As I write this Prince Fielder is hitting .190, Yasiel Puig is hitting .235, Cliff Lee has a 1.56 WHIP, and Homer Bailey has an 8.16 ERA. Brandon Belt already has six home runs, Charlie Blackmon is hitting .426, and Alfredo Simon has a 0.86 ERA and 0.81 WHIP in 21 innings.
We’re nearing May and you’ve fought the itch to overreact to early season events like those mentioned above. Surely those statistics above are starting to mean something, right?
Here are the strategies I try to apply early in the season in order to avoid making a mistake that could cause damage to my team:
Understand the Effect of Small Sample Sizes
What does it mean that Prince Fielder is only hitting .190 three weeks into the season? He has 12 hits in 63 at bats. Do you know what effect three bloop singles would have on his average? Or maybe he has hit three line drives that ended up being snagged by great defensive plays? Well if Fielder happened to have 15 hits on 63 at bats he’d be hitting .238. Not that it’s that much better. But it is nearly 50 points higher! Three hits. 50 points! If he was hitting .238 instead of .190 would you really even be that worried?
These are the sample sizes we are dealing with. Very small variances in early season performance can look bad. But we must take these early season statistics with a grain of salt. A struggling player is often only a handful of bloop singles away from career norms or a hot week away from carrying your team batting average.
This same principle applies to pitchers. A pitcher with 25 innings pitched and 15 earned runs allowed would have an ERA of 5.40. That pitcher is a three run home run away from a 4.32 ERA. By that same line of thinking, David Price could be one gust of wind away from a low 3.00 ERA (currently 4.39) or Stephen Strasburg could be a couple of pitches from a mid 4.00 ERA (currently at 6.00).
This is not to say that these guys’ home runs allowed have been unlucky (Strasburg gave up a BOMB to Giancarlo Stanton), just to illustrate some pitchers are literally only two bad pitches from ERAs two runs lower.
Let Preseason Projections Still Guide Decisions
The warnings of small sample sizes above should have you asking, “If I can’t trust early season statistics, then what can I use?”. And the answer is still preseason projections.
They are simply the BEST resource we have to estimating what a player will do this year. Many projection systems use three full seasons of data in their calculations. So if a projection system requires three years of data for optimal accuracy, you sure as heck better not be making major roster decisions based on just three weeks of statistics.
Be Mindful of the Advice You Come Across
Earlier in the week I came across some fantasy advice that essentially said, “I don’t trust Brandon Belt‘s hot start. He has a strikeout rate of over 30% and he has a walk rate of only 2%. His numbers are going to plummet because of his poor plate discipline”.
So here we have someone using more advanced metrics who sounds like they know what they’re talking about. And I even happen to greatly respect the work of this fantasy expert. But in the four games since Belt has struck out 3 times and walked twice in 18 plate appearances. This was enough to change his strikeout rate to 26.8% and raise his walk rate to 4.2%. Not overwhelmingly great numbers, but you can see how close he is to career averages of 9.8% walk rate and 23.2% strikeout rate.
Even seemingly great sounding analysis that uses more advanced metrics and component stats needs to be used carefully this early in the season.
Monitor Changes in Situation
Many great projection models use the underlying metrics like strikeout rate, walk rate, and batted ball profiles. So if we can’t rely on those component stats this early in the season, what kind of changes can we be on the look out for?
Instead of spending a lot of time early in the season digging through advanced metrics, I think you should be focusing on purely monitoring baseball news, playing time battles, and batting lineups. Playing time can be fickle too. But it’s my belief that playing time is the most likely variable to launch a player into fantasy relevance or fantasy stardom. And it’s also pretty easy to understand.
We don’t need to be expert sabermatricians to know that Abraham Almonte has batted leadoff for the Mariners 16 straight games to start the season. Or that Andrelton Simmons recently batted lead off for the Braves after starting the season batting eighth.
Those kind of things matter and can significantly affect player value. And all it takes is a little bit of hard work to separate yourself from others in your league.
Prey on the Weaker Minded
So now that we’re more comfortable with trusting our projections and can exercise patience, we can really put the screws to the rest of the league by waiting for other people to make mistakes. If some faint-of-heart owner drops Homer Bailey and his 8.00+ ERA, I’m all over it. I’m out on the trade market looking to see f I can acquire a struggling Bryce Harper, Robinson Cano, or Prince Fielder at a 10% discount.
Maybe I’m even quoting or linking to seemingly sound fantasy advice, like that mentioned above, that I realize is just falling victim to the same sample size issues as batting average.
Ask Myself, “What if This Was August?”
One of my favorite approaches to use when making roster decisions in April is to simply ask myself, “What if this was August?”. If Mark Buehrle put together three straight dominating outings in August, would I even care? Or better yet, “WOULD I EVEN NOTICE?”.
If a formerly unknown player hit .400 and had 4 HR over three weeks in August, would I be dropping a player with a history of proven performance? No way. When there are already four months of at bats to hide the effect of a hot 50 ABs, things don’t stand out so much. So why do we want to let the same batch of 50 ABs overweight projections that are often founded on averages of the last three years of data?
I’m not promising that Prince Fielder is going to end up with the 35 HR and 115 RBI that many expected in the preseason. But just like in Poker, the way to winning in fantasy baseball is through making decisions that increase the odds in your favor by even small margins.
Over time these seemingly small decisions add up and give you an edge over the competition. The decision to stick with and even try to acquire players struggling in April is often advantageous.