Read the rest of this entry »

Currently in possession of a 68-56 record and standing just a half-game out of first place in the NL West, the Colorado Rockies are in the midst of an objectively good season. Actually, the 2017 and -18 versions of the club have the best combined two-year winning percentage for any pair of Rockies teams in history, so one could make the argument that this is Colorado’s finest run ever. They’ve had two MVP candidates in the starting lineup both seasons and the starting pitching, long a team bugaboo, ranks ninth in the majors by WAR over that time period. Things in Colorado aren’t bad, per se.

But they could be better, it seems, without much effort. One real problem for the Rockies has been the team’s lack of offensive depth. It’s an issue they’ve shown little interest in addressing. And it’s costing them real wins.

With Nolan Arenado and Charlie Blackmon in 2017 and Arenado and Trevor Story in 2018, Colorado’s top-end offensive talent has been as dangerous as that of any team in baseball. Once you look past the top of the roster, though, things become a bit more frightening. Despite the team’s respectable raw numbers, the club’s offensive line reads like a gothic horror story after you factor in our old friend, Coors Field.

Even with the impressive performances by the brand names — most notably Nolan Arenado, who has been a legitimate MVP contender both seasons — the Rockies rank near the bottom of baseball in offense. At five of the eight main offensive positions — I’m not considering pitcher hitting or the DH for interleague road games — the Rockies have ranked 25th or worse in baseball by wRC+.

Now, DJ LeMahieu is a very ordinary offensive second baseman, outside of his .348/.416/.495 campaign in 2016, but he more than makes up for any bat-related shortfall with his defense. You can’t say that for the other positions ranking near the bottom of baseball.

Read the rest of this entry »

Matt Kemp is in a funk. I’m not talking about the kind of funk endorsed by Sly Stone or Parliament. Kemp’s funk is more like the time a friend of mine left a chicken salad sandwich in my car over a hot weekend and it fermented into a noxious cloud of nauseating death-barfiness. Or funk metal.

I’m nearly certain the Dodgers didn’t originally expect to ever have Kemp on the roster in 2018. LA acquired him from Atlanta in exchange for Charlie Culberson, Adrian Gonzalez, Scott Kazmir, Brandon McCarthy, and cash — and if the the deal were to have occurred in 2012, with those names, it would have been a blockbuster. In the winter of 2017, however, Kemp wasn’t so much a player as a tax loophole, the maguffin in a trade that was primarily about teams aligning their year-to-year payrolls in such a way as to avoid luxury tax.

But a funny thing happened on the way to the luxury tax: Kemp became interesting. For once, one of those articles about a player looking amazing in spring training actually bore real fruit. Kemp showed up to spring training in excellent shape, having lost a non-trivial 40 pounds and gained a renewed interest in playing defense.

The homecoming to Los Angeles, after a lot of hurt feelings years ago, turned out to be a positive one. When Kemp slugged .561 in spring training while also exhibiting improved defense and a real effort to be a mentor to the younger players, he gave the Dodgers enough justification to keep him on the team as a role player.

Los Angeles struggled early. Kemp, however, did not. With one of the club’s top batting marks and the promise of better defense — or at least decidedly less-atrocious defense — fulfilled, Kemp received more at-bats. Unlike in previous seasons with Atlanta and San Diego, Kemp’s playing time in this case was earned on the merits of his play and not his reputation or salary. He started in the All-Star Game.

Since the All-Star Game, though, things have not gone well for Kemp. Standing at .310/.352/.522 when baseball took its midsummer respite, Kemp’s OPS has bled about 100 points in just a month, and he’s stalled at 1.1 WAR for the 2018 season. Neither ZiPS or Steamer are optimistic about a turnaround, projecting him to finish at 1.3 and 1.2 WAR, respetively, the primary difference between the two being playing time.

Before Wednesday’s 2-for-4 performance, Kemp last had a multi-hit game on July 23rd and now has hits in four of his last 19 games. Overall, he’s 5-for-58 from that date with only a lone double. The result? A .086/.191/.103 line.

So, what happened to Kemp’s 2018?

Read the rest of this entry »

Projecting the Rookie of the Year award is simultaneously easier and more difficult than the Cy Young. It’s easier in the sense that there are fewer rookies than non-rookies and that, in most seasons, there’s a definite top tier of candidates that crowds out the rest of the pack.

What makes it a bit trickier is that the standards for rookies are applied a bit more haphazardly by writers. Because rookie ballots feature only three players — as opposed to five for the Cy Young and 10 for MVP — we see fewer players actually included in the final voting. Ideally, you’d like to bring in all the voters, crack open their skulls, and somehow read their brains to see how everyone would rank at least the top 10 rookies. My lawyers, however, inform me that this is extremely illegal and also totally gross.

In the end, I’m less confident about the Rookie of the Year model than the MVP or Cy Young versions. While, historically, ZiPS identifies about seven of the top 10 MVP and Cy Young vote-getters, the model only gets three of the top five rookies. Hopefully, as the electorate becomes more and more analytically inclined, I’ll be able to improve the model.

The greatest challenge of projecting the AL race is figuring out what to do with Shohei Ohtani. There’s no guidance available on how two-way players ought to be treated, so there’s a lot more guesswork than usual. Comparing apples to oranges is tricky enough — although rendered less tricky by the fact that they’re frequently right next to each other at the grocery store — but how do you treat something that is an apple and an orange at the same time? Applange and orpple both sound terrible.

Read the rest of this entry »

Read the rest of this entry »

Last week, I ran down the latest ZiPS projections for the MVP awards. The Cy Young tends to be a bit simpler to project than the MVP, for a number of reasons: there are no positional differences for which to adjust (outside of starter vs. reliever), no consideration of defense, fewer candidates (because hitters aren’t eligible for the award, where pitchers can win the MVP), and finally, fewer pitching stats from which to measure performance. One historical note of interest is that, while team quality plays a factor, it appears to be a less predictive factor than it is for MVP, also serving to make things a bit less difficult.

Before I begin, one clarification from the MVP post: the award percentages that appear here aren’t based on each player’s mean final projections, but from the whole array of possibilities, 1st to 99th percentile, for each player. So, for example, Giancarlo Stanton’s projected award chance of 2.8% isn’t predicated on him winning the award based on his predicted final line, but from the better scenarios in which he exceeded that current projected final line.

There’s a popular perception that Chris Sale wears out down the stretch. The perception is supported by his career splits: Sale has a career ERA below three for each of the first four months of the season with a 3.22 ERA for August and a 3.78 ERA in September. This data isn’t quite as robust as some suggest, but assuming for the sake of argument that it reflects something real, it’s also worth nothing that Sale’s been used more carefully than in past seasons. At this point in 2017, Sale had thrown at least 110 pitches in 16 starts after having crossed that threshold 10 times in both of the previous two seasons. This year, he’s at only four. The FanGraphs Depth Charts have Sale finishing at just 187 innings for the season, his lowest figure since 2014. ZiPS projects Sale to finish the year with the most strikeouts and best ERA in the AL while also tying for the highest WAR mark. The combination of his own performance plus the strength of the Red Sox make him a strong bet to go over the top.

Read the rest of this entry »

Once baseball’s non-waiver trade deadline passes, you start to see the conversations shift from fantasy to terra firma. Almost all the big-name players who are likely to make Chicxulub-sized impacts on team rosters have already been traded. The focus shifts squarely back to the pennant races, and with them, talk of individual player awards.

It will come as no surprise to most readers that I love working on predictive models. It’s not just about trying to predict the future — though that is inevitably a large part of it — it’s also about dissecting things to see how they work. Awards are something I’ve always found fascinating because they not only deal with truths in baseball but also with the psychology and mindset of the people covering the sport. We talk a lot about baseball writers believing more in stats like OBP and SLG, and eventually WAR, but the proof in the pudding is in the eating. If advanced stats don’t budge how writers are judging the best players in the league, are they truly accepted?

We’ll start this trio of pieces with the current MVP races. I’ve spent a lot of time over the past decade modeling MVP votes, and the truth is that things have, in fact, shifted considerably. Slugging percentage and wins above replacement do have more predictive value than they did in the past, as does on-base percentage (albeit to a lesser extent). The defensive players who get a larger share of the vote than one expects tend to be players who do well in the sabermetric defensive measures. Team quality and the Triple Crown stats still play the largest role, however, and even though the MVP award doesn’t specify hitters over pitchers, pitchers still make far less of a dent than one would expect from their impact.

Read the rest of this entry »

Language is rich with words and terms that recognize the contributions of historical figures. This can be a good thing, but also a bad one, depending on what’s being commemorated. You’d rather go down in history as the namesake of a popular sweater (like James Brudenell, 7th Earl of Cardigan) or a certain type of legal protection (Ernesto Miranda) than for those traits by which Nicholas Chauvin or Ned Ludd are best remembered. In baseball, former utility infielder Mario Mendoza belongs to the latter category. Thanks to some creative but cruel teammates on the 1979 Mariners, Mendoza’s name has become synonymous with hitting futility. To fall below the Mendoza Line is to record a batting average below .200.

For a hitter both to qualify and to finish below the Mendoza Line actually represents a notable feat of ineptitude. One must not only fight the influence of the Regression Gods attempting to pull the hitter into the respectable company of the .200s, but also to play sufficiently well otherwise not to lose his job. It’s something Mario Mendoza himself never actually even achieved, coming closest in the black-magical 1979 season, but falling short due to manager Darrell Johnson’s mercy: Mendoza was frequently pinch-hit for in his third time up, was pinch-hit-for five times in his second plate appearance, and lost significant playing time to Larry Milbourne late in the season.

Read the rest of this entry »

Through Monday morning, José Ramírez has nearly accomplished something very unusual, which to-date has only been done once in baseball history. With just 0.1 WAR in separation, or roughly one run, Ramírez has almost caught Mike Trout. While one might ideally like Derek Fisher – or at least someone named Fry, Cook, or Giantbearpaw – to best a Trout, it’s Ramírez, the still-young Cleveland third baseman, overshadowed as a prospect by Francisco Lindor, who is leading the charge.

Trout, of course, hasn’t just been squeaking out WAR leads, as the average result during his five full, healthy seasons has been a 1.1 WAR lead over the second-place position player, terrorizing the error bars as thoroughly as major-league pitching. Only Bryce Harper has matched Trout so far, a player linked with the Angels center fielder while a prospect but one who has generally fallen short.

Even in Trout’s injury-trimmed 2017, during which he only played 114 games, he still finished fourth among hitters with a 6.9 WAR. That was good enough to be the fifth-best season for a player who appeared in fewer than 120 games since 1901, behind George Brett, Joe DiMaggio, Ted Williams, and Mike Schmidt, all players you may have heard of.

Ramírez didn’t explode onto the baseball scene as Trout did; in fact, he has been a bit of a surprise. The ZiPS projection season liked him as a prospect, but it didn’t fall CPU-over-RAM in love with him, seeing him more as a cordial fan. Trout is still the favorite, but ZiPS gives him a 35% chance of being surpassed by Ramírez in the year-end WAR count, something that’s awfully hard to do.

After his 2013 stint in Double-A, Ramírez got his first official ZiPS projection for the 2014 season. The computer pegged him at .267/.308/.346 at second base, good for 1.5 WAR in 116 games. While a league-average projection for a 21-year-old is impressive, I can’t make any claim to ZiPS predicting players achieving superstar status.

In 2015, Ramírez’s projection was similar at .263/.304/.360, good for a 90 OPS+ and 2.2 WAR over 138 projected games. This was largely thanks to a .302/.360/.441 Triple-A season in 2014 followed by a .262/.300/.346 line in his first real stint with the Indians. That was enough to get him almost into the top 20 in ZiPS (he would’ve been a top-five prospect in my 2015 list if he hadn’t already lost his Rookie of the Year qualifications).

In 2015, Ramírez began the season as the starting shortstop for Cleveland, and while I don’t have any inside knowledge of Cleveland’s reasoning, I would imagine they wanted to see what they had in Ramírez at short before Lindor hit the majors and cleared out the rest of the suitors, as Odysseus did in legend upon his return to Ithaca.

And what he showed wasn’t that much, at least initially. His defense wasn’t great, at -6 in UZR and -2 in DRS over fewer than 400 innings in 2015. When demoted in June, he was only hitting an Alcidian .180/.247/.240. By the time he returned to the majors in August, Ramírez had lost the shortstop job to the aforementioned Lindor for good, and hit .259/.337/.438, mostly filling in for Jason Kipnis and Giovanny Urshela, both nursing shoulder injuries.

ZiPS still believed in Ramírez going into 2016, but didn’t see quite as much upside, with his 2015 dropping him to 43rd in rest-of-career ZiPS WAR with a .262/.316/.383 line and a 1.9 WAR projection.

By now, you know how the rest of this story went. Ramírez burst into stardom in 2016, hitting .312/.363/.462, good for a 121 wRC+ and 4.8 WAR. Then he got even better in 2017. And this year, he’s ridden the saber-limousine all the way to Crazyville. As of Monday morning, Ramírez stands at .300/.409/.631 with 33 home runs, 26 stolen bases, a 175 wRC+ and 7.5 WAR, all already career highs (except for batting average) and by substantial margins. You can even make a case that he’s actually been a little unlucky, doing all of this with a .277 BABIP, down 23 points from his career average of .300. From his hit profile, ZiPS thinks Ramírez “ought to” have a .315 BABIP this year, or .302 by Andrew Perpetua’s model.

My colleague Jay Jaffe wrote about the greatest seasons by a third baseman a few months ago, so I won’t go too into detail, but I will touch on a few important points.

For position players as a group, the 10-WAR seasons in history can be distributed into a few buckets:

• Hall of Famers, plus one that ought to be on performance alone (Barry Bonds)
• Players who are not yet eligible (Alex Rodriguez, Trout)
• Norm Cash
• Fred Dunlap

Out of those 52 seasons, 50 were players in the first two categories, all with nearly indisputable Hall of Fame talent. Cash isn’t quite as obvious, but he’s at least a borderline category and better than quite a few other Cooperstown-immortalized sluggers. That just leaves Dunlap and his curious 1884 for the St. Louis Maroons, as he hit .412/.448/.621 as a 25-year-old and never hit .275 again. I am unable to find any accusations of performance-enhancing tincture or tonic, so I’ll assume that this was another case of 1800s-baseball-gonna-1800s-baseball, given the variability of play quality at the time.

The mean projection for Ramírez by ZIPS for the rest of the season puts him at 9.7 WAR, tantalizingly close to the 10-WAR barrier, which has been broken by exactly zero third basemen in MLB history. Only Darrell Evans, Adrian Beltre, Rodriguez, and Ron Santo made it to within a half-win of ten. What this means is that as of right now, ZiPS is projecting José Ramírez to have a 53% chance at the greatest season by a third baseman in history and a 20% chance of being the first to hit ten WAR.

Ramírez’s season is not being driven by some freakishly high defensive WAR number, something you occasionally see given how volatile defensive measurements tend to be. The former shortstop’s defense only amounts to 8.7% of his total WAR, 51st among the 85 seven-WAR seasons by third basemen. A few players have seen this share go to over a third, headed by Brooks Robinson’s 1968, in which fielding makes up 58% of his WAR, (Graig Nettles and Robin Ventura both had seasons over 33%).

Naturally, all this uncontrolled awesomeness has altered Ramírez’s career-trajectory, tempora mutantur, nos et mutamur in illis and all that jazz. In this year’s FanGraphs trade value series, Kiley McDaniel ranked Ramírez as the player with the greatest trade value in baseball, up from 15th in last year’s edition.

Unsurprisingly, the occasionally anthropomorphic projection system on my desk has also adjusted its gaze. Ramírez’s 2016 campaign got him back up to 25th in future WAR and over 2000 hits for his career. His 2017 rocketed him up to 8th, third-best of players 25 or older for the 2018 season, behind only Trout and Mookie Betts (and their 2500 hits).

As you probably have guessed, 2018 shoots Ramírez even higher in the long-term projections. His top-five near-age offensive comps are all Hall of Famers or should be (Santo, Ernie Banks, Scott Rolen, Chipper Jones, and Brett). Over the next five years, ZiPS projects 32.5 WAR, more than six wins a year, and 52 total WAR remaining, third among position players only behind Trout and Lindor. He already passed the 2500 hit mark in the projections; now he’s up at 350 homers as well. And with some simple math, all of a sudden, his projected final rank among third basemen puts him in Hall of Fame territory.

You can shift these rankings around a little depending on how you categorize players like Rodriguez and Miguel Cabrera, but there’s only limited play possible with these numbers, and however you shuffle the deck, Ramírez is one of the very few players projected to finish with a career of historical, plaqued significance. Projection systems just aren’t designed for exuberance.

José Ramírez is heading on a course that could see him end up as one of the greatest Indians players in history and one of the greatest third basemen, period. And thanks to some shrewd wheeling-and-dealing by the team’s front office to sign him to a five-year, $26 million contract after his initial breakout in 2016, he’s going to be doing it in Cleveland through the 2023 season. Ramírez is now in his third consecutive year of doing things we didn’t know were possible for him; if he manages to pull this off for a fourth, he may just break baseball.