The Erroneous Streak

Errors might be the most flawed statistic in baseball. Wins and batting average don't paint nearly the complete picture they are given credit for, saves are arbitrarily defined, RBIs are greatly influenced by a hitters place in the line-up and a catcher's caught stealing percentage is badly skewed by the pitcher who is throwing to him. Still, I think errors tell you the least of any popular metric about what happens on the field.

Bill James famously broke down the difference between a good and bad defensive shortstop for John Dewan's Fielding Bible and his analysis purposefully removed errors from the equation. An error only measures when a player makes a mistake on a ball that they got to. It can't take into account the difference in range, which is the most important overall defensive factor. 

Your baseball career wouldn't last very long if you did this, but a shortstop could theoretically stand in the same place for the entire game and as long as they caught every ball within their reach, they would never be docked with an E6. 

It's great that the Yankees just set the Major League record for most consecutive games without an error. It means that everyone has performed their position somewhat competently over that stretch. At least that's what you would assume, right?

Just as one example, take Brett Gardner's play during the fifth inning of Sunday's game on Asdrubal Cabrera's line drive. It was a ball that he easily could have caught, had he read the ball correctly off the bat. Instead he misjudged it, took a step forward, then retreated and reached towards the ball as it sailed past him for a double. Ironically, had he been slightly closer to making the play, and it tipped off his glove, it would have been scored an error. While not an error in official baseball scoring, it still made the game recaps and highlight reels and was cast as an obvious mistake. 

Since defensive statistics are relatively new and difficult to quantify, we can't go back and see what the best 18 game streak of defensive play in the history of the MLB was. Dewan's plus/minus system has only been around since 2004. UZR only goes back to '04 as well. I did a pretty entensive search for a study looking into the best fielding teams of all times and came up empty aside from straight sorts based on fielding percentage. The nature of batted balls and the dimensions of parks have changed throughout the years, making it even more difficult to quantify. 

If you had a Retrosheet Database, it would be easier, and more to the point to find the lowest amount of runs allowed by a pitching staff over that same time period and adjust for era. During the last 18 games, the Yankees pitchers have allowed only 3 1/3 runs per outing, a feat which I find far more remarkable.

The Fallacy Of The Predetermined Outcome

I'm a little late to the party on this one, but Lar at wezen-ball (fellow resident of Shysterville) asked one of the most loaded sports-related questions one can ponder on Tuesday: Have Two Baseball Games Ever Played Out Identically?

In other words, are there any two games in baseball history that, if I were to pick up the scorecards for each, they would be indistinguishable?

The difference between Lar and pretty much everyone else in the world is that he actually set out to crack this mystery. He created a Retrosheet database and methodically broke down the potential possibilities resulting from each plate appearance and quickly came to the conclusion that the odds would of two games playing out exactly the same would be astronomical.

If you think about the numbers involved, it's not surprising in the least: with at least a dozen possible standard outcomes available for each plate appearance (and another dozen or more possible, but highly unlikely outcomes), and with 60 or 70 or even 80 plate appearances per game, the odds become fantastic that two games would be identical.

Furthermore, if it were possible to have each plate appearance broken down by pitch, it would become even more impossible to find two identical games. Click through to see how similar the pairs of games he isolates actually were. You might be surprised.

-----

Take into account that it is wildly unlikely that any two baseball games have ever played out in exactly the same way, and remember that the next time some tries to apply hindsight to something that almost (or supposedly should have) occurred in a sporting event of any kind. "Oh man, if the ump had called that obvious strike, then Pettitte gets out of that inning, those two runs don't score, and the Yanks would have won." Those are the same people who think "everything happens for a reason" and that everyone has a soulmate.

The title of this post comes from something that Michael Kay likes to say during Yankee broadcasts. I've had some fun with Mr. Kay 'round here, but I must agree with him on this one. There is NO chance that the game would have played out the same way and to believe so is to subscribe to the fairy tales of Destiny and/or Fate.

Lar went back over the entire history of recorded MLB game data and couldn't find anything close to two identical scoresheets. That database consisted of 100,000 games. When people try to isolate one event and keep the rest of the game the same, they are saying that not only is the imaginary divergent outcome of the game is the same as another that occurred in baseball history (essentially impossible), they are saying that it is the same as the actual outcome of THAT game (essentially impossible multiplied by 100,000+).

This is the kind of stuff that makes life endlessly fascinating. Even in a sport like baseball where things are endlessly categorized and recorded, it is almost impossible (no, fuck it, it IS impossible) to have two games which are the same, even on a scorecard.

That's why I hate the concept of "everything happens for a reason". Ironically the people who say this are usually talking about things beyond human control, which are actually the exact things that don't happen for a reason. There is plenty of stuff in life that happens without any reasonable explaination at all, and the best we can do is try to make sense of it logically.