“People ask me what I do in winter when there’s no baseball. I’ll tell you what I do. I stare out the window and wait for spring.”
– Rogers Hornsby
One nice thing about the offseason is that it gives the obsessive baseball fan a chance to reflect on the season past and consider questions that were neglected during the heat of the pennant race. While some of these questions may not rise to the level of PrOPS, DIPS, UZR and other acronyms for which sabermetricians are famous, the credo of the baseball fan is to leave no stone unturned where numbers are involved. And so it is in that spirit that I offer this week’s analysis.
I’m sure many of you remember the controversy engendered last August when Larry Krueger, a radio talk show host on the Giants flagship station KBNR in San Francisco, went on a rant (as radio talk show hosts will do) that included criticism of the club and its “brain-dead Caribbean hitters hacking at slop nightly.”
What followed was simply this: Giants manager Felipe Alou and his players criticized the remarks. Alou refuses to do his pre-game interview on the station. Then lots of people call for Krueger’s dismissal. Krueger is then suspended for a week and apologizes. Alou does not accept the apology, saying that “there’s no way to apologize for such a sin.” And finally Krueger, his producer and a program manager are fired after they replayed parts of an Alou interview from ESPN a few days later in which Alou called Krueger “a messenger of Satan.” It was a busy week.
At the time there were several moderate defenses of Krueger to go along with plenty of recrimination. Most of those were on both sides, however, when they used numbers at all focused on the 2005 Giants and their seven Caribbean players or the 2005 season as a whole. While we all know that it’s often said that Caribbean players “don’t walk off the island,” I wondered (leaving aside whatever racial component might be involved) just how true historically that bit of conventional wisdom is.
To look at this question I used the Lahman database to find the 992 “Caribbean” players born since 1930. The cutoff 1930 was used because players born at that time would be coming of age just after Jackie Robinson broke the color barrier, a barrier that also prevented dark-skinned Caribbean players from entering the majors.
I included those born in the following places.
Dominican Republic 386 Puerto Rico 214 Venezuela 167 Cuba 149 Mexico 97 Panama 44 Colombia 7 Bahamas 5 Aruba 4 Jamaica 3 Honduras 1
Now of course not all these countries are properly Caribbean but I’ve included them because of the general perception that players from these countries possess similar traits.
As you might expect, the Dominican Republic is to baseball what Kenya is to middle- and long-distance running, producing 386 major league players, 106 of whom were active in 2004. And as in Kenya where the tiny Nandi region, representing 1/2000th of the world population, has produced more than 20% of all winners of international distance running events, 65 of those 386 players hail from the city of San Pedro de Macoris, including Sammy Sosa, Alfonso Soriano, Tony Fernandez, George Bell and Rico Carty.
But can you name the lone player from Honduras?
Why yes it was Gerald Young who debuted July 8, 1987 and played eight seasons for Houston, Colorado and St. Louis. I knew you’d get that one.
Admittedly looking only at birth country isn’t the best approach since some of these players, while born in Caribbean countries, were actually raised in the United States. The most prominent current example is Albert Pujols. But be that as it may, this is the best approach I could conjure up given the data at hand, and since I don’t have complete information on who was raised where, I did not make any adjustments. I should also mention that from other sources I’ve seen it would appear the Lahman database is not complete in tracking the birth countries of players. For example, one source indicates there were over 420 Dominicans who had played in the majors through last season. The database does contain enough data, however, to get a feel for group differences if they exist.
The Caribbean players included in this list span the career of pitcher Raul Sanchez, who was born in Cuba in 1930, debuted with the Washington Senators in 1952 and pitched just 89 and two-third innings over his three-year career, all the way to Dioner Navarro of the Dodgers born in 1984 in Venezuela and who debuted in 2004.
For comparison purposes I used all of the non-Caribbean players born since 1930—which comprised the careers of 7,510 players—as a control.
When comparing these two sets of players the first thing that jumps out at you is the increasing number of Caribbean players in the game; it rose from around 6% in 1955 to almost 20% today. And as you can see in the following graph, that trend has visibly accelerated since the late 1980s.
That acceleration is the result of teams becoming more actively engaged in Latin America and the Caribbean by opening baseball academies and scouting bureaus in hotspots like San Pedro de Macoris in the Dominican Republic. According to an excellent three-part series recently run in the Denver Post, 28 of the 30 major league teams have baseball operations in the Dominican, where the teams spent a total of $75 million. The Rockies, for example, operate a complex in Boca Chica, a 40-minute drive to the east of Santo Domingo, where they currently host 38 players.
The first question of course is: Do Caribbean players really walk less than non-Caribbeans? The cumulative totals of both sets of players seem to indicate that this is indeed the case.
Count AVG SLUG OBP OPS BB/PA SO/PA All Players 7510 0.259 0.394 0.324 0.718 0.086 0.150 Caribbean Players 992 0.265 0.395 0.320 0.715 0.073 0.143 Non-Caribbean Player 6518 0.258 0.394 0.325 0.718 0.089 0.151
As you can see from the table Caribbean players recorded .073 walks per plate appearance while non-Caribbeans had .089. Over 700 plate appearances this translates to a difference of 11 walks (51 to 62). Graphically you can see that the difference between the two groups remains fairly constant over time; the gap shrinks in the mid-1970s and then grows again in the mid-1980s before shrinking again some more.
But you can also see from the table that although they draw fewer walks, Caribbean players also strike out less often, averaging around five fewer strikeouts per 700 plate appearances.
You’ll also notice that although they walk less often, Caribbean players hit for a higher average, and as a result record a higher slugging percentage, while non-Caribbeans have a higher Isolated Power (ISO). The combination of a higher ISO and walk rate means that non-Caribbeans also enjoy a slightly higher OPS. This also partly explains the lower strikeout rate. Caribbean players make fewer outs and strike out less frequently per plate appearance.
One way to interpret these results would be to say that they’re consistent with the idea that, as a group, Caribbean players are more aggressive at the plate. This results in a higher batting average but fewer walks and a little less power, based on the theory that more selective hitters will more often get pitches that they can drive for extra bases.
When you look at the data over time, you find that whereas there were once gaps in slugging percentage and OPS, those have narrowed considerably to the point where Caribbean players now outpace non-Caribbeans in those categories. In fact, in 2004 the 243 Caribbean players recorded a .270/.328/.438 line, while the 1,022 non-Caribbeans were at .265/.332/.424.
So while Caribbean players close the gap in all other areas, they continue to lag behind in drawing walks. So maybe there is something to this bit of conventional wisdom after all?
At first blush I had assumed that this general narrowing in offensive performance resulted from the fact that in the 1960s and into the 1980s Caribbean players were generally cast as middle infielders. This bias, so went my theory, would have the effect of major league teams selecting Caribbean players who fit that mold, a mold that also correlates with lower power and on-base numbers. As scouting became more sophisticated in these areas, the distribution of talent would naturally expand, which in turn would produce Caribbean players with a wider range of skill sets.
Alas, to paraphrase “Darwin’s Bulldog” Thomas Henry Huxley, it was a beautiful theory spoiled by an ugly fact.
I produced the following table that shows the number of games played by Caribbean players at the various positions by decade, which illustrates that while the percentage of games played at shortstop has indeed trended downward over time, the mix has not changed nearly as substantially as I had imagined. For example, the percentage of games played in the outfield has remained essentially constant while those at first base, a position usually associated with greater offensive production, have actually declined.
Position 1955-1965 1966-1975 1976-1985 1986-1995 1996-2004 1B 12.2% 7.0% 8.0% 7.7% 6.3% 2B 13.6% 21.5% 15.2% 17.5% 15.4% 3B 6.4% 9.4% 7.1% 4.9% 10.4% C 2.6% 6.3% 6.5% 9.6% 12.2% DH 0.0% 2.1% 4.9% 5.6% 5.1% OF 34.8% 33.4% 33.7% 29.3% 29.0% SS 30.3% 20.3% 24.6% 25.4% 21.6%
But what this does indicate is that as a group, Caribbean players are more valuable than non-Caribbeans. In the last decade they’ve matched or surpassed the offensive production of non-Caribbeans. They continue to play the more demanding defensive positions on the right side of the defensive spectrum, logging 37% of their games at shortstop and second base, whereas non-Caribbeans man these positions just 19% of the time.
It would appear from the data that, as with many stereotypes, there is a grain of truth to the idea that Caribbean hitters are generally more aggressive than non-Caribbeans. Although the difference doesn’t appear to be shrinking, Caribbean players are beginning to outpace non-Caribbeans in overall production anyway while continuing to play more demanding defensive positions.
In the end Krueger was fired not because he linked Caribbean players to less frequent walks, but because he used the term “brain dead” in the same sentence. And as I’ve discussed before on my blog, the linkage of intelligence with race (and sometimes any racial difference at all), however tenuous, is a subject that is taboo in our society.