Michael Rand started RandBall with hopes that he could convince the world to love jumpsuits as much as he does. So far, he's only succeeded in using the word "redacted" a lot. He welcomes suggestions, news tips, links of pure genius, and pictures of pets in Halloween costumes here, though he already knows he will regret that last part.

Follow Randball on Twitter

Mid-day talker: When is a small sample size no longer a small sample size?

Posted by: Michael Rand under Professional baseball, Target Field, Twins fans Updated: June 11, 2014 - 2:05 PM

Through seven innings in Wednesday afternoon's game against the Blue Jays, Joe Mauer was 2-for-3 with a walk. That's a .750 on-base percentage for the slumping Mauer, and if he could keep that up for the rest of the season he would be the MVP.

But most of us understand those numbers are what we call a "small sample size" -- a sometimes relevant set of data, but numbers that nonetheless can't be extrapolated to inform us of a trend.

In the larger context, Mauer is having a poor season. But his diminished output only represents about 5 percent of his career at-bats. Are these two-plus months of Mauer still a small sample size?

We asked the honest question on Twitter: when does a small sample size for a hitter magically become an adequate sample size? Because while most of us like to toss around the "small sample size" phrase these days, very few of us are actually well-versed in what it means.

Here is the ENTIRE SAMPLE SIZE of the responses to our second tweet:

 

 

The most consistent response to that and a previous tweet pointed us to Fangraphs, which has attempted to tackle this very question. A study suggests the following benchmarks "when certain statistics stabilize for individual hitters":

 50 PA: Swing % 100 PA: Contact Rate 150 PA: Strikeout Rate, Line Drive Rate, Pitches/PA 200 PA: Walk Rate, Groundball Rate, GB/FB 250 PA: Flyball Rate 300 PA: Home Run Rate, HR/FB 500 PA: OBP, SLG, OPS, 1B Rate, Popup Rate 550 PA: ISO

In essence, the size of a relative sample is relative to what you're measuring. With someone like Josmil Pinto, with limited career at-bats, this is fairly cut and dried. With Mauer, though, it's still complicated. Do we choose to believe the greater sample -- more than 5,000 career plate appearances, which suggest Mauer is a very good hitter -- or the smaller but still relevant sample size from this season?

That's the crux of the Mauer debate.

ADVERTISEMENT

Toronto 0 Top 2nd Inning
NY Yankees 0
Washington - G. Gonzalez 3:05 PM
Cincinnati - J. Cueto
St. Louis - S. Miller 3:05 PM
Chicago Cubs - J. Arrieta
Baltimore - B. Norris 3:10 PM
Seattle - C. Young
Arizona - J. Collmenter 6:05 PM
Philadelphia - C. Lee
NY Mets - J. Niese 6:10 PM
Milwaukee - W. Peralta
San Diego - O. Despaigne 6:10 PM
Atlanta - J. Teheran
Miami - T. Koehler 6:10 PM
Houston - J. Cosart
Boston - J. Lackey 6:10 PM
Tampa Bay - J. Hellickson
Chicago WSox - C. Sale 6:10 PM
Minnesota - L. Darnell
Cleveland - Z. McAllister 6:10 PM
Kansas City - J. Guthrie
Oakland - S. Gray 7:05 PM
Texas - N. Tepesch
Pittsburgh - J. Locke 7:10 PM
Colorado - T. Matzek
Los Angeles - C. Kershaw 8:05 PM
San Francisco - R. Vogelsong
Detroit - J. Verlander 8:05 PM
LA Angels - M. Shoemaker
Sporting Kansas City 6:00 PM
Toronto FC
Columbus 6:30 PM
New England
Calgary 26 FINAL
Edmonton 22
Winnipeg 23 FINAL
Brt Columbia 6
Ottawa 6:00 PM
Hamilton
Toronto 9:00 PM
Saskatchewan
Winnipeg 7/31/14 6:00 PM
Hamilton
Toronto 8/1/14 6:00 PM
Montreal
Brt Columbia 8/1/14 9:00 PM
Calgary
Saskatchewan 8/2/14 6:00 PM
Ottawa
Los Angeles 3:00 PM
Seattle
Indiana 7:00 PM
San Antonio
New York 9:00 PM
Phoenix

ADVERTISEMENT

ADVERTISEMENT

ADVERTISEMENT

ADVERTISEMENT

question of the day

Poll: Should the Twins replace Ron Gardenhire?

Weekly Question

ADVERTISEMENT

Connect with twitterConnect with facebookConnect with Google+Connect with PinterestConnect with PinterestConnect with RssfeedConnect with email newsletters

ADVERTISEMENT