Probably the most common word we use over at Rotowire this time of the year is “patience.” Plenty of people freak out over slow starts from their star players, and start seeing their fantasy seasons flash before their eyes because Carlos Beltran has only one stolen base through April, or because Eric Chavez‘s home run total isn’t matching his price tag. The point we have to keep emphasizing is that a baseball season is a marathon, not a sprint, and that a player expected to put up big numbers in all likelihood will do so, once that marathon ends in the fall. Too many owners give up and start breeding lilacs when their teams are still title contenders, because they don’t have patience to wait for production that’s just over the horizon.
Is it true though? Is a slow start really just a statistical blip which will sort itself out over the next five months? It’s never a bad idea to challenge conventional wisdom, and every self-respecting Baseball Prospectus writer needs only the feeblest of excuses to start playing around with numbers. So let’s test the hypothesis that production comes in fits and spurts, not steadily over the course of the season, and see if there’s anything that can be learned in the process.
I compiled a list of every player who hit 40 or more home runs in any of the last three seasons, or stole 40 or more bases over the same stretch. The former list is longer than the latter of course (27 40+ home runs seasons, versus 16 40+ stolen base seasons), but I’m more interested in sifting through for possible patterns than proving a rigorous hypothesis, so the sample sizes on the data will do. I broke each of those seasons down by month (March stats being folded into April, and October into September) and figured out what percentage of their final total they accumulated in each month.
Before you peek at the numbers though, a trivia question: who are the only three players to reach 40 home runs or 40 steals in each of the last three seasons? I’ll give you answer at the bottom of the page.
Here were the results for the speedsters.
Player SB Mar-Apr May June July August Sept-Oct Scott Podsednik 2004 70 17.1% 14.3% 12.9% 14.3% 14.3% 27.1% Carl Crawford 2004 59 20.3% 20.3% 18.6% 13.6% 11.9% 15.3% Juan Pierre 2004 45 13.3% 20.0% 6.7% 17.8% 22.2% 20.0% Carlos Beltran 2004 42 16.7% 9.5% 9.5% 11.9% 38.1% 14.3% Bobby Abreu 2004 40 0.0% 25.0% 12.5% 27.5% 17.5% 17.5% Juan Pierre 2003 65 15.4% 18.5% 24.6% 21.5% 7.7% 12.3% Carl Crawford 2003 55 12.7% 9.1% 12.7% 16.4% 30.9% 18.2% Alex Sanchez 2003 52 5.8% 13.5% 23.1% 13.5% 21.2% 23.1% Scott Podsednik 2003 43 2.3% 16.3% 18.6% 14.0% 27.9% 20.9% Carlos Beltran 2003 41 2.4% 17.1% 26.8% 26.8% 0.0% 26.8% Dave Roberts 2003 40 25.0% 20.0% 15.0% 0.0% 27.5% 12.5% Luis Castillo 2002 48 18.8% 25.0% 10.4% 14.6% 16.7% 14.6% Juan Pierre 2002 47 19.1% 19.1% 12.8% 8.5% 23.4% 17.0% Dave Roberts 2002 45 11.1% 20.0% 11.1% 24.4% 20.0% 13.3% Alfonso Soriano 2002 41 14.6% 24.4% 17.1% 17.1% 14.6% 12.2% Vladimir Guerrero 2002 40 17.5% 7.5% 17.5% 12.5% 20.0% 25.0%
From a statistical perspective, there’s nothing funny going on here at all. About as many months fall within one and two standard deviations of the mean–16.7%, or an even one-sixth, of course–as you’d expect, and there’s no apparently meaningful pattern to specific months being good or bad. From a fantasy, and especially a Beltran-owner perspective, the news is bright. No less than three players (Bobby Abreu in 2004, and Beltran and Dave Roberts in 2003) reached the 40 steal plateau even though they went an entire calendar month without swiping a base, and in fact April features more “bad” months (four players more than one SD below the mean, and three that were more than two) than any other period.
In other words, a slow start or an off-month for a player with 40-steal talent doesn’t seem to hinder a player’s chances at reaching that mark. (If it did, fewer bad Aprils than expected would show up in the sample, one of the advantages of using a self-selecting group–although that can’t be considered a really firm conclusion without looking at a larger data sample).
The data for the sluggers, however, is a bit more surprising.
Player HR Mar-Apr May June July August Sept-Oct Adrian Beltre 2004 48 14.6% 10.4% 14.6% 20.8% 27.1% 12.5% Adam Dunn 2004 46 17.4% 8.7% 23.9% 13.0% 23.9% 13.0% Albert Pujols 2004 46 15.2% 17.4% 10.9% 19.6% 26.1% 10.9% Barry Bonds 2004 45 22.2% 8.9% 13.3% 15.6% 24.4% 15.6% Manny Ramirez 2004 43 11.6% 20.9% 14.0% 16.3% 20.9% 16.3% Jim Edmonds 2004 42 11.9% 14.3% 9.5% 31.0% 23.8% 9.5% Jim Thome 2004 42 16.7% 11.9% 35.7% 9.5% 19.0% 7.1% Paul Konerko 2004 41 12.2% 12.2% 24.4% 17.1% 12.2% 22.0% David Ortiz 2004 41 12.2% 14.6% 24.4% 14.6% 14.6% 19.5% Alex Rodriguez 2003 47 19.1% 12.8% 10.6% 10.6% 31.9% 14.9% Jim Thome 2003 47 8.5% 17.0% 19.1% 12.8% 21.3% 21.3% Barry Bonds 2003 45 17.8% 11.1% 20.0% 24.4% 15.6% 11.1% Richie Sexson 2003 45 20.0% 15.6% 13.3% 15.6% 15.6% 20.0% Javy Lopez 2003 43 9.3% 20.9% 23.3% 16.3% 11.6% 18.6% Albert Pujols 2003 43 11.6% 23.3% 18.6% 14.0% 18.6% 14.0% Carlos Delgado 2003 42 19.0% 19.0% 23.8% 11.9% 7.1% 19.0% Frank Thomas 2003 42 11.9% 11.9% 16.7% 19.0% 31.0% 9.5% Jason Giambi 2003 41 12.2% 14.6% 26.8% 19.5% 12.2% 14.6% Sammy Sosa 2003 40 12.5% 2.5% 10.0% 32.5% 20.0% 22.5% Alex Rodriguez 2002 57 15.8% 14.0% 12.3% 21.1% 21.1% 15.8% Jim Thome 2002 52 13.5% 13.5% 19.2% 13.5% 21.2% 19.2% Sammy Sosa 2002 49 16.3% 22.4% 16.3% 10.2% 22.4% 12.2% Barry Bonds 2002 46 17.4% 21.7% 15.2% 8.7% 23.9% 13.0% Rafael Palmeiro 2002 43 14.0% 11.6% 14.0% 25.6% 23.3% 11.6% Lance Berkman 2002 42 23.8% 16.7% 19.0% 14.3% 14.3% 11.9% Shawn Green 2002 42 7.1% 23.8% 28.6% 7.1% 23.8% 9.5% Jason Giambi 2002 41 9.8% 24.4% 17.1% 17.1% 9.8% 22.0%
To make the pattern clearer, I’ll break it down in terms of “good/great” months (at least more than one SD above the mean), “average” months (within one SD) and ‘bad/awful’ (more than one SD below) months.
Mar-Apr May June July August Sept-Oct
Good(Great) 1(0) 3(0) 8(2) 4(2) 10(2) 1(0)
Average 22 20 15 18 15 21
Bad(Awful) 4(0) 4(1) 4(0) 5(0) 2(0) 5(0)
The numbers in parentheses are months where the player was more than two SDs above or below the mean. In checking out one piece of conventional wisdom, I seem to have stumbled onto another one: that offense heats up in the summer months. While the “bad” months stay normally distributed over the course of the year, there’s a marked increase in “good” months for the 40+ home run crowd through June and August. (If I had to hazard a guess as to why July doesn’t see a similar pattern, I would say that the plate appearances lost to the All-Star break might play a role). It’s important to note that there’s no corresponding increase in “bad” Aprils and Mays though, nor does there seem to be a correlation between big summers and poor springs. As a group, hitters who slam 40-plus home runs are simply prone to doing more than their share of slamming during the summer.
Of course using a self-selecting group has disadvantages as well as advantages. We have a good idea that players who reach a certain nice round number in home runs or steals can do it without sprinting out of the gate. What we don’t know yet is whether a slow start has a predictable impact on a player’s bottom line. A 40-home run player isn’t less likely to have had a bad April, but is a player who has a bad April less likely to reach 40 home runs? That one I’m going to take a swing at next week.
Trivia answer: I tried to phrase it so you’d think it was a trick question, but it isn’t–nobody appeared twice on the home run list and once on the steals list, for instance. Barry Bonds and Jim Thome hit 40 or more home runs in 2002-2004, while Juan Pierre stole 40-plus bases in each of those years.
Erik Siegrist is a beat writer for RotoWire, covering the Marlins, Nationals and White Sox.
Thank you for reading
This is a free article. If you enjoyed it, consider subscribing to Baseball Prospectus. Subscriptions support ongoing public baseball research and analysis in an increasingly proprietary environment.
Subscribe now