Predictive pitching stats key to DraftKings success

When it comes down to it, winning the Official Mini Fantasy Game of MLB.com from DraftKings is about making accurate predictions; you need to forecast player performances more accurately than the rest of the field. There are various statistics that can help in that process.

Not all stats are created equal, though. Some stats can explain past events really well. ERA is one such stat; it can give you a really strong idea of which pitchers have helped their teams win in the past. But there are better options out there to tell you which hurlers will continue to pitch at a high level. That's why we're searching for predictive stats -- those that best predict future performance. The more predictive a stat, the more useful it is in daily fantasy baseball.

That said, let's take a look at a few stats to use and avoid when selecting your pitchers in the Official Mini Fantasy Game of MLB.com from DraftKings.

Use WHIP
WHIP -- walks plus hits per inning pitched -- is a simple and common stat with which most casual fans are familiar. Although it isn't without its faults, a quick glance at a pitcher's WHIP will give you a decent idea of how many hits, walks and, ultimately, runs he can be expected to yield in the future.

Avoid ERA
Although ERA is the most common pitching stat that's available, it's one of the worst in terms of its predictive power. Here's the problem: Although runs are clearly important, they're also pretty volatile; we see pitchers give up many hits and few runs quite often. While the narrative is often that "he was able to scatter X hits" or "he dug down deep to work through trouble," the pitcher really just got lucky. Over the long run, WHIP is going to predict future earned runs much more accurately than ERA.

Avoid home runs allowed
Giving up home runs is obviously devastating to a pitcher's fantasy production, but because home runs are such low-frequency events, it's really difficult to use past home run rate allowed to predict future home runs allowed.

Use fly-ball rate
Instead of looking at home run rate, examine how many fly balls a pitcher yields. Because fly balls occur at a much greater rate than home runs, there isn't as much variance in the totals. Over the long run, most pitchers allow a home run on right around 10 percent of their fly balls.

If you're examining a pitcher who has allowed 100 fly balls on the season, the most likely result is around 10 home runs. If he's given up significantly more long balls than that, chances are he's been pretty unlucky and should improve his performance in the future from luck alone.

Use K/9 rate
Finally, look at a pitcher's strikeouts per nine innings as a predictor of future strikeouts. The beauty of K/9 rate is in its simplicity. Unlike ERA, K/9 rate isn't significantly affected by the ballpark in which a pitcher is playing or other similar variables.

Plus, K/9 rate is one of the most consistent stats in baseball; pitchers who have struck out many batters in the past are probably going to continue to do it in the future. When selecting pitchers for the Official Mini Fantasy Game of MLB.com from DraftKings, identifying candidates for high-strikeout games is critical.