This is the best pitcher you don't know about

It's OK if you don't know about Taylor Rogers, because up until about May or so, there was really no reason to. It's OK if you aren't sure if we're talking about Taylor Motter, Christopher Austin, Tyler Kinley or Tyler Duffey, all members of the 2018 Twins, or if you hadn't realized that the 103-loss 2016 Twins, who had a 4.94 bullpen ERA -- the fourth-worst that year -- may have been actually hiding three future relief stars.

Through the end of May, Rogers had a 5.48 ERA as a nondescript middle reliever for the Twins, and a 3.84 ERA in 149 generally forgettable games over parts of three seasons. He'd collected zero professional saves, not that that matters very much. You didn't notice him. Why would you?

Obviously, there's going to be more to this story. Fast forward to the end of the year, and look past the traditional measures of saves, wins and even ERA. When we did, we noticed something incredible. By one powerful Statcast™ measure, Rogers was one of the 10 most effective pitchers in the game in 2018.

274 pitchers faced 250 hitters this year. The Top 10 in Expected wOBA are "9 awesome guys and whoa, I need to pay attention to Taylor Rogers, don't I?"https://t.co/iW1kOkyd4Y

1 Diaz
2 Treinen
3 Ottavino
4 Hader
5 Sale
6 Betances
7 Verlander
8 Pressly
9 deGrom
10 Rogers ?!

— Mike Petriello (@mike_petriello) October 29, 2018

It's true. That's a list of pitchers sorted by Expected Weighted On-Base Average; while you can and should read about how it works, the short version of it is that is measures a pitcher's performance in terms of quality of contact (i.e., exit velocity and launch angle) and amount of contact, including strikeouts, walks, and hit-by-pitches, all independent of ballpark or defense, and puts it on a scale similar to on-base percentage.

The Major League average for all pitchers was .311. For relievers, it was .305. For Rogers, it was .245.

Most importantly, just look at the quality of the names on that list. Jacob deGrom won the National League Cy Young Award, and Chris Sale and Justin Verlander each had strong cases to do the same in the AL. Edwin Diaz, Blake Treinen, Josh Hader, Dellin Betances and Adam Ottavino were, by any measure, among the most dominant relievers in the game. So was Thomas Pressly. We'll get back to him.

So, also, is Rogers, just ahead of Max Scherzer and Wade Davis, of all people. You may not have even known what team he was on before a few moments ago. What's going on here?

A midseason breakout
When we used "through the end of May" as a breakpoint above, that wasn't an arbitrary figure. On May 16, Rogers introduced his new slider for the first time, getting Tommy Pham to pop out. He didn't throw it again until May 31, the first game in which he threw it multiple times. From there on out, it became a regular part of his repertoire, as Rogers threw it 20 percent of the time.

Now, realize what happened after May. Let's set aside the expected metrics we shared above and look at what actually happened. There were 266 pitchers who threw at least 40 innings from June 1 on; if we sort them by lowest wOBA allowed, you'll very much notice where Rogers ranks.

Lowest wOBA, from June 1, minimum 40 innings
.170 -- Treinen, A's
.191 -- Rogers, Twins (that's a line of .146/.206/.231)
.209 -- Sale, Red Sox
.211 -- Diaz, Mariners
.226 -- deGrom, Mets

Those are four of the 10 or so best pitchers in baseball. There, also, is Rogers. Let's simplify this even further, into standard ERA. Check out what Rogers' monthly ERA marks looked like.

April: 8.44
May: 2.53
June: 1.69
July: 4.09
August: 0.00
September: 0.00

Another way of saying that is that on July 28, Rogers allowed two earned runs to the Red Sox in Boston. He appeared in 28 more games for the rest of the season, and allowed zero earned runs. He struck out 29 and walked only three. He didn't start collecting saves because the Twins primarily used Trevor Hildenberger in that role after trading Fernando Rodney, but if Andrew Miller has taught us anything, it's that collecting saves does not define the quality of a reliever.

OK, so, Rogers ended up having a quietly great season, a shockingly good one for a 2012 11th-round pick who had risen through the Minnesota system as a relatively unheralded starter. (In 2015, MLB Pipeline described him as "the quintessential pitchability lefty," and that he was "on the brink of bringing his command and average stuff to a big league rotation near you.")

How did he pull that off? It can't just be that he found a new pitch, right? Let's call this a three-step process.

1) Rogers was already tough on lefties.
In 2014, as a starter at Double-A New Britain, Rogers allowed lefties only a .555 OPS. (Righties had a .703 mark.)

In 2015, as a starter at Triple-A Rochester, Rogers dominated lefties, holding them to a .177/.209/.193 line (.402 OPS), while righties crushed him: .326/.374/.457 (.831 OPS).

In his first two seasons in the big leagues, it was more of the same. Lefties hit .186/.257/.296 (.553 OPS) against him; righties hit .288/.355/.435 (.790). Start with that, because Rogers had brought a strong skill to the bigs -- but also a weakness that, if unresolved, would limit him going forward.

2) The slider solved his platoon problems.
Here's how Rogers had attacked righties during his career through the end of May: 60 percent sinkers, 25 percent curveballs, 15 percent four-seam fastballs, changeups, and sliders.

Here's how he attacked them from June 1 on: 40 percent sinkers, 30 percent sliders, 25 percent curveballs, five percent four-seams.

All of a sudden, Rogers had a weapon against righties, who hit just .220/.267/.377 against him in 2018, and only .174/.216/.322 after June 1.

3) He used his high-spin curve more effectively.

Pitching is hardly entirely about spin rate, but just about the most 2018-19 thing you can do in baseball right now is to identify a pitcher with the raw talent of a high spin rate and help them to use it more effectively.

Last year, there were 69 lefty pitchers to throw at least 100 curves. Rogers had the seventh-highest spin rate, and if we'd made the minimum 200, he'd have been third behind only Dodgers lefties Rich Hill and Caleb Ferguson. It's a skill that smart teams are beginning to be able to shape into effectiveness on the mound.

In Rogers' case, his curve has been more of a side-to-side sweeper. Last year, 236 pitchers threw at least 100 curves, and only seven got more horizontal movement than Rogers did -- names you want to be associated with, like Hill, Charlie Morton, Aaron Nola, and Pressly. But he also added vertical drop as well, reportedly after working with former bullpen coach Eddie Guardado.

When we looked at 2018's most dominant pitches last week, there was Rogers's curve, high atop the list. Hitters had just a .136 average and a .152 slugging against it.

Now, if this all sounds familiar, it should. We've spent all this time talking about how in 2018, the Twins had a previously nondescript middle reliever take a huge step forward to become one of the better bullpen arms in the game, thanks in part to changing his pitch usage to emphasize breaking balls and to better utilize a high-spin curveball.

The thing is, that also applies to Pressly, who, after several decent years in the Twins bullpen, started strong for Minnesota and closed the season with a run of dominance upon being traded to the Astros in July, and is perfectly described by all of the above.

The 2019 Twins could have had Pressly, Rogers and Trevor May, who had a 36-to-5 strikeout-to-walk ratio in 25 1/3 innings after returning from Tommy John surgery. The 2016 Twins did have that trio, and they combined for a 4.17 ERA that year. The best teams can take talented pitchers and make them better. We might be seeing the beginnings of that in Minnesota, though the Twins may regret having traded Pressly for a pair of prospects who aren't in the current MLB Pipeline Twins Top 10.

You may not have noticed Rogers in 2018, and it's perfectly understandable why that would have happened. It won't be the same in 2019, however. He's on everyone's radar now.