I admit it -- I was never a Belieber. At least not when it came to Shane.
Shane Bieber was on my radar plenty as a prospect; you couldn't miss those walk rates as he never walked more than 2.5% of opposing batters at any level until he got to Triple-A, where his control went completely off the cliff and he walked all of ... uh, 3.4% of batters he faced. He was always working with fairly average stuff, though, and never missed all that many bats, which was partly evidenced by strikeout rates that were good-not-great, usually below one per inning. He was homer-prone, however, which is unsurprising for a guy who throws a ton of strikes and doesn't have a plus pitch.
This year, however, Bieber has been a different pitcher thanks to tweaks to several of his pitches. He's still homer-prone, but he's missing so many more bats that he's minimized the damage. He throws harder now than he did in college, and the velocity was there in his unremarkable debut last year. He tweaked the grip on his changeup, getting five more inches of horizontal break on the pitch this year, but his slider was his most effective pitch in 2019 -- 43% of his swings and misses this year came on his slider. Most important, I think, is that he cut the use of his four-seamer, so that he's now throwing his three secondaries a majority of the time. He gave up 18 of his 30 homers this year on the four-seamer, meaning he gave up 60% of his homers on a pitch he threw 44% of the time, so using it less is a smart move.
Bieber is fifth in the American League in fWAR as I write this and seventh in Baseball-Reference.com's WAR, so he has undeniably been an ace this season. If his slider, which is actually breaking a little less this year than in 2018, really is this kind of swing-and-miss pitch, I see he no reason he can't be one for several years to come.
So he leads my annual column of players I was wrong about -- always on the low side, as it's no fun for anyone to talk about players I thought would be good but ended up flaming out, especially not for those players. You can see previous versions of this column for 2018 | 2017 | 2016 | 2015 | 2014 | 2013 | 2012.

Matthew Boyd, RHP, Detroit Tigers. Speaking of homer-prone prospects, when Boyd was a fringe prospect in the Blue Jays' system, he didn't even project as so much as a fifth starter; he was a control guy without a clearly average pitch, and while he wasn't homer-prone in the minors, it looked very likely he would become so in the majors. He was: 17 homers in 51 innings in 2015, 17 homers again in 92 innings in 2016. Since 2017, however, he's redesigned his slider, improving its vertical break by more than 15 inches between 2017 and 2019, and it's a legitimate out pitch, with pitch values of 30 runs above average total since the start of 2018. He's still homer-prone, and none of his other pitches grades out as average, but he's a viable major-league starter now, whereas when he was still a Jays prospect, I didn't think he would become one.

Jeff McNeil, IF/OF, New York Mets. McNeil never made any of my prospect rankings, as he neither performed well as a young pro player nor showed any plus tools. By the time he started to produce, he was 26 years old in Double-A last year, way too old for those stats to be meaningful predictors, and then he continued to hit in the hitters' paradise of Las Vegas in Triple-A. A small shift in where he starts his hands unlocked some more power, enough that he's made consistently hard contact since his call-up just over a year ago. One statistical note I find interesting, although it could merely be a sample-size issue, is that since this year's All-Star break he's had a totally different profile as a hitter, losing 126 points of BABIP from the first half but adding almost that quantity of isolated power, with 14 homers in 50 games and a .267/.352/.550 line. I'll be very curious to see how McNeil, who had never showed even average power prior to 2018, balances his natural ability to put the ball in play with this newfound power.

Luis Castillo, RHP, Cincinnati Reds. I really didn't think Castillo could be an effective starter, let alone a No. 2, without a viable breaking ball -- and his arm slot will always make it difficult for him to throw one. He's done so despite the lack of that average third pitch because his fastball and changeup are so good, and he only uses his below-average slider about 17% of the time, and just under 10% of the time when he's facing left-handed batters. He's also been much more effective in 2019 at keeping the ball in the park thanks to a much-improved ground ball rate, something he does with both his fastball and his plus-plus changeup, which Fangraphs has as the most valuable changeup in baseball this year, a ridiculous 9 runs better than the next-best (Hyun-Jin Ryu's). Castillo doesn't even have above-average command or control, which I might have assumed was a requirement for a de facto two-pitch pitcher to become this good of a starter; if he ever develops that, he would be an ace. His success has made me more bullish on Padres starter Chris Paddack, who also has a plus fastball/70 changeup/45 breaking ball, with less sink but better control. (It doesn't work quite as well when the two pitches are a fastball and curveball or slider, because such pitchers nearly always show large platoon splits.)

Eugenio Suarez, 3B, Cincinnati Reds: I just have to add that I never in a million years thought Eugenio Suarez would come close to 50 homers. If you had told me when I saw him as a prospect with the Tigers that he'd get to 30 homers, I would have probably said you were being wildly optimistic. Even with the juiced ball and Cincinnati's home park, I wouldn't have guessed he'd get past 45. And I remember saying nice things about him back then, and when the Reds acquired him in a trade. He's just become two grades better as a player than I ever thought possible.