What separates these two groups of hitters, besides their stats?
| 2010-14 | BA | OBP | SLG | OPS | OPS+ | BAbip | SO% | WAR* | oWAR* | dWAR* |
| Group A | .259 | .327 | .412 | .739 | 102.7 | .304 | 19.4% | 2.30 | 2.27 | -0.10 |
| Group B | .268 | .329 | .414 | .743 | 103.9 | .304 | 16.4% | 2.66 | 2.58 | 0.19 |
These groups cover all non-pitchers with 500+ plate appearances over the last five years. Each group has over 140 players, totaling more than 200,000 PAs. Group OPS+ is weighted by playing time. The WAR figures are rates per 650 PAs; the ratio between the groups would be about the same if expressed per 162 games.
Assignment to Group A or Group B was not based on any baseball statistic. What was it based on?
Here are their counting stats, per 650 PAs:
| 2010-14 | AB | R | H | 2B | 3B | HR | RBI | BB | IBB | SO | HBP | SH | SF | GDP | SB | CS |
| Group A | 581 | 76 | 150 | 30 | 3 | 17 | 72 | 55 | 4 | 126 | 6 | 3 | 5 | 13 | 11 | 4 |
| Group B | 586 | 73 | 157 | 30 | 3 | 16 | 72 | 51 | 5 | 107 | 5 | 4 | 5 | 15 | 11 | 4 |
To round out the picture, here’s a quick look at the pitchers, divided on the same basis:
| 2010-14 | ERA+ | OPS+ | SO% | WAR* |
| Group A | 106.2 | 97.8 | 19.3% | 2.09 |
| Group B | 108.2 | 95.9 | 20.3% | 2.21 |
This is all pitchers with 100+ innings over the last five years. Each group has over 140 pitchers, totaling more than 40,000 innings. Group ERA+ and OPS+ are weighted by playing time. WAR is the rate per 200 innings.
__________
Congratulations to Joseph, who determined that Group A is players born in the United States,* and Group B is those born elsewhere.
This quiz grew from wanting to check the old chestnut that “you don’t walk off the island.” Of course, not all foreign-born players hail from the Caribbean area. But more than half of the foreign-born hitters in this study came from Caribbean Islands, mainly the Dominican Republic (46), Cuba (13) and Puerto Rico (13).* And most of the rest came from countries on the Caribbean — Venezuela (48), Colombia (3), Panama (3), Mexico and Nicaragua (1 each).
Anyway … The recent data do show a slightly lower walk rate among foreign-born hitters. But there are even bigger differences:
- Strikeout rate is much lower among the foreign-born — 16.4% vs. 19.4%, a margin of almost 20 Ks per 650 PAs — leading to a 9-point edge in BA despite the same BAbip. Their gains via contact more than offset the lower walk rate, and came with very little loss in the power departments: U.S.-born players averaged one extra home run per 650 PAs, while doubles and triples were even.
Overall, the foreign-born hitters had a slight edge in OBP and slugging, driven by BA. The rest of the season rates are extremely close. - Foreign-born players had more defensive value, with an edge of about 0.3 dWAR per 650 PAs, accounting for almost all of their overall edge in WAR. This could be from actual skill, or from a higher concentration at the more valuable defensive positions. It doesn’t seem to be age-related, as the unweighted average of both groups is just under 29.
Foreign-born pitchers had a higher strikeout rate, although by a smaller margin (20.3% vs. 19.3%). This might grow out of their developing years: Since a certain degree of strikeout skill is needed for success, foreign-born pitchers might have had to hone this skill more sharply because they faced hitters who make more contact.
Finally, here’s a longer view of the hitters, covering 1995-2014 and all those with 300+ PAs:
| 1995-2014 | BA | OBP | SLG | OPS | OPS+ | BAbip | SO% | WAR* | oWAR* | dWAR* |
| U.S.-born* | .267 | .339 | .427 | .766 | 101.4 | .304 | 17.2% | 2.17 | 2.14 | -0.08 |
| Foreign-born | .272 | .335 | .427 | .762 | 100.6 | .304 | 15.8% | 2.31 | 2.28 | 0.15 |
WAR figures above are per 650 PAs, as are the following:
| 1995-2014 | AB | R | H | 2B | 3B | HR | RBI | BB | IBB | SO | HBP | SH | SF | GDP | SB | CS |
| U.S.-born* | 577 | 81 | 154 | 31 | 3 | 18 | 77 | 59 | 4 | 112 | 6 | 3 | 5 | 13 | 11 | 4 |
| Foreign-born | 583 | 80 | 159 | 31 | 3 | 18 | 77 | 52 | 5 | 103 | 5 | 4 | 5 | 14 | 11 | 5 |
A similar picture on strikeouts and value, but milder.
And now we throw the floor open to comments.
_____
* Puerto Rico is a U.S. commonwealth whose people are U.S. citizens. But it’s treated as non-U.S. by my data source, Baseball-Reference.com, and there is some logic behind that. No slight is intended.
Just a guess: Group A is visiting team, Group B is home team.
Paul, that’s an interesting guess, but not correct.
Group A is younger than Group B? (Say, minus/plus age 30?)
Age is not a factor.
Date of birth? Jan-June vs July-Dec.
Right-handed vs left-handed? If so, I’m guessing Group A is the righties.
Not righty-lefty. And no dates are involved, birth or otherwise. Also, not height or weight.
Born in the USA vs. born someplace else?
DING-DING-DING! Good job, Joseph. How did you deduce it?
It was a total guess–playing a hunch.
The difference in dWAR is almost entirely due to higher concentrations at more valuable positions. oWAR + dWAR – WAR = positional adjustment. When you subtract the positional adjustment from dWAR you get about the same Rfield numbers for both groups. Alternatively, WAR – oWAR tells you the same thing.
Joe, I had to walk myself through it a couple of times, but I now believe you are right.
When I use your method on the first set of WAR numbers in this post — not the rounded ones published, but the precise ones in my spreadsheet — I get an edge of about 0.5 Rfield per 650 PAs for the foreign-born. Which is probably insignificant.
And the “concentration” hypothesis jibes with the stereotype, e.g., Dominican Republic as the cradle of shortstops. But not to rely on the stereotype, I counted the players listed as *2, *6 or *4 in each group (meaning C, SS or 2B in 2/3 of team games). US group had 27% at those 3 spots (100/365), foreign group 43% (64/148). (FWIW, foreign group had 30 SS and just 5 2B.)
I’m a little surprised there isn’t much difference in the walk rates. Back in the day when Alfredo Griffin, Ozzie Guillen and Damaso Garcia were playing you heard a lot about “Nobody ever walked off an island”…
I wonder how recent the change is (if the old adage about not walking off of an island was ever true), or if the 2010-2014 is just an odd cut of the data.
Looking at the league leaders in walks, from 2010-2014 I see the following foreign born players at the top of one of the two leagues (just looking at the #1 player in each league): Votto (twice), Bautista, and Carlos Santana.
Other than David Ortiz (2006-2007), I don’t think any other foreign born player has ever led either league in walks (maybe there is another Canadian player in the mix somewhere, or someone born overseas to parents who were serving in the military).
What’s James’ quote from Gould about evolution? Over time the differences between the population disappears? As executives put more value on OBP could that be driving these changes? It would
Tabbed to the submit button too soon.
Given that there is a “bigger” difference in walks between 1995-2014 (7 walks between the two groups) and 2010-2014 (4 walks between the two groups), it appears that the difference is shrinking over time (because the 1995-2014 data include the 2010-2014 data). How difficult is it to run the numbers for just 1995-1999?
Mexico is not part of the Caribbean http://en.wikipedia.org/wiki/Caribbean
Simmer down, Lou.
http://www.answers.com/Q/What_bodies_of_water_border_Mexico
The southeastern tip of Mexico has a Caribbean frontage, specifically the state of Quintana Roo.
As it happens, Luis Cruz — the only Mexican-born player with 500 PAs in the last 5 years — was born in Navojoa, Sonora, which is on the Sea of Cortes (Gulf of California). But how do we know he didn’t grow up in Chetumal, Cozumel or Cancun?