In this age of super-computer statistics, ERA remains one old-school stat that actually works pretty well. After all, Clayton Kershaw has four ERA titles. Felix Hernandez, David Price, Zack Greinke and Jake Peavy each have a couple. Before them, Pedro Martinez led his league five times, Randy Johnson and Greg Maddux four times, and Roger Clemens led a staggering seven times.

Of course, single-season ERA is also subject to the whims of the defense behind the pitcher, limiting hits with runners in scoring position, random luck and even unearned runs. So baseball history is also littered with obscure ERA champions. Remember Steve Ontiveros? The pitcher, not the third baseman. Allan Anderson won an ERA title for the Twins in 1988. Rick Honeycutt won an ERA title before a long career as a reliever. Sammy Stewart? Craig Swan? Buzz Capra? Sounds like a trio of silent-movie stars from 1922.

Kyle Hendricks and Aaron Sanchez, last year's ERA champs, weren't household names entering the season. Hendricks began the season as the fifth starter for the Chicago Cubs, coming off a 3.95 ERA in his first full year in the big leagues. Sanchez had a power arm but had performed much better in relief for the Toronto Blue Jays in 2015 than as a starter, and many believed he would end up in the bullpen. Hendricks posted a 2.13 ERA while Sanchez had a 3.00 mark.

There are two questions here: Will they be as good in 2017? Are they for real?

The answer to the first question is almost certainly "no." I checked out the previous 50 ERA champions. Only six had a lower ERA the following season, and one of those -- Josh Johnson -- made just nine starts. The other five were Maddux, Randy Johnson, Kershaw and Martinez twice. The average ERA increase was 0.51 for the National League pitchers and 1.08 for the American League guys.

What we really want to know, however, is whether Hendricks and Sanchez will remain top-echelon pitchers, even as the projection systems forecast major regression:

Hendricks

Steamer: 3.51 ERA

ZiPS: 3.21 ERA

PECOTA: 3.76 ERA

Sanchez

Steamer: 4.00 ERA

ZiPS: 3.35 ERA

PECOTA: 4.40 ERA

In Hendricks' case, the systems look at his entire career and sees a pitcher who had never been a big strikeout guy. But he did rank a solid 22nd among qualified starters. He doesn't throw hard and the rest of his stuff doesn't jump out in any classic sense, so how has he improved his strikeout rate from 14.6 percent as a rookie to 22.8 percent in 2016? Sure, his command and location are top shelf, and the Dartmouth grad has earned "The Professor" nickname, but Hendricks also excels at something called pitch tunneling.

Baseball Prospectus recently had an excellent package of stories on this idea. Greg Maddux used to talk about it: You want all your pitches to look as similar as possible for as long as possible. Think about how batters learn to hit: Through repetition of thousands and thousands of pitches, the brain learns to distinguish fastballs from curveballs from sliders and so on. Some brains are better at this than others. But if the brain can't tell what pitch is coming and where it will break, the pitcher has a huge advantage, even if he's not throwing 95 mph.

Maddux once told Thomas Boswell his goal was to "make all of my pitches look like a column of milk coming toward home plate." As Jeff Long, Jonathan Judge and Harry Pavlidis wrote at BP:

Greg Maddux is one of the founding fathers of the pitch tunnels concept, explained through his "column of milk" analogy in the above excerpt. Maddux's approach is typically what we think of when we think of "pitch tunnels": the ability to fine-tune pitches so that they all fly through a small window a certain distance from home plate. When these pitches eventually diverge, it's too late and the hitter is unable to adjust to variations in speed or movement. The result, the theory goes, is weak contact and the (more than) occasional whiff.

At his peak, Maddux would use his full repertoire to change the eye level of the batter, change speeds, and work both sides of the plate. His willingness to throw all of his pitches in any count and to any part of the plate was amplified by the fact that all of his pitches looked the same to the batter. Maddux's pitches didn't have to move a lot, because he already had won the battle before the ball even reached home plate.

This is one reason why he induces so much soft contact (he had the lowest hard-hit rate among starting pitchers last season, according to ESPN Stats & Info). "In effect," the BP authors wrote in an in-depth breakdown, "Hendricks' lackluster stuff plays to his advantage, allowing him to generate weak contact because hitters are only slightly miscalculating when swinging at his offerings."

Is Hendricks the next Maddux? No, I'm not so foolish as to suggest that. Maddux actually had above-average fastball velocity early in his career while Hendricks' velocity is well below average. Hendricks did benefit last season from good defense behind him and he was especially tough with runners in scoring position (.184 average allowed). I would expect regression, but I wouldn't be surprised to see him post more seasons with an ERA under 3.00.

Sanchez is a more conventional power pitcher who averaged 94.7 mph on his fastball -- a two-seamer that he threw 74 percent of the time. The projection systems see a so-so strikeout rate -- he ranked 43rd out of 74 qualified starters -- along with inconsistent results in the minors, and don't believe he can maintain such a low batting average against (.224). What the projection systems don't see is what the eyes tell you: That's one nasty sinker.

Even though he threw it three-quarters of the time -- only Bartolo Colon threw his fastball more often among starters -- hitters did little damage against it. Sanchez was eighth in wOBA allowed on his fastball among qualified starters and sixth in home run percentage. Hitters have trouble elevating it, so it limits their power, even if they aren't striking out.

Moving forward, it will be interesting to see how Sanchez adapts. He's young and strong, but most pitchers eventually lose velocity, and we don't know how he'll respond to throwing 190 innings year after year. As good as that power sinker is right now and as good as his curveball is as a wipeout offering (.159 average allowed, 45 percent strikeout rate), he'll likely need to continue refining his changeup as a strong third offering.

Few pitchers surprised me in recent seasons as much as Hendricks and Sanchez in 2016. One showed how to win with command and below-average stuff and one showed why you always give a kid with a great arm the opportunity to start until he proves he can't. I think both will outperform the projections in 2017.