Last night, Armando Galarraga was robbed of a perfect game with two outs in the ninth because of a missed call by first base umpire Jim Joyce.  To Joyce's credit, he admits that he blew the call and apologized to Galarraga.  But anyone who watches a replay of that ground ball knows that the runner was out at first and that Galarraga should have been credited with the 21st perfect game in the history of Major League Baseball.

So why isn't AG credited with a perfect game?  If an official scorer sees video replay postgame that refutes one of his decisions, he changes the scoring of the play.  Can't an umpire who changes his mind do the same given the special circumstances? (A perfect game has been amended before, as you will soon discover).

Whether or not June 2, 2010 is considered a perfect game in the official history books, it is considered a perfect game by Baseball Evolution.  Actually, if you look at the list of Baseball Evolution perfect games, you'll note several discrepancies from the official list.

Before we examine some of the other additions and subtractions from the official list of perfect games, there are a couple of final points that need to be made about the Joyce-Galarraga game.  First, there wouldn't even be controversy here if the Detroit Tigers employed a second baseman.  They do not, instead choosing to place the Statue of Carlos Guillen on the right side of the infield.  Miguel Cabrera had so little confidence in Guillen's ability to field what was really a routine grounder to second that he ranged far to his right to cut the ball off and make a throw with his momentum carrying him away from the play. 

Had any other second baseman been there for that play, Cabrera would have allowed him to make the routine play by charging the ball and making a strong throw, with Cabrera making a proper first baseman's stretch on the other end.  The play would not even have been close enough to have been blown.  If Jim Leyland wants to emphasize offense over defense and use Guillen as his starting second baseman, that's his prerogative.  But with a perfect game at stake, how can he justify not having brought in Adam Everett - quite possibly the best defensive shortstop of this generation - to play shortstop and shifting Ramon Santiago to second base after Detroit pulled ahead to a 3-0 lead?  That non-move is every bit as inexcusable as Joyce's call.

Second, maybe this is all for the best.  Having three perfect games thrown in a four-week period after only 18 had been thrown in the previous 130 seasons might have cheapened the accomplishment for this generation of baseball fans.  However, as we are about to see, those statistics are actually quite debatable.  Let's edit the official list of perfect games, shall we?

Potential Additions:

Milt Pappas, 9/2/1972

This is probably the instance of a lost perfecto most applicable to the Galarraga game.  Pappas had a perfect game through 26 at-bats and an 0-2 count on hitter #27.  He then proceeded to throw four straight balls to ruin the perfect game, but got batter #28 to pop out to second base, preserving the no-hitter.

Pappas was livid with umpire Bruce Froemming for those ball calls.  While Pappas' ire has been consistent, his story has changed over the years.  Sometimes all four balls are borderline and Pappas believes that Froemming should have given him the benefit of the doubt under the circumstances.  Sometimes two of the balls are borderline and the other two are definitely strikes, with Froemming intentionally messing with Pappas as a cruel joke. 

Basically, it is a case of Pappas shifting blame.  Just like a hitter has to protect the plate with two strikes to prevent an umpire from calling him out on a borderline pitch, a pitcher who is up 8-0 (with a win probability of 100%, according to Baseball-Reference.com) while pitching a perfecto has to protect that by not nibbling at the plate and not throwing three straight sliders to Larry Stahl, a pinch-hitter who was batting .232 at the time and finished his career with the very same batting average mark.

This, unlike Galarraga's effort, is most certainly not a perfect game.

Harvey Haddix, 5/26/1959

In one of the most famous games of all-time, Harvey Haddix pitched 12 perfect frames before an error allowed a runner to reach in the 13th.  A subsequent sacrifice and intentional walk brought Joe Adcock to the plate with runners at first and second, one out, and the no-hitter still in tact.  Adcock blasted a pitch over the right-center fence, but was called out at second for passing Hank Aaron on his trip around the bases, which changed the scoring of the play from a homer to a double.   Either way, the unearned run scored, erasing the no-hitter and shutout, plus saddling Haddix with a loss.

This is most assuredly a perfect game, though.  Of the 20 official perfect games, how many would have remained perfect if the pitcher had to continue for three or four more innings past the ninth?  Few, if any.  Haddix' performance here is certainly not less than that of the other perfect game authors, and is arguably greater.

Ernie Shore, 6/23/1917

How can a reliever pitch a perfect game?  In this instance, Babe Ruth started the game, walked the first batter, then was promptly ejected for arguing balls and strikes.  After the ejection, Ruth actually punched the home plate umpire in the head and received a suspension and a fine for his barbaric act.

But this was one game in which Babe Ruth was the B story rather than the A story.  The runner Ruth had walked was caught stealing shortly after Ernie Shore came on in relief.  Shore retired each of the 26 batters he faced and was on the mound for all 27 recorded outs.  Shore was recognized as having pitched the fifth perfect game in baseball history, third of the modern era.

It wasn't until years later that a commissioner-appointed committee ruled that it was not a perfect game because of the walk, merely a combined no-hitter.  Baseball Evolution agrees with the ruling.  In every other perfect game, the perfect pitcher needed to face 27 batters; Shore only faced 26.  It might be more of a grey area had Shore picked the runner off himself rather than have the runner caught stealing by the catcher.  In some ways, Shore's accomplishment is more impressive than other perfect games in that he did not have time to prepare himself to pitch nine innings that day, but a perfect game, it was not.

But this instance gives some hope to Mr. Galarraga.  If a committee can turn a perfect game into a non-perfect game, why can't it do the converse?

Kerry Wood, 5/6/1998

Most perfect games included at least one spectacular play on defense.  Kerry Wood's 20-strikeout game in his fifth career start was not a perfect game because of one spectacularly bad play on defense.  Leading off the third inning, a weakly tapped grounder to the left of third baseman Kevin Orie kicked off his glove, but was ruled a hit.  The runner advanced to third on a sacrifice and a walk, but remained stranded there.

This play was clearly an error by Orie.  Even if the official scorer didn't get a good look at it live, he could have easily changed it to an error soon after the fact.  No one knew in the third inning what Wood was about to accomplish.  He could also have changed the ruling after the game, although in doing so it would look like favoritism from the home scorer to give Woody the no-hitter.

Even with the poor scoring decision, a runner reached third base and Wood faced 28 hitters, so this was not a perfect game.  It was a no-hitter and the most dominant pitching performance in baseball history, but it was not a perfect game.

Subtractions:

Lee Richmond and Monte Ward, June of 1880

In 1880, pitchers threw underhanded and batters were able to request whether they wanted a high or low pitch.  Pitchers threw from a "box" rather than a mound, and this box was only 45 feet from home plate, which was a 12-inch square.  It took eight balls to walk a batter and foul balls did not count as strikes.   

Basically, the game called Base Ball in 1880 was nearly as different as the Baseball we know today as American Football is from European Football (or soccer).  How many pitchers would have thrown perfect games in the past 100 years or so if they could throw seven balls in an at-bat and not have that count as a walk?  Who knows.  What is clear is that whatever Richmond and Ward accomplished in one week in 1880 isn't anything like the feat Cy Young performed 14 years later under modern rules.  These were not perfect games - at least, not how we know them to be now.

Don Larsen, October of 1956

Trevor Hoffman needs five saves to get to 600 for his career.  Roger Maris broke Babe Ruth's single-season home run record in 1961.  Greg Maddux finished his career with one more win than did Roger Clemens  We hold these to be truths.

Why then, is Don Larsen's World Series game in which he did not allow anyone to reach base mentioned along with all of the other regular season perfect games?  If postseason stats count, then Clemens and Maddux are tied in career wins, Maris only tied Ruth's output of 62 home runs from 1927, and Hoffman is one elusive save away from 600.  Mariano Rivera is nipping at his heels, as the whopping 39 saves he has recorded in the postseason give him a total of 576 for his career.

Perhaps posteason stats should count in the record books among regular season and career totals, but that is a debate for another day.  The fact is that they do not.  Larsen's performance in the 1956 World Series may have been more important and more impressive than any other perfect game in history, but it does not belong with the other perfect games in the record books any more than Ruth's two World Series homers in 1927 gave him 62 dingers for the season.

So there you have it.  Richmond, Ward, and Larsen are out.  Haddix and Galarraga are in.  That gives us 19 perfect games in 110 years of Major League Baseball under modern rules.  Three of those perfectos have come in the past 24 days and four have occurred in the past 12 months.  Those are the facts.  Baseball Evolution has spoken.