The Small Sample Size Heroes of April 10

Learning to love the momentary masters of the leaderboards.

Apr 10, 2013, 7:27 PM UTC

Marc Serota

Is your favorite player struggling?

Did he show promise in Spring Training, only to leave you disappointed with just three hits in six, with an 0-for-19 stretch? First, why are you a Jeff Keppinger fan? Are you his dad or something? Second, Mr. Keppinger, just one reminder: Do not panic.

Has a slugger of unexpected prominence emerged on your favorite team?

Does he have five home runs and 17 RBI, one of the hottest starts in all of history? Just seven games into the season and he’s already proven himself to be the offensive juggernaut that your team so desperately needed -- he even has more RBI than some teams have and he’s just one guy! Let’s talk in June and see how that’s going.

Is your team’s ace giving up grand slams and walking an annoyingly high percentage of batters?

Are you creating conspiracy theories about his new mustache being the culprit of his 10.97 ERA and his dramatic change in groundball-to-flyball ratio? Maybe if he shaves he’ll be fine ... or perhaps some things just take time.

Whether it's a Daniel Nava first at-bat grand slam, University of Michigan guard Spike Albrecht's 17-point explosion off the bench in the NCAA National Championship game on Monday, or Jeff Samardzija's suddenly brilliant command, we can't stop ourselves from reacting. We're drawn to small sample sizes like magnets to refrigerators, cats to laser pointers, or soccer moms to minivans.

Unlike most sabermetricians or statisticians, I’m not here to inform you of the evils of small samples. The truth is, small sample sizes are a blast. Did you know that Willie Mays started his career 1 for 26? Or that when Kevin Maas was called up in 1990, he hit a record 10 home runs in his first 25 games? There are plenty more stories like that; Ted Berg (formerly of SNY now of USA Today) even wrote a small sample size song last season that has over 27,000 views on YouTube -- which is a large enough sample to confirm that I am not alone in my admiration of small sample sizes.

Consider small samples the car crashes of the baseball world: You can’t look away, and the best you can do is hope that the hitters aren’t permanently damaged when you see the wreckage that is a .160 batting average. Like many things that bring us pleasure, with small samples you have to exercise some restraint. The philosophy should be that if you’re willing to read all of the Surgeon General’s warnings -- DO NOT ANALYZE THIS DATA and DO NOT EXTRAPOLATE FROM THIS SAMPLE and SUBJECT TO CHANGE VERY RAPIDLY AS MORE GAMES ARE PLAYED -- then by all means, observe, pontificate, and even snark if you’d like. The samples aren’t big enough to hurt you.

I created a series of April 10^th leaderboards, a snapshot of how hitters and pitchers fared from Opening Day through April 10in 2011, 2012, and this season. It’s the ultimate small sample size -- some players had just five or six games -- so while we can’t do a Monte Carlo simulation, we can at least let our hair down and take a peak.

Home Runs

2011		2012		2013
Nelson Cruz	5	David Freese	3	Justin Upton	6
Howie Kendrick	4	Carlos Beltran	3	Mike Morse	6
Ryan Braun	4	Yoenis Cespedes	3	Coco Crisp	5
Mark Teixeira	4	Jay Bruce	3	Dexter Fowler	4
Ian Kinsler	4	Omar Infante	3	Chris Davis	4
Miguel Cabrera	4	Corey Hart	3	Will Middlebrooks	4
		Miguel Cabrera	3	Mark Reynolds	4
				John Buck	4
				Alex Rios	4

While I would never advocate relying on small samples, power numbers are probably the most reliable -- regardless of what people say about Ichiro Suzuki, most players cannot hit home runs whenever they want to. In 2011, everyone on that list hit at least 20 home runs, except for Howie Kendrick who hit 18. Mark Teixeira finished third on the home run leaderboard, and Ryan Braun finished in 10^th place.

Last season, Yoenis Cespedes dazzled the baseball world by hitting three home runs in his first four games and finished the season with 23. But of the seven hottest home run hitters at the start of the season, only one finished in the top 10, and that was American League MVP Miguel Cabrera.

This season, Justin Upton has already matched 35 percent of his home run productivity from last season, and Mike Morse is already at 28 percent. While it’s fun to imagine that Upton could be the first player to hit 138 home runs in one season, those dreams will be dashed soon enough. Still, he’s on the April 10^th leaderboard and in all small samples, anything is possible.

Adam Jones (Joy R. Absalon-USA TODAY Sports)

Batting Average

2011		2012		2013
Miguel Montero	.500	Austin Jackson	.562	Adam Jones	.500
Yunel Escobar	.476	Zachary Cozart	.471	Carlos Santana	.500
Joey Votto	.455	Josh Hamilton	.450	Lance Berkman	.480
Paul Janish	.444	David Freese	.444	Carl Crawford	.458
Nick Hundley	.440	Ryan Sweeney	.444	Jean Segura	.458
Matt Kemp	.438	Rafael Furcal	.435	Jose Reyes	.444
Shane Victorino	.417	Emilio Bonifacio	.421	Adrian Gonzalez	.435

Batting averages can take wild swings in small samples, because good production in a one-day period is enough to make today's sub-Mendoza hitter tomorrow's Babe Ruth. For instance, on Tuesday morning Brett Gardner was hitting .167/.242/.267. He went 4-for-5 last night and is now hitting .257/.333/.400. Robinson Cano entered the same game hitting .222/.300/.481 after his two home run game on Monday. He went 4-for-6 last night with two doubles and a home run, and now he's hitting .303/.361/.667.

When looking at charts of small sample sizes, which I did for the better part of yesterday because, frankly, I didn’t have anything better to do, it seemed like there was a trend emerging. Of course, I was only looking at charts for seven different metrics, which is clearly a small sample size, but there were three distinct things happening:

First, there are players that appear on the leaderboard that make complete sense. For instance, in 2011 Matt Kemp hit .438 at the start of the season and stayed hot throughout the month of April, hitting .368/.446/.613. Kemp was on fire again at the end of the season and finished second in the National League MVP voting.

Second, you see names of players that have been streaky throughout their careers, getting hot for a short periods and finishing the season around league average. Think Shane Victorino in 2011, Emilio Bonifacio in 2012, and probably, sadly, Carl Crawford in 2013.

Finally, there are always a few names that are small-sample hot streaks personified: They show up on the April 10^th leaderboard only to disappear into oblivion, sometimes even before the season ends. Ryan Sweeney started the 2012 season with the Boston Red Sox hitting .444, but he suffered a string of injuries, including a self-inflicted fractured right pinkie after punching a door, and now he's on a minor-league contract in the Cubs system.

Sergio Romo (Thearon W. Henderson )

Saves

2011		2012		2013
Huston Street	6	Javy Guerra	5	Sergio Romo	5
Brian Fuentes	5	Brandon League	4	Craig Kimbrel	4
Leo Nunez	5	J.J Putz	4	Rafael Soriano	3
Neftali Feliz	5	Jim Johnson	4	Joel Hanrahan	3
Mariano Rivera	5	Craig Kimbrel	4	Addison Reed	3
Jonathan Broxton	5			Jason Grilli	3
				Rafael Betancourt	3
				Tom Wilhemlsen	3

Given the volatility of relief pitchers, it should be a no surprise to anyone that the pitchers who show up on the Saves list have interesting stories. Just two years removed from the 2011, only two of the pitchers on the April 10 list are still closers. The others? Well, one retired, one is injured, and the other -- Leo Nunez -- isn't even named Leo Nunez anymore, nor is he a closer. And even though Javy Guerra was the first to reach five saves last season as the Dodgers' Opening Day closer last year, he started this season in the minors. But in even larger datasets for closers the same things happen: Wannabes and proven closers get stuck in the revolving door of duties, job expectations, and performance.

In keeping with the theme, this is just a small sample size of the April 10^th leaderboards from the past three seasons. If you're interested in seeing the standings for RBI, Strikeouts, Walks, Lowest Starter ERA after 2 starts, they can all be found in a spreadsheet here. You'll be surprised to learn that Chris Narveson was the best starting pitcher through two starts in 2011, and that Jonathan Sanchez was on the strikeout leader board that year as well.

Bottom line, we shouldn't be afraid to discuss small sample sizes. There's no harm in saying that Chris Davis might have 400 RBI before the season is over, as long as it's in jest. And as long as you realize that your chances of getting hit by a bus are greater than Jed Lowrie batting .500 for the rest of the season, I give you my blessing to embrace the small sample size.