What I Didn't Expect in Baseball in 2017

With less than a week until this year's record-early Opening Day, I'll be sharing my annual predictions for the upcoming baseball season pretty soon. But if last year's predictions are any indication, you should probably ignore them. (I'm really good at marketing, guys.) After looking back at how my political forecasts fared in 2017, I do the same for baseball below, and... it's not pretty. Come, let's all laugh at my terrible prediction skills together:

(You can read my full predictions for the 2017 American League season here; the 2017 National League season is here.)

Prediction: The AL division winners would be the Red Sox, Indians, and Astros; the Blue Jays and Rays would win the Wild Cards. In the NL, the Nationals, Cubs, and Dodgers would win their divisions, with the Mets and Cardinals nabbing Wild Cards.
What Really Happened: I got all the division winners right—although everyone else did, too (those six teams had it in the bag since virtually Opening Day)—but the Yankees, Twins, Diamondbacks, and Rockies won the Wild Cards. I had projected the Twins and D'backs for 68 and 74 wins, respectively, two of my worst predictions on the year. One of my predicted playoff teams, the Mets, instead lost 92 games. (Amazingly, though, I still nailed every other team in the NL East within two wins.) Overall, I got 13 teams within five games of their eventual win totals, and the average error of my predictions was 7.7 wins.

Prediction: Greg Bird and Aaron Judge would be the modern Mantle and Maris, going back and forth all season as the Yankees team home-run leader.
What Really Happened: This line allows me to take credit for calling Judge's breakout season, right? The towering right fielder obviously came close to winning the AL MVP with his 52 home runs. Bird, though, laid an egg: .190/.288/.422 with just nine homers.

Prediction: José Altuve would captivate America with a 50-game hitting streak.
What Really Happened: Altuve's longest hitting streak of the season was 19 games, but I doubt he was disappointed—he won the AL MVP and got to death-stare President Trump.

Prediction: Giancarlo Stanton would be the first Marlin ever to top 50 home runs, leading the National League.
What Really Happened: Stanton hit 59 homers, best in not only the Senior Circuit, but all of baseball.

Prediction: By the end of the season, Jeb Bush would be the proud new owner of the Marlins.
What Really Happened: Poor Jeb can't win anything, can he? The group headed by Bruce Sherman and Derek Jeter instead won the bidding and bought the team in August.

Prediction: Keon Broxton and Domingo Santana would strike out a combined 300 times, but both would get on base at .350 clips despite .250 batting averages. Broxton would hit 20 homers with 30 steals, and Santana would produce a mirror-image 30/20 season.
What Really Happened: I was pretty close. Broxton hit exactly 20 homers but stole "only" 21 bases. A bigger problem was his average and OBP: .220 and .299, respectively. Santana hit exactly 30 homers and stole "only" 15 bases. He hit .278 and got on base at a .371 clip.

Prediction: Bryce Harper, Adam Eaton, Trea Turner, Anthony Rendon, and Daniel Murphy would each be worth more than 4.0 WAR.
What Really Happened: Three surpassed that threshold, according to FanGraphs: Rendon (6.9), Harper (4.8), and Murphy (4.3). Turner wasn't too far behind at 3.0, but Eaton tore his ACL at the end of April, cutting off his season at 0.5 WAR.

Prediction: The Rockies rotation would be the best in club history.
What Really Happened: They posted 11.8 FanGraphs WAR (fifth in club history) and a 91 ERA− (second in club history). Jon Gray (3.2 WAR and a 73 ERA−) and Germán Márquez (2.4 and 87) led the way.

Prediction: Tigers manager Brad Ausmus would be fired in May. With the team hovering around .500 at the trade deadline, ownership would finally give the OK to blow it all up and rebuild.
What Really Happened: Ausmus held on the whole season, but he was quasi-fired in September when the Tigers announced they wouldn't renew his contract. The Tigers entered the July trade deadline at 47–56 and the August deadline at 58–74; it was at the second one when they finally fire-sold off Justin Upton and Justin Verlander.

Prediction: The San Francisco offense would post its worst offensive season since the 2011 squad's 91 OPS+.
What Really Happened: The 2017 Giants managed just an 83 OPS+, the worst in baseball and the Giants' worst score since 2009.

Prediction: Pablo Sandoval would scream back to relevance with 25 home runs and a positive number of Defensive Runs Saved.
What Really Happened: This is one of those predictions that makes you realize just how long ago March 2017 was. Sandoval hit .212/.269/.354 with the Red Sox, was released in July, returned to the Giants, and basically put up the same slash line for them. He hit nine home runs total with −7 DRS.

Prediction: Every Orioles starting pitcher except Wade Miley would give up more runs in 2017 than in 2016.
What Really Happened: They all did it—including Miley.

Prediction: Robbie Ray would take a huge step forward, shaving more than a run off his ERA and leading the league in strikeouts.
What Really Happened: Ray went from a 4.90 ERA to 2.89. Ray's 218 strikeouts were "only" good for third in the Senior Circuit (Max Scherzer led with 268), but Ray did lead in strikeouts per nine innings (12.11).

Prediction: Every Dodgers starting pitcher would miss at least eight starts as the injury bug plagued Los Angeles.
What Really Happened: Every Dodgers starting pitcher missed at least five starts. Their pitchers lost 1,051 days to the disabled list in total, and the entire roster led MLB with 1,914 days missed.

Prediction: Jean Segura would be a huge bust. Mitch Haniger would turn out to be the more valuable addition from the Taijuan Walker trade, even in the short term.
What Really Happened: Segura went from 5.0 WAR to 2.9 WAR—hardly a bust, and still better than Haniger. However, I was at least right that Haniger would distinguish himself right away: he accumulated 2.5 WAR and wOBA-ed .360.

Prediction: The Mets rotation would be fully healthy and dominant, getting 200 innings out of Noah Syndergaard, a sub-3.00 ERA from Steven Matz, and even a respectable year out of Matt Harvey.
What Really Happened: A wonky elbow held Matz to 13 starts with a 6.08 ERA. A torn lat muscle kept Syndergaard out for five months. And Harvey was lucky not to be non-tendered after his 6.70 ERA performance.

Prediction: Jordan Zimmermann would rue signing with Detroit as he became a pure contact pitcher (setting a career low in strikeout percentage) but the Tigers' league-worst defense failed to convert them into outs.
What Really Happened: Exactly that. Zimmermann's 14.5% strikeout percentage was not only a career low, but it was also fifth-worst among all MLB pitchers with at least 160 innings pitched. As a result of the Tigers' AL-worst −69 DRS, Zimmermann mustered just a 6.08 ERA.

Prediction: Two former Rangers prospects would experience resurgences. Jurickson Profar would win the batting title, and Delino DeShields Jr. would sport a .350 OBP and 30 stolen bases.
What Really Happened: Whoops—Profar hit .172 in only 58 at-bats. But DeShields came eerily close to my projections: his OBP was .347, and he stole 29 bases.

Prediction: Greg Holland wouldn't notch a save all season.
What Really Happened: He led the NL in them with 41.

Prediction: José Berríos and Byron Buxton would finally live up to their potential—which would be good for the Twins, since Brian Dozier would hit just .210 with 10 home runs and nearly 200 strikeouts.
What Really Happened: Berríos went from walking nearly as many as he struck out in 2016 to 14–8 with a 3.89 ERA. Buxton hit a decent .728 OPS but, thanks to stellar defense, amassed 3.5 WAR. Dozier, though, was his usual excellent self, slashing .271/.359/.498 with 141 strikeouts and only mild regression in the homer department (34).

Prediction: Wade Davis would struggle with his control in his recovery from injury, and Kyle Hendricks would regress to league average.
What Really Happened: Davis did indeed walk a career-high 11.6% of batters he faced; I'm nervous for him in 2018. Hendricks regressed from a 2.13 ERA to 3.03, but that was still good for an ERA+ of 144.

Prediction: I called Dansby Swanson winning NL Rookie of the Year "the safest prediction on this page." However, I did expect Cody Bellinger to "force himself into the lineup in mid-siummer."
What Really Happened: Swanson didn't even get Rookie of the Year votes, as he finished with just 0.1 WAR. Bellinger came up on April 25 and didn't look back, collecting 39 home runs and the ROY trophy.

Prediction: The Rangers would have a losing record in one-run games, and they would lead the AL in days spent on the DL.
What Really Happened: Texas did indeed go 13–24 in one-run games, the worst mark in baseball. Their players spent an above-average 1,271 days on the DL, but the Rays led the AL with 1,644.

Prediction: Jason Heyward would bounce back with a .290/.350/.450 slash line, 20 DRS, and a 5.0 WAR. Ben Zobrist, on the other hand, would run into a brick wall. His modest value with the bat would be offset by the worst defensive season of his career.
What Really Happened: Heyward was only marginally better than his disappointing 2016 with the bat (.259/.326/.389), and while he did put up 18 DRS, UZR was much less kind to him. As a result, he had a FanGraphs WAR of just 0.9. Zobrist joined him in the Cubs' trash heap, though not for the reasons I foresaw. He was putrid at the plate (.232/.318/.375) but maintained a (barely) positive defensive value (1 DRS, 1.7 Fielding Runs Above Average).

Prediction: Jorge Soler would lead Royals position players in WAR.
What Really Happened: At −1.0, he was instead dead last.

Prediction: Sonny Gray would continue to be a 75 ERA+ pitcher, Jharel Cotton would pitch 160 innings with a 3.40 ERA, and All-Star Sean Manaea would be the first of the 2017 season to throw a no-hitter.
What Really Happened: Gray returned to form with a 123 ERA+, good enough to be traded to the Yankees. Cotton stunk up the joint to the tune of a 5.58 ERA in 129 innings. Manaea was better (a 4.37 ERA) but no All-Star—although he did throw five no-hit innings in just his third start of the season. (He was removed with the no-hitter intact because he had already thrown 98 pitches.)

Prediction: Jay Bruce would finally win over Mets fans by giving them a .750 OPS, while José Reyes would be banished from Flushing for good by the end of the season.
What Really Happened: Bruce gave Mets fans a .841 OPS, earning a trade to Cleveland, but New York welcomed him back as a free agent in January. Despite being far worse (a .315 OBP), Reyes was allowed to bat 501 times for the Mets, and he too was re-signed in January.

Prediction: The Dodgers would have the NL's stingiest bullpen, followed by the Marlins.
What Really Happened: The Dodgers did rank first in the NL in bullpen ERA (3.38), but the Marlins ranked 10th (4.40).

Prediction: Félix Hernández would post a career-low strikeout rate and flirt with his career-high ERA of 4.52. However, Drew Smyly would make up for it with a 3.20 ERA. James Paxton would finally pitch to his 2.80 FIP.
What Really Happened: Hernández's 21.2% strikeout rate wasn't the lowest of his career, but his ERA did soar to 4.36. Smyly didn't pitch an inning all season, going under the knife in June. Paxton did indeed have that breakout season, posting a 2.98 ERA, but it still didn't catch up to his FIP, which was an outstanding 2.61.

Prediction: David Dahl would surpass outfield-mates Carlos González and Charlie Blackmon in WAR.
What Really Happened: Dahl never played in the majors all year long, so obviously he failed to do so. It would've been easy to beat out González (−0.2 WAR), but Blackmon got MVP votes with his 6.5 score.

Prediction: Steven Souza would finally have that 20/20 breakout season, and Colby Rasmus and Matt Duffy would match their career-high WARs for the Rays.
What Really Happened: Rasmus "stepped away from baseball" halfway through the year, and Duffy never even played in the majors. Souza did break out, but not in quite so balanced a proportion: he hit 30 home runs and stole 16 bases. Both were career highs, and he was traded to the Diamondbacks for his efforts.

Prediction: Jonathan Villar would hit just .240, and his runs scored and stolen bases would both be slashed in half from 2016.
What Really Happened: Villar hit .241. His runs scored went from 92 to 49, and his stolen bases went from 62 to 23.

Prediction: Chris Archer and Blake Snell would be a formidable two-headed monster at the front of the Rays rotation, but the AL Cy Young Award would go to KC's Danny Duffy (a 2.50 ERA and 240 strikeouts).
What Really Happened: Archer actually regressed from 2016 with a 4.07 ERA, and Snell was close behind at 4.04. Duffy was a little better, boasting a 3.81 ERA, but only punched out 130. None of the three received any Cy Young votes.

Prediction: Jameson Taillon's 2.50 ERA would put him squarely in the NL Cy Young conversation. However, Pirates teammate Jung Ho Kang's personal and legal problems would end his major-league career.
What Really Happened: Taillon instead took a huge step back with a 4.44 ERA, although a 3.48 FIP and .352 BABIP suggests he didn't deserve that fate. And so far so bad for Kang.

Prediction: José Bautista would bounce back so convincingly that he would be as valuable as his 2016 self and Edwin Encarnación combined.
What Really Happened: Bautista was awful. His −0.5 WAR was far worse than the 1.4 he accrued in 2016. Encarnación was worth 2.5 WAR in his first season in Cleveland.

What I Didn't Expect in Politics in 2017

It's snowing out and nothing else is really going on, so I'm taking care of some site housekeeping today. Every year, I make a certain number of predictions on these webpages, and every year I try to look back at how I did. This is the first of two posts on that subject—the one that will focus on politics.

Every fall, I issue race ratings, inspired by those at Inside Elections, for every downballot constitutional office up for election. For those types of elections (not so much for politics in general), 2017 was a pretty quiet year: only the Virginia lieutenant governor, Virginia attorney general, and Louisiana treasurer were on the ballot. Here were my ratings for those three races, originally issued in October and kept current (although they never changed) through November 6. They predicted a status quo election, with Democrats holding onto the two offices they already owned, and Republicans successfully defending their one seat.

The small number of races meant I had fewer opportunities to make a bone-headed mistake, and as a result the ratings validated quite nicely.

• Democrats won two out of two races I rated as Lean Democratic.
• Republicans won the one contest I rated as Solid Republican.

The final results are also pretty close to how one might express a "Lean Democratic" or "Solid Republican" race quantitatively. Here's the Democratic margin of victory or defeat in each of the three elections:

In a well-calibrated world, the Virginia average of D+6.2 is probably right on the border between Lean Democratic and Likely Democratic. Likewise, the Louisiana treasurer margin of R+11.5 is on the Solid side of Likely Republican. All in all, pretty close, though.

As I mentioned, getting these races right is no great achievement: last year offered a small number of fairly predictable races. The big challenge will be 2018; midterm cycles are the absolute busiest for downballot constitutional offices. My goal this year is to merely handicap all 142 of them before November, let alone get them all right. Wish me luck!

My Model Nailed This Year's Hall of Famers—The Vote Totals, Not So Much

About one month ago, Chipper Jones, Vladimir Guerrero, Jim Thome, and Trevor Hoffman were elected to the Baseball Hall of Fame, which means two things: (1) this Hall of Fame election post mortem is almost one month overdue, and (2) for the first time in three years, my forecasting model correctly predicted the entire Hall of Fame class.

You'd think that would be cause for satisfaction (and I suppose it is better than nothing), but instead I'm pretty disappointed with its performance. The reason is that an election model doesn't really try to peg winners per se; rather, it tries to predict final vote totals—in other words, numbers. And quantitatively, my model had an off year, especially compared to some of my peers in the Hall of Fame forecasting world.

First, a brief rundown of my methodology. My Hall of Fame projections are based on the public ballots dutifully collected and shared with the world by Ryan Thibodaux (and, this year, his team of interns); I extend my gratitude to them again for sacrificing so much of their time doing so. Based on the percentage of public ballots each player is on to date, I calculate his estimated percentage of private (i.e., as yet unshared) votes based on how much those two numbers have differed in past Hall of Fame elections. These "Adjustment Factors"—positive for old-school candidates like Omar Vizquel, negative for steroid users or sabermetric darlings like Roger Clemens—are the demographic weighting to Ryan's raw polling data. And indeed, they produce more accurate results than just taking the Thibodaux Tracker as gospel:

My model's average error was 1.6 percentage points; the raw data was off by an average of three points per player. I didn't have as many big misses this year as last year; my worst performance was on Larry Walker, whom I overestimated by 5.0 points. My model assumed the erstwhile Rockie would gain votes in private balloting, as he had done every year from 2011 to 2016, but 2017 turned out to be the beginning of a trend; Walker did 10.5 points worse on 2018 private ballots than on public ones. I also missed Thome's final vote total by 3.5 points, although I feel better about that one, since first-year candidates are always tricky to predict. Most of my other predictions were pretty close to the mark, including eight players I predicted within a single percentage point. I came within two points of the correct answer for 17 of the 23 players forecasted, giving me a solid median error of 1.3 points. For stat nerds, I also had a root mean square error (RMSE) of 1.9 points.

All three error values (mean, median, and RMS) were the second-best of my now-six-year Hall of Fame forecasting career. But that's misleading: during the past two years, thanks to Ryan's tireless efforts, more votes have been made public in advance of the announcement than ever before. Of course my predictions are better now—there's less I don't know.

Really what we should be measuring is my accuracy at predicting only the 175 ballots that were still private when I issued my final projections just minutes before Jeff Idelson opened the envelope to announce the election winners. Here are the differences between my estimates for those ballots and what they actually ended up saying.

The biggest misses are still with the same players, but the true degree of my error is now made plain. I overshot Walker's private ballots by more than 12 percentage points, and Thome's by more than eight. Those aren't good performances no matter how you slice them. If we're focusing on the positives, I was within four percentage points on 16 of 23 players. My average error was 3.8 points, much better than last year when I had several double-digit misses, but my median error was 3.2 points, not as good as last year.

But where I really fell short was in comparison to other Hall of Fame forecasters: Chris Bodig, who published his first-ever projections this year on his website, Cooperstown Cred; Ross Carey, who hosts the Replacement Level Podcast and is the only one with mostly qualitative predictions; Scott Lindholm, who has been issuing his projections alongside me since day one; and Jason Sardell, who first issued his probabilistic forecast last year. Of them all, it was the rookie who performed the best: Bodig's private-ballot projections had a mean and median error of only 2.2 percentage points. His RMSE also ranked first (2.7 points), followed by Sardell (3.1), Carey (3.9), me (4.6), and Lindholm (6.3). Bodig also came the closest on the most players (10).

Overall, my model performed slightly better this year than it did last year, but that's cold comfort: everyone else improved over last year as well (anecdotally, this year's election felt more predictable than last), so I repeated my standing toward the bottom of the pack. Put simply, that's not good enough. After two years of subpar performances, any good scientist would reevaluate his or her methods, so that's what I'm going to do. Next winter, I'll explore some possible changes to the model in order to make it more accurate. Hopefully, it just needs a small tweak, like calculating Adjustment Factors based on the last two elections rather than the last three (or weighting more recent elections more heavily, a suggestion I've received on Twitter). However, I'm willing to entertain bigger changes too, such as calculating more candidates' vote totals the way I do for first-time candidates, or going more granular to look at exactly which voters are still private and extrapolating from their past votes. Anything in the service of more accuracy!

Hall of Fame Projections Are Getting Better, But They're Still Not Perfect

You'd think that, after the forecasting debacle that was the 2016 presidential election, I'd have learned my lesson and stopped trying to predict elections. Wrong. As many of you know, I put myself on the line yet again last month when I shared some fearless predictions about how the Baseball Hall of Fame election would turn out. I must have an addiction.

This year marked the fifth year in a row that I developed a model to project Hall of Fame results based on publicly released ballots compiled by Twitter users/national heroes like Ryan Thibodaux—but this was probably the most uncertain year yet. Although I ultimately predicted that four players (Jeff Bagwell, Tim Raines, Iván Rodríguez, and Trevor Hoffman) would be inducted, I knew that Rodríguez, Hoffman, and Vladimir Guerrero were all de facto coin flips. Of course, in the end, BBWAA voters elected only Bagwell, Raines, and Rodríguez, leaving Hoffman and Guerrero to hope that a small boost will push them over the top in 2018. If you had simply taken the numbers on Ryan's BBHOF Tracker at face value, you would have gotten the correct answer that only those three would surpass 75% in 2017.

But although my projections weren't perfect, there is still a place for models in the Hall of Fame prediction business. In terms of predicting the exact percentage that each player received, the "smart" model (which is based on the known differences between public and private voters) performed significantly better than the raw data (which, Ryan would want me to point out, are not intended to be a prediction):

My model had an overall average error of 2.1 percentage points and a root mean square error of 2.7 percentage points. Most of this derives from significant misses on four players. I overestimated Edgar Martínez, Barry Bonds, and Roger Clemens all by around five points, failing to anticipate the extreme degree to which private voters would reject them. In fact, Bonds dropped by 23.8 points from public ballots to private ballots, and Clemens dropped by 20.6 points. Both figures are unprecedented: in nine years of Hall of Fame elections for which we have public-ballot data, we had never seen such a steep drop before (the previous record was Raines losing 19.5 points in 2009). Finally, I also underestimated Fred McGriff by 5.4 points. Out of nowhere, the "Crime Dog" became the new cause célèbre for old-school voters, gaining 13.0 points from public to private ballots.

Aside from these four players, however, my projections held up very well. My model's median error was just 1.2 points (its lowest ever), reflecting how it was mostly those few outliers that did me in. I am especially surprised/happy at the accuracy of my projections for the four new players on the ballot (Rodríguez, Guerrero, Manny Ramírez, and Jorge Posada). Because they have no vote history to go off, first-time candidates are always the most difficult to forecast—yet I predicted each of their final percentages within one point.

However, it's easy to make predictions when 56% of the vote is already known. By the time of the announcement, Ryan had already revealed the preferences of 249 of the eventual 442 voters. The true measure of a model lies in how well it predicted the 193 outstanding ones. If you predict Ben Revere will hit 40 home runs in 2017, but you do so in July after he had already hit 20 home runs, you're obviously benefiting from a pretty crucial bit of prior knowledge. It's the same principle here.

By this measure, my accuracy was obviously worse. I overestimated Bonds's performance with private ballots by 13.4 points, Martínez's by 11.5, and Clemens's by 9.8. I underestimated McGriff's standing on private ballots by 12.9 points. Everyone else was within a reasonable 6.1-point margin.

That was an OK performance, but this year I was outdone by several of my fellow Hall of Fame forecasters. Statheads Ben Dilday and Scott Lindholm have been doing the model thing alongside me for several years now, and this year Jason Sardell joined the fray with a groovy probabilistic model. In addition, Ross Carey is a longtime Hall observer and always issues his own set of qualitatively arrived-at predictions. This year, Ben came out on top with the best predictions of private ballots: the lowest average error (4.5 points), the lowest median error (3.02 points), and the third-lowest root mean square error (6.1 points; Ross had the lowest at 5.78). Ben also came the closest on the most players (six).

(A brief housekeeping note: Jason, Scott, and Ross only published final projections, not specifically their projections for private ballots, so I have assumed in my calculations that everyone shared Ryan's pre-election estimate of 435 total ballots.)

Again, my model performed best when using median as your yardstick; at a median error of 3.04 points, it had the second-lowest median error and darn close to the lowest overall. But I also had the second-highest average error (4.8 points) and root mean square error (6.2 points). Unfortunately, my few misses were big enough to outweigh any successes and hold my model back this year after a more fortuitous 2016. Next year, I'll aim to regain the top spot in this friendly competition!

What I Didn't Expect in Baseball in 2016

There has been a lot to reflect on in 2016. Some of those year-in-review pieces are serious; others are more light-hearted. This is the latter. Having already shared my thoughts on politics, I now turn to another year-end tradition around these parts: revisiting my preseason baseball predictions. Because predicting ball is an exercise in futility, grading my MLB predictions is always a humbling experience; more often, it's a hilarious one, filled with "boy, was I wrong" moments as well as the occasional "jeez, that wild guess was scary accurate." Let's dig into my 2016 American League and National League predictions to see how I did.

Prediction: The AL playoff teams would be the Blue Jays, Rays, Indians, Astros, and Rangers. The NL playoff teams would be the Mets, Nationals, Cubs, Cardinals, and Giants.
What Really Happened: The Blue Jays, Indians, and Rangers made it, but the Red Sox and Orioles replaced the Rays and Astros. In the NL, I missed only the Dodgers, who replaced the Cardinals. Overall, seven out of 10 playoff teams was pretty good! I also estimated the win totals of 14 teams within five; my average error of 6.4 wins was better than any of the five years I've been making official predictions.

Prediction: The Dodgers would lead baseball in days on the disabled list. The injury bug would be particularly devastating to their starting rotation, with only Clayton Kershaw surpassing 200 innings en route to another Cy Young Award.
What Really Happened: One of my eeriest predictions. The Dodgers' snakebitten starting rotation was the story of the summer in Los Angeles, and not even Kershaw was immune: a herniated disk in his back interrupted an historic season and cost him over 10 starts, enabling Washington's Max Scherzer to steal the Cy Young trophy. Kenta Maeda ended up leading the Dodgers with only 175.2 innings, and LA set new major-league records for most players put on the DL in one season (28) and man-days spent on the DL (2,418).

Prediction: Three Indians pitchers would throw no-hitters: Corey Kluber, Carlos Carrasco, and Danny Salazar.
What Really Happened: Only one pitcher league-wide tossed a no-no in 2016: Jake Arrieta on April 21.

Prediction: The Orioles would slug 250 home runs as a team—a number not seen since the 2010 Blue Jays—but would be led in OBP by slap hitter Hyun Soo Kim.
What Really Happened: Baltimore hit 253, 28 more than the next most powerful team. Kim's .382 OBP led all Orioles with at least 16 plate appearances.

Prediction: Julio Teheran would struggle his way to a 4.00 ERA, clearing a path for Ender Inciarte to be the most valuable Brave. He would even be better than the man he was in part traded for: Shelby Miller. Patrick Corbin and maybe even Robbie Ray would allow fewer runs than their new Arizona teammate.
What Really Happened: Inciarte put up 3.8 WAR, third-best on the Braves. Freddie Freeman (6.5 WAR) led the squad, and Teheran righted the ship to the tune of a 3.21 ERA, 4.07 K/BB ratio, and 4.9 WAR. As for Miller? He was worth just −0.8 WAR for the Diamondbacks. Corbin (5.15 ERA) and Ray (4.90) both had terrible seasons, but not nearly the disaster that was Miller's (6.15).

Prediction: The Royals would have a better record when Raúl Mondesí Jr. starts at shortstop than when Alcides Escobar does.
What Really Happened: Trick question: Mondesí never started at shortstop but instead played almost exclusively second base. Meanwhile, Escobar and his .261/.292/.350 batting line inexplicably started every single game the Royals played. Yet sure enough, Kansas City's record in just Mondesí starts was 23–17, and overall (i.e., in Escobar starts) it was just 81–81.

Prediction: Jon Gray would be dominant on the road—posting a 2.80 ERA—but would be unable to solve Coors Field, stumbling to a 5.20 home ERA.
What Really Happened: The Rockies phenom actually struggled more on the road: a 4.91 ERA. At home, he was a surprisingly good 4.30 ERA pitcher, limiting hitters to an excellent .241/.291/.383 line at altitude.

Prediction: Kevin Cash and Bruce Bochy would be voted 2016's Managers of the Year. Terry Francona would be runner-up in the American League. Buck Showalter would have such a disappointing season he would be shown the door.
What Really Happened: Neither Cash nor Bochy came close to sniffing the award, which went to Francona in the Junior Circuit. Showalter kept his job, but he did take heat after failing to use Zach Britton in a key situation in the Wild Card Game.

Prediction: Age would catch up to Justin Verlander and Ian Kinsler in Detroit. Verlander's fastball velocity would tick down until he led the Tigers with the highest WHIP on the team. Meanwhile, Mike Pelfrey would boast the league's highest ERA.
What Really Happened: Verlander's WHIP did lead the team—in a good way. It was the lowest in the entire AL, in fact. According to PITCHf/x, his fastball averaged 93.7 miles per hour, his best mark since 2013. Likewise, the highest ERA in the American League didn't belong to Pelfrey—but rather his teammate, Aníbal Sánchez. Far from deteriorating, Kinsler upped his WAR for the fourth year in a row (to 6.1) and won a Gold Glove.

Prediction: Bryce Harper would disappoint in the follow-up to his insane MVP season of 2015. He would maintain the same beastly rate stats, but he would miss a third of the season due to injury.
What Really Happened: Bryce stayed healthy for the second full season—or at least that's what he claimed. His OPS mysteriously dropped a staggering 295 points to .814, perhaps the result of playing through a right shoulder injury for… yep, a third of the season.

Prediction: Three Twins would finish in the top five for AL Rookie of the Year: José Berríos, Byung Ho Park, and eventual winner Byron Buxton. In the NL, Corey Seager would waltz home with the trophy.
What Really Happened: None of the three Twins even got a single vote. Berríos started 14 games with a hellish 8.02 ERA, Park hit just .191, and Buxton notoriously scuffled through his first two cups of coffee with a .193/.247/.315 slash line before coming to life (.287/.357/.653) in September. The Tigers' Michael Fulmer, of course, ended up winning the award in the AL. In the Senior Circuit, was there ever any doubt? Seager's .877 OPS and 6.1 WAR made him an easy choice.

Prediction: Jay Bruce would be one of the only good hitters on the Reds, earning him a trade out of town. Brandon Phillips, meanwhile, would slip to a .300 OBP and cease to be an asset on defense. His final WAR: 0.0.
What Really Happened: Bruce put up an .875 OPS for Cincinnati, better than anyone not named Joey Votto, and on August 1 he was traded to the Mets. Phillips held on in the OBP department (.320), but for the first time since 2006 he put up a negative DRS (−7) to finish with a 0.8 WAR.

Prediction: The penny-pinching Astros would keep Ken Giles out of the closer's role in an effort to suppress his salary in arbitration—but this would also allow them to use him in the highest-leverage situations.
What Really Happened: Indeed, the Astros handed Luke Gregerson, then Will Harris the closer's role before giving Giles a crack; he finished with 15 saves. Giles didn't lead the AL in leverage index like I predicted, but his 1.83 LI was higher than any other Astro.

Prediction: Marcus Semien would be one of the Athletics' best players and would even be a net positive on defense.
What Really Happened: Semien slugged 27 home runs and was worth 3.0 WAR, both tops among Oakland hitters. On defense, he was barely an asset (0.1 dWAR), thanks to the positional adjustment of playing shortstop.

Prediction: The worst infield defense in the AL would doom Rick Porcello's second season with the Red Sox. Meanwhile, in Chicago, Chris Sale would sail to his first Cy Young Award with a sub-2.50 ERA.
What Really Happened: The Boston infield actually saved nine runs defensively, helping Porcello along to a 3.15 ERA and the Cy Young. Sale was no slouch either, posting a 3.34 mark for the Pale Hose. By December, of course, who was better is a bit of a moot point: they're now teammates in Boston.

Prediction: Prince Fielder and Shin-Soo Choo would turn in carbon copies of their superb 2015 seasons.
What Really Happened: Fielder had the worst season of his career, hitting just .212/.292/.334 through 370 plate appearances before being forced to retire. Choo had a similarly snakebitten season, going on the disabled list four times with four unrelated freak injuries. On April 10, he was shelved with a calf strain; he returned May 20, and then pulled a hamstring after just three innings. On July 20, he hit the DL again with back inflammation, returning on August 4. On August 15, he was hit by a pitch that broke his forearm and ended his season.

Prediction: Pittsburgh wouldn't need to worry about slipping offensively with the losses of Pedro Alvárez and Neil Walker. John Jaso and David Freese would prove shrewd free-agent signings.
What Really Happened: The Pirates went from a 97 OPS+ in 2015 to a 95 OPS+ in 2016. Jaso hit an above-average .268/.353/.413, and the team also rewarded Freese's .270/.352/.412 line with a two-year, $11 million extension.

Prediction: Mookie Betts would contend for AL MVP, but Carlos Correa would actually win it. The shortstop's 40 home runs would distract voters yet again from Mike Trout. In the National League, Paul Goldschmidt would finally step out of others' shadows and claim his first MVP award.
What Really Happened: Trout deservedly won the AL hardware for just the second time. Correa still turned in a great season, but he slammed just 20 taters. For his part, Betts more than contended for MVP: he came darn close to winning it, with nine first-place votes and a second-place finish. As for Goldschmidt, a down year (for him) doomed him to "just" an 11th-place finish in the NL race.

Prediction: The Diamondbacks would regret the Jean Segura trade. His bat would continue to drag down the lineup, and his poor defense would contribute to Arizona tumbling from 63 DRS to a neutral fielding team.
What Really Happened: Segura had one of the best come-out-of-nowhere seasons in a long time, hitting .319/.368/.499 for a 5.7 WAR. He was average on defense, although the D'backs did fall all the way to −12 DRS.

Prediction: A 5.0 K/BB ratio by Anthony DeSclafani would see him selected to the All-Star Game.
What Really Happened: DeSclafani was injured for much of the first half and did not debut until a month before the All-Star Game, but if he had frontloaded his first 16 starts (8–2, 2.93 ERA, 1.14 WHIP, 4.1 K/BB ratio), he surely would have earned a ticket to San Diego.

Prediction: Marco Estrada and J.A. Happ would fall dramatically back down to Earth, with 9.0 hits per nine innings and a 90 ERA+ respectively. Aaron Sánchez would show he belonged in the bullpen all along by struggling as a starter.
What Really Happened: Estrada followed up a 2015 in which he led the AL with 6.7 hits per nine with a 2016 in which… he led the AL with 6.8 hits per nine. All Happ did was win 20 games with a 135 ERA+. Sánchez led the league with a 3.00 ERA as one of the breakout starters in all of baseball.

Prediction: Under the tutelage of hitting coach Barry Bonds, Marcell Ozuna would take his game to the next level, setting career highs in all three slash categories.
What Really Happened: Ozuna had a good year but not quite a full breakout. His .321 OBP was a career high, but his .266 average and .452 slugging percentages were each three points shy of his historical best. Oh, and Bonds was fired at the end of the season for allegedly losing interest in the team.

Prediction: Andrew Heaney would establish himself as the one sure thing in an injury-plagued Angels rotation, while Jered Weaver would shockingly retire midseason when it became apparent he couldn't throw above 80 miles per hour.
What Really Happened: Heaney got in just one start all year—a six-inning, four-run effort against the Cubs—before feeling elbow discomfort. He had Tommy John surgery in July. Weaver did indeed have a tough time getting outs with his 84.0-mile-per-hour "heater," but he stuck it out the whole year and ended with a 5.06 ERA.

Prediction: Ian Desmond would look so lost in the outfield that the Rangers would bench him. He would enter career purgatory, bouncing around on the free-agent market as a utility man for the rest of the decade.
What Really Happened: With −4 DRS, Desmond wasn't an asset in the outfield, but he ably remade himself into a useful player there, amassing 2.7 WAR. As for his financial future, Desmond just signed an inflated five-year, $70 million deal with the Rockies.

Prediction: Domingo Santana would put up a vintage Adam Dunn season: a .230 average but a .340 OBP, 180 strikeouts but 30 home runs.
What Really Happened: Santana lost significant time to two injuries in 2016, but he still did the following in 281 plate appearances: a .256 average, .345 OBP, 11 home runs, and 91 strikeouts.

Prediction: Milwaukee would be the only team in baseball with zero complete games in 2016.
What Really Happened: Indeed, no Brewer starter pitched a complete game. However, three other teams also shared this ignominious distinction: the Marlins, Yankees, and Blue Jays.

Prediction: Jason Heyward, Kris Bryant, and Anthony Rizzo would all rank among the NL's top 10 position players by WAR. Ben Zobrist, meanwhile, would continue to decline thanks to poor defense.
What Really Happened: Bryant (at 7.7 WAR, the NL MVP) and Rizzo (5.7) ranked first and fifth, respectively, in WAR, but Heyward's contract infamously proved a bust, as he could muster just a 70 OPS+ and 1.5 WAR. Zobrist did cost his team multiple runs defensively for the third straight year, but his excellent hitting (.272/.386/.446) made him Chicago's fifth-best position player (3.8 WAR).

Prediction: The White Sox would be a fountain of youth for Mat Latos, who would be worth 2.0 WAR, and Melky Cabrera, who would begin to justify his $42 million contract.
What Really Happened: Latos bombed out of Chicago, and then Washington, with a 4.89 ERA and 0.1 WAR. Yet in a development no one outside the South Side noticed, Cabrera was great in 2016, putting up an .800 OPS that almost matched his 2014 performance.

Prediction: Sonny Gray's 2015 luck would reverse itself, with a .340 BABIP leading to a 3.95 ERA.
What Really Happened: Gray was unlucky—and then some. He offended A's fans with a 5.69 ERA, far worse than an already poor 4.67 FIP. His BABIP was a not-great .319, but the real culprits were a terrible 64% strand rate and artificially inflated 1.4 home runs per nine innings.

Prediction: Yasiel Puig would rediscover his 2013–2014 form, and Joc Pederson would discover new heights.
What Really Happened: Puig had his worst season in the majors yet, hitting just .263/.323/.416 between hamstring injuries. He fell so out of favor with the club that they demoted him to AAA in August, and they still found room to criticize him for his behavior while in Oklahoma City. Back in California, Pederson continued to blossom, pairing his previous on-base ability and home-run power with more well-rounded hitting: more doubles and more selective baserunning.

Prediction: Mike Leake would be the one weak link in the Cardinals' rotation, with a 100 ERA+, and Jaime García would land on the DL yet again.
What Really Happened: García pitched a full season for the first time since 2011, starting 30 games. Leake had his worst season yet: an 87 ERA+. He wasn't even the worst St. Louis pitcher, though, as Michael Wacha spat out an 81 ERA+ despite a pretty good 3.91 FIP.

Prediction: Trea Turner would grab ahold of the Nationals' shortstop job so surely that Stephen Drew wouldn't even collect 100 plate appearances.
What Really Happened: Turner had nothing to do with Drew amassing just 165 plate appearances for Washington's $3 million investment. The infield prospect remained exiled to AAA until mid-July, when he finally came up… only to be moved to the outfield. He certainly made a statement, though, with his .342 average and 33 stolen bases.

Prediction: Rich Hill would prove to be an illusion, finishing with a 4.00 ERA.
What Really Happened: Hill extended his four-start dominance from the end of 2015 into his first 14 starts of 2016 with Oakland, posting a 2.25 ERA. Then he was traded to the Dodgers as one of the deadline's biggest gets and did even better: a 1.83 ERA. He went into the offseason as the top pitching prize on the free-agent market.

Prediction: Led by Craig Kimbrel and Carson Smith, the Red Sox bullpen would strike out a quarter of the batters it faced, second in the league only to the hated Yankees.
What Really Happened: It took Smith just 2.2 innings to succumb to season-ending injury, but the Sox bullpen still struck out 25.4% of batters. That's not as impressive as it sounds, though; five other bullpens, including the Yankees' (27.1%), fanned more.

Prediction: Tanner Roark would be mediocre, and Noah Syndergaard would go under the knife at midseason.
What Really Happened: Both Roark and Syndergaard garnered downballot Cy Young votes. Roark finished with a 2.83 ERA, and Syndergaard had the game's strongest peripheral stats (a 2.29 FIP) in his 30 starts.

Prediction: Yovani Gallardo and Kevin Gausman would swap 2015 ERAs, with Gallardo finishing at 4.25 and Gausman at 3.42.
What Really Happened: The two Orioles hurlers did undergo a freaky Friday situation, with Gallardo regressing to a 5.42 ERA and Gausman improving to 3.61, establishing himself as the team ace.

Prediction: Ray Searage would not be able to fix what ails Ryan Vogelsong or Jon Niese, but Juan Nicasio would thrive in Pittsburgh.
What Really Happened: All three reclamation projects fell flat. Vogelsong pitched to a 4.81 ERA in 82.1 innings, and Niese mustered just a 4.91 ERA before being traded back to the Mets. Nicasio boasted strong strikeout numbers (10.5 per nine innings), but he had just a 4.50 ERA in a swingman role.

Prediction: PETCO Park would help James Shields to a bounceback season of a 3.30 ERA, a 1.20 WHIP, and 2.0 walks per nine innings.
What Really Happened: Shields limped through 11 starts in San Diego with a 4.28 ERA before getting traded to the White Sox, where he was lit up in the bandbox that is U.S. Cellular Field. He finished with a 5.85 ERA, 1.60 WHIP, and 4.1 walks per nine—all career worsts.

Prediction: Breakout seasons by Steve Pearce, Blake Snell (who would strike out 10 batters per nine innings), Drew Smyly, and Chris Archer would lead the Rays to the World Series.
What Really Happened: Pearce did return to his 2014 self, slashing .288/.374/.492, and Snell struck out 9.9 per nine (so close!). However, Smyly finished with a 4.88 ERA and Archer lost 19 games, the most in baseball. The Rays stunk up the joint to the tune of 94 losses—my biggest whiff on any team.

Prediction: Even-year magic would strike again. New ace Johnny Cueto would make up for free-agent bust Jeff Samardzija (0.5 WAR), and the Giants would win the World Series for the fourth time in seven years.
What Really Happened: Samardzija did not disappoint (2.7 WAR), and Cueto was even better (5.6 WAR), but the Giants were bounced from the playoffs in four games by their atrocious bullpen and the eventual World Series winners—the Chicago Cubs.