Rounding the Bases: Sample Size & What It Means

Posted by Ray Flowers on 07/11/2012 | 1 Comments

Tags: Adam Jones, Mike Napoli, Alex Gordon, Tim Lincecum, Jason Kubel, Tyler Colvin, Brian McCann, Daniel Murphy, Alex Gordon, Eric Hosmer, Mike Napoli, Dustin Pedroia, Alexei Ramirez, Jimmy Rollins, Justin Upton, Delmon Young, Michael Young

Tim Lincecum is the worst pitcher ever. Alex Gordon has been a massive disappointment. Mike Napoli is flipping killing my fantasy team. I hear comments like that on a daily basis from people. Maybe all of the above is true, but there might also be something else going on here. What is that something else? The most obvious situation that has to be addressed is expectations. Were your expectations for a player reasonable given his skill level, age, club situation etc. Second, it's sample size. A quick example. Adam Jones has been a superstar this year, a top-25 performer overall, hitting .289 with 20 homers, 44 RBI, 54 runs and 11 steals. However, were you aware that since the start of June that he's hit .252 with 10 RBIs an a .681 OPS? Yeah, he's been pretty bad of late. So that brings me to the heart of today's article --- sample size. What does it mean, when is it important, and how should you work with it?

Sample Size Talk

Numbers can be manipulated. That's the dirty little secret we deal with in baseball. Take a snapshot of time and I can “prove” to you something that may not even be true. Examples.

The last four weeks Alex Rios is a better fantasy player than Ryan Braun and Carlos Gonzalez. Would you rather have Rios than the other two players on your fantasy squad? Didn't think so.

The past three weeks Jason Kubel has been more productive than Jose Bautista. Uh, yeah, I'd rather have Bautista too.

The last two weeks Tyler Colvin has been a bigger fantasy performer than Matt Holliday. Pretty obvious who you'd rather have on your club in the second half.

This is the heart of the problem when it comes to breaking down players. What sample size do we want to use when looking at players? Remember, if we were to look at the past six weeks Adam Jones is barely a replacement level outfielder. If we back that up to the start of the season Jones becomes an elite player. So, should you look at two weeks, two months, two years when breaking down a player? Honestly, there is no “right” answer, but in general the larger the sample size the more valid the data is. Let's take the case of Colvin, who again, has been a top-10 overall hitter than last two weeks thanks to a .348 average, six homers, and 17 RBIs.

If we look at him the last four weeks that ranking drops down to about 40th overall.
If we go back to the start of the season we're at about 125th.
If we take it back and look at the 2011 season he was barely inside the top 900.

So which player is Colvin? I'd posit that he is none of those players. He's certainly not an elite hitter. I also feel pretty confident he isn't a top-50 overall fantasy player. I have a hard time thinking, given his skill set, that he's even a top-125 player, but I also know there is no way he could be as awful as his numbers last season suggested (.151-6-20 with a .509 OPS). What I would prefer to do with Colvin is the following – I'd look at the players skills, and his total body of work. A review of his big league skill set shows the following. Here are some career numbers (837 plate appearances).

0.26 BB/K ratio
.237 AVG
.289 OBP
.471 SLG
.760 OPS
.273 BABIP
17.5 line drive rate
16.5 HR/F

For some perspective, here are the league averages since 2009 when Colvin began his career.

0.45 BB/K
.261 average
.329 OBP
.411 SLG
.740 OPS
.295 BABIP
19.0 line drive rate
9.0 HR/F

See what I'm saying with Colvin? He's not the worst hitter in the world like he was last year, and he's also not the elite performer he has been the past two weeks. In total, Colvin is a league average big league hitter, with solid pop, since he began his career.

Let's take the case of Brian McCann. A week ago he had nine homers and 35 RBIs and didn't look  like his former power hitting self. A 20 homer, 70 RBI bat in four of the past five seasons (he had “only” 18 homers in 2007), McCann was a huge disappointment. So what did he do last week? All he did was blast a homer in each of his last four games while knocking in 11 runners. All of a sudden he has 13 homers and 46 RBI, and now he's on pace for his customary 24 homers and 85 RBI after, literally, being on pace for 17 homers and 68 RBI just seven days ago.

As I'm fond of saying, these things tend to even out over the long haul. The problem is, no one can really quantify that. Is “the long haul” a month, two months, six months, a year? I'm fairly certain that if given enough time Albert Pujols will show himself to still be a .300 hitter. He was so awful though at the start of the season that he hit the All-Star break hitting .268. Does he have enough time to get his average back up to .300 in 2012? Maybe, maybe not. I'd lay even money though if you add his 2010, 2011 and 2012 seasons together that we're still going to find a guy who is a .300 hitter.

But that's the problem in fantasy sports – we need/want instant gratification. If a guy slumps for three weeks we drop him. If he goes off for three weeks, think Bryan LaHair, we inflate that players value in our heads to make him something he isn't. The truth of the matter is that baseball is a 162 game season. Even the best players often go into funks for weeks. It's just the nature of the game. My best advice is to take a holistic approach. Don't get so caught up in what is happening right around you, and remember this – by the time you realize a player is hot, it's too late to grab him. Say you just noticed that Daniel Murphy is hitting .420 the past three weeks so you run out to the waiver-wire to grab him. Let's say he then hits .050 for you the next week. What the hell happened? Things just evened out, I mean, over the last four weeks Murphy still hit .310 in this scenario – you just got him at the tail end of an unsustainable run. The point being that if you end up chasing production you'll likely never get it. Target players with a track record of success and stable skills, or at least target players whose current production doesn't match the skills that they possess. It's really the only way to stay ahead of the curve.

The Last Calendar Year

Let me illustrate what I'm talking about here. Let's look at the last 365 days and see what some of the numbers are for some of those men on the diamond that have been disappointing first half players. If we extend out our sample size to the last 365 days (denoted by “365” in numbers below), are the players still disappointing?

Alex Gordon
2012: .274-5-27-48-3
365: .288-18-67-100-14

I won't say that Gordon hasn't been disappointing after a nearly 20/20 season with 100 runs last year, but over the past 365 days those numbers are still borderline special for an outfielder.

Eric Hosmer
2012: .231-9-39-36-8
365: .268-20-84-77-17

Hey, if a first baseman almost goes 20/20 we've got big time fantasy love for that guy, don't we? He's still been very productive the past calendar year, even if his numbers the first three months this season are poor.

Mike Napoli
2012: .228-12-30-37-0
365: .299-30-72-81-3

The batting average this year is dreadful, but if we go back 365 days he's still hitting .299 which would be .039 points better than his career average. This is a perfect example of a player's performance coming back to earth to match his skill set. Even if Napoli were to hit .230 the rest of this season he'd still end up with a batting average in 2011-12 that is still better than his career mark.

Dustin Pedroia
2012: .266-6-33-41-6
365: .300-17-81-89-16

Injuries have derailed his 2012 effort, but his year long marks are still elite.

Alexei Ramirez
2012: .266-2-44-32-10
365: .265-9-76-62-14

His 365 day marks are, guess what, pretty darn close to his 2008-2011 average of .279-17-71-75-12 aren't they? I know he was terrible to start off the 2012 season, but again, given a long enough outlook...

Jimmy Rollins
2012: .256-8-32-48-14
365: .262-16-60-84-25

Rollins hit 239 with an OPS of barely .600 the first two months of the season, and people were thinking his days as a top level shortstop were over. Well, he killed it in June (.303-6-16-20-4) to bring up his 365 day total to a darn near identical match for his actual 2011 numbers (.268-16-63-87-30). Imagine that.

Justin Upton
2012: .273-7-37-54-10
365: .277-23-79-105-17

When a 23 year old goes off for a .289 average, 31 homers, 88 RBIs, 105 runs and 21 steals, the expectations are about off the charts, and when that same guy throws up a crappy ass first half of a season as a 24 year old there are plenty of ticked off people. Going back 365 days it's not like Upton's numbers are overly impressive, but they still are close enough to last year's numbers, other than the loss of homers, that they would be well within the realm of expectations.

Delmon Young
2012: .271-10-37-27-0
365: .274-20-81-60-0

Do you know how many 20 homer seasons Young has in his career? One as he went deep 21 times in 2010. Do you know how many 80 RBI season Young has in his career? The answer is two (93 in 2007 and 112 in 2010). Maybe be hasn't been that bad after all the past 12 months.

Michael Young
2012: .270-3-35-38-2
365: .307-6-82-84-4

Young's effort thus far has been poor. There is no way around that. At the same time, his 12 month look shows him to have posted a line that is pretty darn close to his career line per 162 games. Take a look:

.302-16-88-97-8

The homers are certainly well off the pace, but the other numbers are close enough to be considered a reasonable possibility given his current skills and age. Don't overlook the fact that Young has failed to hit more than 12 homers in three of his last five seasons – power just isn't his game.

Maybe that player of yours that is “struggling” right now will be just fine if you open up the sample size that you're looking at when evaluating the player meaning that he might, just might, still have time to bring things back up to “expectations” in the second half of the season.

Ray Flowers can be heard daily on Sirius/XM Radio on The Fantasy Drive on Sirius 210 and XM 87, Monday through Friday, 5-8 PM EDT. Ray's baseball analysis can be found at BaseballGuys.com and his minute to minute musings can be located at the BaseballGuys' Twitter account Twitter account.