There will be a very short planned maintenance outage of the site tonight (7/22) at 11 PM ET
November 9, 2006
"He was something like zero for twenty-one the first time I saw him. His first major league hit was a home run off me, and I'll never forgive myself. We might have gotten rid of Willie forever if I'd only struck him out." --Warren Spahn on his "first contact" with Willie Mays, who homered on May 28, 1951
In 1961, astronomer Frank Drake created an equation for estimating the number of intelligent civilizations in our galaxy. This equation, appropriately monikered (The Drake Equation), includes seven factors, ranging from the number of stars in the Milky Way to the fraction of the universe's life during which an average communicating civilization communicates with radio waves. By making assumptions to estimate each of these seven factors, astronomers have variously estimated that there are between 10,000 and 1,000,000 planets in the Milky Way alone housing civilizations that are able to create radio broadcasts (and therefore communicate using electromagnetic waves).
All of this helped lead to the creation of the SETI (Search for Extraterrestrial Intelligence) project, whereby scientists search the cosmos for these radio signals. Unlike the movie Contact however, a "first contact" with one of these civilizations has obviously yet to be realized. As long ago as 1950 physicist Enrico Fermi asked the question, "Where is everybody?"
The question is interesting, because if there really are tens or hundreds of thousands of planets in our own galaxy capable of making contact (not to mention the millions or billions that would be so in our light sphere), it would seem highly improbable that the skies are quiet. After all, it's probably not reasonable to assume that all such civilizations have only arisen recently and had the power to communicate as we have for 100 years. If indeed the SETI assumptions are valid, there must be thousands of worlds that have gone far beyond our technological capability and not either destroyed themselves or developed and obeyed "the prime directive" (in the lingo of Star Trek).
So we have an apparent contradiction in ideas. On the one hand, there's Drake's suggestion that there is a high probability for the existence of extraterrestrial civilization, and on the other, there's the lack of evidence or contact with those civilizations. This is known as the Fermi Paradox. While there are other explanations that one might consider, ranging from problems of distance, time, and technology, for my money, the fact that we aren't already aware of other civilizations indicates that we are very likely indeed alone in the cosmos.
I've been cogitating a bit on the subject of "first contact" after listening to an exchange between two baseball analysts and authors several weeks ago, culminating in this column on the nature of baseball's "first contacts".
Schwarz and Neyer
Although not quite the conundrum that the Fermi Paradox is, the question of whether the hitter or the pitcher has the upper hand when the two face each other for the first time on a major league field is a question open to debate. That debate was touched on in an ESPN Baseball Today podcast during the World Series.
For those not familiar, this season ESPN began offering the daily podcast, hosted by author Alan Schwarz. Each show runs 15 to 20 minutes, generally covers the day's news, and usually calls on one from among a few regular ESPN contributors for analysis and opinions. You know, guys like Steve Phillips, Jerry Crasnick, or Rob Neyer. I was late to the party and only discovered the podcast in the last month, and unfortunately they ceased at the end of October. Schwarz always kept it entertaining, and it was well worth the investment each day as I made my way to work, so here's to hoping that the ESPN honchos elect to bring it back next season.
Be that as it may, on this particular day Schwarz and Neyer were discussing one of the postseason series when Schwarz made the comment that he assumed a particular pitcher would have the advantage in his upcoming start, since most if not all of the opposing team's hitters had never seen him before. (Forgive me my paraphrasing him, but the podcast is no longer available, nor is there a transcript.) Neyer quickly chimed in with words to the effect that "if it were true, and I am not at all sure it is," before continuing on with his analysis of the matchup.
That exchange made me wonder whether Schwarz was correct in his assumption that pitchers have the upper hand in "first contact" situations, or if Neyer's skepticism was justified.
To find out, I devised a small study devoted to the question using play-by-play data from 2000 through 2005. To begin, with the help of the Lahman database I identified 437 position players and 562 pitchers who debuted from 2000 through 2004 whose data I could correlate with the play-by-play data. (The two data sets use different identifiers for players, and the Lahman database did not contain a cross reference for players who debuted in 2005.) The number that debuted each year can be seen in the graph below.
Using the play-by-play data, we can then calculate the performance of those players over the years 2000 through 2005. For the curious, walks here do not include intentional walks, and on base percentage only includes non-intentional walks, hits, and hit batsmen.
PA AVG OBP SLG OPS K/AB BB/PA HBP/PA Position Players 266056 .264 .326 .413 739 .192 .071 .010 Pitchers Against 362887 .267 .338 .430 768 .186 .085 .010 League 2000-2005 1126132 .265 .332 .425 757 .189 .080 .010
As you can see, the position players in our sample fared more poorly than did the overall population of hitters who faced our set of pitchers, although both performed more poorly than the league as a whole. The disparity between the position players and pitchers might indicate that it takes longer for pitchers to get established in the majors. It might also simply mean that, along with the larger number of pitchers, this is an indication that the talent level among pitchers is more variable.
Now that we have our samples in place, what we want to examine is how this particular set of hitters fared against this particular set of pitchers in their "first contact" situations versus subsequent times they squared off. Drum roll please…
PA AVG OBP SLG OPS K/AB BB/PA HBP/PA First Contact 30518 .258 .331 .405 736 .218 .084 .012 Subsequent PA 78666 .270 .333 .426 759 .179 .074 .010 Total 109184 .267 .332 .420 753 .190 .077 .011
In over 30,000 initial plate appearances, the hitters hit 12 points worse, slugged 21 points worse, but had an on-base percentage just two points worse than they would subsequently, accounting for a 23-point OPS deficit. It would appear that Schwarz's intuition was indeed correct.
Within the data, it's interesting that while hitters struck out 22% more often in first contact situations than they did in other plate appearances, they also walked 13% more often and were hit by pitches 17% more often. One theory that might account for these facts is that hitters are unsure of a pitcher's stuff in their first plate appearance against him. This might result in a less aggressive approach by the hitter, and therefore increase his tendency to take more pitches and get a read on his new foe, resulting in both more strikeouts and walks, as well as fewer balls put into play. Furthermore, having not seen a pitcher before, the hitter might be less able to react to pitches as quickly, and so be hit more often. It's noteworthy that in first contact plate appearances, the hitters saw an average of 3.85 pitches, while in subsequent plate appearances they saw an average of 3.71 pitches.
So, how big is that difference in overall productivity between first contact and subsequent plate appearances? Well, using the simple Runs Created Basic formula, the first contact batting line works out to 4.76 runs created per 27 outs while subsequent plate appearances are at 5.13, an increase of 7.8%. Apply that over the course of 162 games in the run environment of 2000-2005, and that's equivalent to approximately six wins.
Now, keep in mind that these numbers do not take into account the fact that some of these hitters may indeed have seen some of these pitchers in the minor leagues, since we're only looking at major league plate appearances. If some hitters had seen some of these pitchers before, one might assume that the difference would continue the trend, and indeed enlarge it. On the opposing side, a mitigating factor to consider is that perhaps a disproportionate number of the first contact situations occur early in a game, when a pitcher is fresh, and that might therefore suppress the opposition's batting line. In looking at the data, however, there is no indication that this is the case, as the average "first contact" plate appearance occurs at nearly the same position as that for all plate appearances.
Baseball, as it is often said, is the only game where the defense has the ball, where the player on defense initiates the action, and to which the offensive players must react. It appears from this study that pitchers do have the advantage the first time around, albeit not by a great deal. That sounds about right to me.