursday, July 24, 2008

Robbed By the Defense: Almost Perfect Games

Tonight I was intently following (via Gameday) the start of one Ian Patrick Kennedy, 23-year old prospect for the New York Yankees. He finished the sixth inning with seven strikeouts, no walks, and no hits. It was the second game of a double-header, so the most Kennedy (or IPK as some like to call him) could expect was a seven-inning perfect game. Well... except for the fact that his teammates made four errors behind him. This fielding futility included two in the seventh inning; second baseman Chris Basak booted a grounder with two outs that would have ended Kennedy's short no-hitter. The next hitter hit one over the head of right fielder Greg Porter, whose botched catch allowed the tying run to score.

All of this got me thinking about no-hitters and perfect games. Now, for anyone who has tried to explain the concepts of "wins" and "errors" to non-baseball fans knows, these concepts are pretty arbitrary and don't really tell us anything about a pitcher's ability. There are still old fashioned folks who declare they'd rather have the pitching stats of a 2008 Livan Hernandez (10-6, 5.29 ERA) than of a 2008 Johan Santana (8-7, 3.05 ERA) because the former has more wins. Others will frequently disguise this stupidity as "he can't pitch to the score."

Similarly odd, yet less talked about, is the concept of the Earned Run. Like pitcher wins and losses, it's an invented statistic attempting to take the blame of a run away from a pitcher whose defense has failed him. Taking the defense out of the statistic should theoretically lead to a stat that more closely belies the true ability of the pitcher. However, recent stat analysis shows us the opposite; a pitcher's stats are more correlated from one season to the next if the "unearned" runs are included. (However, it is with taking the other extreme and considering only the outcomes that are independent of fielding- home runs, walks, and strikeouts- that are most predictive of pitcher ability).

Now, let's return to the concept of the "perfect game," as defined- all batters in a game of at least nine innings, retired without reaching base. The essence of the rule regarding official perfect games is constructed to make it a team effort, and I do not disagree with that. There is something that makes twenty-seven up, twenty-seven down special, and it includes the defense (and the offense scoring at least one run). However, there is also something special about a pitcher going nine innings without allowing a base-runner of his own accord. I was inspired by IPK's performance to investigate whether any pitchers have been "robbed" of a perfect game by their defense. Here's the play index search: 9 innings pitched, 0 hits, 0 walks, and more than one baserunner.

The search returned eight games, since 1956, in which the pitcher went 9 innings and had no hits, no walks, but did have some baserunners. Of these, five pitchers allowed a baserunner by hitting batsmen; these I believe to be against the spirit of the search and shall be excluded. Starting with the earliest:

Dick Bosman- July 19, 1974

Dick Bosman was a mediocre pitcher (1974: 88 ERA+) on a mediocre Indians team (77-85), who was long displaced from his best seasons with the Washington Senators. He did have good control, walking just 2.74 men per 9 innings over the course of his career. In his previous start in Minnesota, Bosman took a no-decision after 6.1 uninspiring innings, allowing 4 runs (3 earned) on 4 hits. On July 19, Bosman took the hill against the Athletics and retired the first eleven men in order. After allowing a man to reach on an error; Bosman retired the final sixteen men in order. His final line: 9 IP, 0 H, 0 BB, 0 R, 4 K. Should this game be included, in spirit, with the seventeen official perfect games?

I say no. The reason lies, specifically, in the reason for the error. You see, with two outs in the fourth inning, A's third baseman Sal Brando hit a ball right back to Bosman, and our potential hero threw the ball into the seats. The lone base-runner of the game was still Bosman's fault. So Bosman should still be in the league with mere mortals throwing mortal no-hitters. According to Wikipedia, Bosman is noteworthy for rebuilding old cars and selling them, including a 1933 Ford to actual perfect game thrower David Wells. The Baseball Almanac notes that Bosman is the only pitcher in recorded history to have ruined his own perfect game with a throwing error.

Jerry Reuss- June 27, 1980

Jerry Reuss was a tad more successful pitcher; over 22 seasons Reuss pitched in 628 games (527 starts) and a dead-average career ERA+ of 100. In 1980, Reuss had one of his best seasons, crafting a 2.51 ERA (141 ERA+) and winning 18 games. He would later finish second in the Cy Young voting to Steve Carlton. In the games leading up to June 27, Reuss was nothing short of spectacular. On June 21, he threw 9 shutout innings against the Mets; on June 16 he went ten innings against the Phillies but still got a no-decision. Amazingly, he already had three shutouts before taking the mound in San Fransisco, eager to top his already brilliant season.

This time, it wasn't long before the lone baserunner reached; it was in the first inning, when a ground ball of the bat of Jack Clark ate up third baseman Bill "No, A Different One" Russell. Reuss then retired the next twenty-five batters in order, winning the game for the Dodgers 8-0. Reuss had twice previously been one batter short of a no-hitter. Mike Robbins, who wrote a book about situations such as these, called "Ninety Feet from Fame: Close Calls with Immortaility" wrote that in 1972, after one of these near no-hitters, Reuss said "If I ever get to the ninth-inning again with a no-hitter, I'm going to get it." So he did. You were perfect in my eyes, Jerry.

Terry Mulholland, August 15, 1990

Terry Mullholland was much more like Bosman than Reuss; though his longevity (20 years) was more like the latter, his performance (93 ERA+, 124-142 record) was more like the former. Still, Mulholland managed to piece together some good seasons, and 1990 was one of them, with a 3.34 ERA (114 ERA+) in 26 starts for the Phillies. In the dog days of the summer, Mulholland had been struggling, allowing four or more runs and ten or more hits in each of his previous four starts. On August 15, things turned around.

While the Lenny Dykstra led Phillies staking him to a lead, Mulholland retired the first eighteen Giants in order, striking out nine. But textbook perfection was not meant to be, as San Fran centerfielder Rick Parker knocked one towards the hole between short and third. With the speedy Parker racing down the line, third baseman Charlie Hayes grabbed the ball and threw wide of John Kruk's considerably large target at first base. It was ruled an error, and Parker was eraced with a double play in the next at bat; Mulholland would retire the final 9 batters in order. According to Robbins, Kruk lobbied to have the error charged to him, but the official scorer did not relent; Hayes also apparently preserved the no-hitter by making a diving catch to end the game.

Said Mulholland after the game: "I'm not Nolan Ryan. I'm not knocking on the door for the Cy Young Award. I'm just Terry Mulholland." And yet Mulholland came closer than ever Ryan did to throwing a perfect game. He is the only man to ever have thrown a no-hitter while facing just twenty-seven batters.

Wikipedia has a good section discussing other, pre-Retrosheet Era games, including a classic pitcher's duel between Christy Mathewson (who threw a no-hit, no-walk game but was failed by two errors) and Mordecai "Three Finger" Brown, who also took a no-no into the ninth, but lost the game. In addition, there are other amazing pitching performances not counted as perfect games. For example, Babe Ruth was famously thrown out of a game in 1917 after walking the game's first batter. Ernie Shore relieved him, and after the runner was caught stealing retired the next twenty-six batters. It was once recognized as a perfect game, but does not fit under the current definition (it is still an official combined no-hitter).

Pedro Martinez pitched one of the best games in recent baseball history in 1995 when he went nine perfect innings, only his Expos teammates had failed to score a run. Martinez came out for the tenth but gave up a leadoff double and was taken out for a reliever.

To close, I give you the most impressive non-perfecto of them all: the twelve perfect innings by Harvey Haddix on May 26, 1959. He retired thirty-six consecutive hitters before an error allowed the 37th batter to reach base. After a sacrifice and an inentional walk (to Hank Aaron), Joe Adcock hit a home run to win the game- the homer was changed to a double because Hank Aaron simply walked off the field, and Adcock passed him on the bases. The Pirates had failed to score any runs for Haddix, despite 12 hits.

These seemingly-freak occurences should be teaching us a lesson about blame and credit and how they are divided among the men who play baseball. Immortality is rightfully given to the men who throw perfect games, but no one ever remembers the defenders who helped them do it, or the hitters that ensured the game only lasted 9 innings. Yet, in a tragedy worthy of the Bard, immortality shuns those pitchers who dare to do all they can to acheive perfection but are robbed by imperfection off the mound. To them, the Tantaluses of baseball, my glass is raised.

UPDATE: As if on cue, posts this song about the "fateful game" about Harvey Haddix, pretty awesome. Why don't we add old Harvey to that list, indeed.


Anonymous said...

I'll disagree with you on Dick Bosman. The pitcher-as-fielder is considered separately from the pitcher-as-pitcher, and that's the way it should be. You're judging his pitching performance, and his fielding isn't relevant.

Think of it this way: if a batter reaches on a pitcher's error, would you count that in the pitcher's DIPS? Of course not, because it had nothing to do with his pitching.

Once the ball is hit to him, he's no longer a pitcher. He's just another infielder. That infielder's bad play shouldn't affect any evaluation of the pitcher's performance.

Matt said...

That's a fair point, and the way I presented all this, in terms of "don't blame the defense," Bosman's defense shouldn't negate his pitching.

Still, there's something not right to me about awarding "perfection" to a man who was not perfect that day-- in the end, it was Bosman's fault that there was a baserunner. But, as The Baseball Project sing in the Harvey Haddix song, all perfection is flawed.

Of course, if Haddix really wanted to have a perfect game, he wouldn't have ground into a double play... ;-)