I’ve discussed the importance of paying for top pitching quite a bit this season. In both 50/50s and GPPs last year, winning teams paid more (a few hundred dollars, on average) for pitching than teams that didn’t cash. Top pitchers are more consistent and have higher upside than middle and bottom-tier pitchers.

However, I think it’s important to make sure you don’t pigeonhole yourself into any particular no-matter-what sort of strategy. When in doubt, you should use general heuristics based on long-term data, but there are many times when heuristics don’t apply or when there’s enough contradicting evidence to go in the opposite direction. So “pay for pitching” is a good rule of thumb, but not an everlasting doctrine.

If you’re blindly selecting mid- and bottom-tier pitchers, though, you’re going to be in trouble. Here are some traits that I emphasize when I decide to save a little money on one (or sometimes both) of my arms.

 

HighK/9

A lot of the high-strikeout pitchers are among the most expensive, but there are always a handful of hurlers who sit down batters with regularity, but don’t otherwise have elite stuff (at least not consistently). Young pitchers who make batters miss but struggle with control at times, for example, can give you strikeout upside at a cheap price.

The good part about K/9 is that it’s extremely consistent. With a high-strikeout pitcher, your job is less about figuring out if he’ll whiff hitters—he usually will for as long as he’s in the game—and more about determining if he can rack up enough innings to realize his potential.

I almost never roster a cheap pitcher if he doesn’t have high upside in terms of Ks.

 

Extreme Splits (GB/FB, L/R)

As mentioned, platoon splits are an easy and effective way to find value. When a player has extreme splits in either direction, his matchups probably aren’t completely priced into his salary. A southpaw who is nasty versus left-handed bats and will be facing a primarily left-handed lineup, for example, will often offer value.

In addition to lefty/righty splits, I also look at batted ball profiles. Remember, ground ball hitters struggle mightily against ground ball pitchers. GB/FB rates are highly unlikely to be factored into pricing on daily fantasy sites, so you can sometimes get a deal on a mid-tier pitcher with a favorable matchup in terms of batted ball profiles.

 

An Unavoidable Stack

I usually begin my lineup creation with pitchers, then move to my favorite stack(s), and then fill in the value bats. But sometimes there’s a particular offensive stack that I really want to target, in which case I’ll build around those players.

If that stack is one that costs a pretty penny, I’ll occasionally try extra hard to find an affordable arm so that I can get the bats I want into my lineup. That’s particularly true when an expensive stack—like the Blue Jays, Tigers, Rockies, etc—is in a perceived poor matchup. If I can get upside without high ownership, I’m going to take it.

 

Start of Season

Daily fantasy pricing tends to fluctuate more at the start of the season than any other time; people overreact to a small sample size of data, and thus you’re more likely to find a quality arm at an affordable price in May than you are in September.

This is where understanding randomness comes into play. I use the same stats as with my batters—HR/FB, BABIP, and so on—to determine how “lucky” a pitcher has been. If we’re looking at a pitcher who has succeeded in the past but has a high HR/FB and BABIP to start the season (and preferably a SIERA that’s lower than his ERA), then there’s a good chance he has unnecessarily dropped in price.

 

Short Slates

During very short slates, you’re going to see massive usage on certain pitchers. If you’re semi-confident in a pitcher who you believe is going to fly under the radar, it can make sense to take a flier on a guy whose lowered usage will make up for his reduced probability of returning value. I still generally side with the top-priced arms in short slates, but I’d say I’m slightly more likely to be contrarian with at least one pitcher in a three-game slate than a 12-game slate.

 

Vegas

Finally, I’ll always take a second look at a player that Vegas really likes. If a mid-priced pitcher is -200 to win a game and has a favorable strikeout prop, I’ll generally trust that the wise guys in Vegas have it right.

DraftKings generally has astute pitcher pricing, but no one is as accurate as Vegas over the long run. If my own research matches what Vegas thinks, that’s even more incentive to bite the bullet and fade the most expensive pitchers.