In my first DFS strategy article of the year, I took a macro view in discussing my process for creating MLB DFS lineups. I am shifting gears a bit this week to discuss the importance of putting the statistics we use into their correct context.

On Sunday, DFS_Fanatic asked for the best way to utilize the hitting and pitching coach tools. Ben Scherr and I had pretty similar responses, but you could probably distill our advice to one sentence: Take the stats in the hitting and pitching coaches and put them into context. That could mean looking for stats that support players you were already considering, or finding a stat to serve as a tiebreaker. As I mentioned in my previous strategy article, I like to use the hitting and pitching coaches in part to see if there are any players I initially dismissed who are worth a second look. In all of those cases, what takes a stat from simply being interesting or eyebrow-raising and makes it actually useful is context.

One of the problems with providing fantasy advice on Twitter is it is often difficult to provide any sort of context in 140 characters. I see a lot of stats thrown around to support various DFS plays that are dubious, at best, and I feel it is important to discuss a few instances in which understanding the context of common stats can make the difference between a smart investment and simply throwing your money away.

The seed for this article was planted back in April when I read a tweet recommending some left-handed hitter against R.A. Dickey because this hitter was doing really well against right-handed pitchers. I nearly threw my phone out the window. I don’t know if the hitter in question was a strong DFS play against R.A. Dickey (I can’t remember the hitter or the person who tweeted the stat, which is probably just as well), but I do know playing that guy against Dickey because of his track record against other righties is silly at best.

The reason to use righty/lefty splits, at the most basic level, is that most same-handed pitchers are similar, so previous success against righties is a reasonable indicator of future success against righties. R.A. Dickey isn’t like most righties. For one thing, he doesn’t throw a breaking ball. In large part due to that fact, he has nearly identical career righty/lefty splits. Lefties are batting .244/.319/.418 against him while righties are at .260/.319/.421. He has allowed a .322 wOBA to lefties and a .323 wOBA against righties.

Perhaps more importantly, hitting a knuckleball is a different skill than hitting any other kind of pitch. For this reason, I rely heavily on batter vs. pitcher data when it comes to knuckleballers. I know guys like R.A. Dickey and Steven Wright are capable of giving up a ton of walks and home runs on any given day, but I don’t think it is worth it to take that chance unless we can point to players who have actually been good against them in the past.

I don’t like to use BvP stats too heavily in other situations because, again, context matters. To use an extreme example, Alcides Escobar is batting .186/.212/.236 this season. He has a career-high strikeout rate and career-lows in walk rate and isolated power. I’m probably not playing Alcides Escobar for the foreseeable future, even if he keeps leading off. I almost don’t care that Escobar is a career .360 hitter against Mike Pelfrey. Escobar was a better hitter when he earned those stats than he is right now. Let me put it another way: If Alcides Escobar got 25 at-bats this week against Mike Pelfrey, would we really expect him to bat .360? I wouldn’t.

The last set of stats I want to try to put into context are streaks and recent stats. These are tricky in DFS because even if you believe a player is hot, you have to pay more for his services. And like anything else, recent stats have to be put into their proper context. Gerardo Parra is 7-for-12 over the last six days going into Thursday’s games. That may be a good reason to use him, except all of those at-bats came at Coors Field. Go ahead and play Parra if you want, but I wouldn’t do so because he is on a hot streak. His recent performance could have more to do with his home ballpark than anything else.

Another reason to potentially be wary of a hot streak is the handedness of the pitchers. If Evan Gattis goes on a hot streak because the Astros have faced a bunch of lefties, that doesn’t tell us a whole lot, especially if his next game is against a righty.

Hopefully by now I have convinced you of the importance of considering stats within their larger context. Fortunately, you don’t always have to do this yourself. The writers here are FantasyAlarm are great at determining what is news and what is noise, and I try to be cognizant of context in my own writing as well. But before you see a random stat in the world and use it to tweak your DFS lineup, I hope you will take a moment to consider the context.