There are a ton of variables you can factor into a hitter’s projection on a daily basis. Some are broad categories like whether the park is hitter-friendly or pitcher-friendly or how well the opposing starting pitcher fares against hitters of a certain handedness. These broad categories tend to get the most attention because they are derived from large sample sizes and thus their impact is more measurable.

From there you can drill down towards minutiae and look at things like how hitters fare against certain pitch types that the opposing pitcher tends to throw. Or you can simply look at how a hitter has performed in the past against a specific pitcher. You can also look at a hitter’s recent performance. But the farther down the rabbit hole you go, the harder it is to be sure of a variable’s impact.

But one rather large factor that doesn’t get enough attention is the quality of the opposing bullpen. Most analysis focuses on the opposing start pitcher, but relievers have faced 37.5 percent of all batters that have come to the plate this season. If you’re a little slow, that’s more than a third of all plate appearances that have nothing to do with the starting pitchers we tend to focus so much analysis on.

Each year the percentage of plate appearances that are against reliever grows. In 2014, the percentage of all batters faced by relievers was 33.6 percent. It then grew to 35.1 percent in 2015 and to 36.7 percent last year. It’s no secret that specialization has become a bigger part of the game, but it’s worth noting that the trend has not yet plateaued.

Figuring out how best to pick on bullpens is a two-step process. The first step is obviously identifying which bullpens are bad. But second, you have to determine which bullpens are likely to see the most work on a given day.

Figuring out which bullpens are bad is relatively simple. The Twins and Tigers are two of four teams whose bullpen has an ERA higher than 5.00, and Minnesota and Detroit have the two worst bullpen xFIPs in the league. The Nationals and A’s also have bullpen ERA’s above 5.00, but their xFIPs indicate that over the long haul those bullpens could trend toward average.

Other bullpens with a toxic mix of a bad ERA and a bad xFIP are the Mets, Phillies and Cardinals, and, to a lesser extent, the Rangers and the Brewers. The bullpens for the Marlins, Giants and Rays have been somewhere near average in terms of ERA so far, but their xFIPs indicate some regression should be forthcoming.

Bullpens you should try to avoid are Cleveland, Los Angeles (NL), New York (AL) and Boston. A cut below those four teams are both Chicago teams, Arizona, Los Angeles (AL) and Toronto with Houston also being in the mix to avoid due to a great xFIP.

Now we turn to how much work a bullpen has to do. At this point, a bad starting pitcher can add to the appeal of the opposing lineup if he’s likely to get chased quickly and the bullpen behind him is bad. You could look at season long averages to see which teams use their bullpens most heavily and which lineups chase starters the earliest, but you’re better off going day-to-day for this step.

One quick and easy way to determine which bullpens are most likely to see the most work on a given slate is to look at our projections for starting pitchers. You can sort by innings pitched and voila…you’ve got an idea about which bullpens should face the most batters for that slate.

Using Thursday’s short slate as an example, you’ll see Rich Hill with the shortest projected start, which is not a plus for the opposing Indians given that LA’s bullpen is one of the best in the league. The first bad bullpen we run into as we go down the projected innings list in ascending order is Milwaukee’s.

Kyle Davies will start for the Brewers against the Cardinals, and he’s already an attractive option for St. Louis hitters given his matching 4.74 ERA and xFIP and his 1.45 HR/9. Davies doesn’t have extreme splits either this year or his career, so he makes for a nice stacking option since hitters from both sides can be used against him.

Other bad bullpens who will be pitching behind a starter with a below average number of innings projected tomorrow are the Phillies and the Giants. Nick Pivetta will start for the Phils, and the bad bullpen only makes the matchup all that more appealing for the opposing Red Sox. And the struggling Matt Moore will pitch for the Giants in Colorado. As if you needed another reason to play Coors…

Look, this isn’t a revolutionary topic we’re discussing here today. The short version of this article: is play hitters who will face bad pitchers. We’re just reminding you that starters aren’t the only pitchers to consider when you’re looking for bad ones to roster hitters against.