Read the rest of this entry »

It’s a rare day when a player looks back on a championship-winning season with something approaching regret, but then, Dustin Pedroia didn’t get much chance to contribute to the Red Sox during their World Series-winning 2018 campaign. After being limited to 105 games in 2017, largely due to ongoing discomfort in his left knee, Pedroia underwent an experimental cartilage restoration procedure that was expected to cost him the first couple months of last season. He returned in late May, but played just three games before inflammation forced him back to the injured list, and he spent the rest of the season as a bystander as the Sox overcame his absence while romping to their fourth title in the last 15 seasons.

“I don’t regret doing it, but looking back and knowing what I know now, I wouldn’t have done it,” Pedroia told reporters in Fort Myers last week, referring to the surgery, called osteochondral allograft transfer surgery, which involves grafting cartilage from a cadaver. While basketball and soccer players had undergone the procedure prior to Pedroia, teammate Steven Wright was the only baseball player who had done so.

The 35-year-old second baseman, who had previously undergone surgery to repair a torn meniscus in his left knee in October 2016 (among his numerous other ailments), believes both that he pushed himself too hard and that some of the expectations with regards to recovery times were “a little off.” In retrospect, he believes that avoiding the surgery in favor of longer rest and rehab might have been a more productive route. “It’s a complicated surgery,” said Pedroia. “The cartilage in my knee is great now. The graft is the thing. You’re putting somebody else’s bone in your body. To get that to incorporate fully, there are so many things going into it that.”

For what it’s worth, Wright, who was limited to 53.2 innings last year by three separate trips to the injured list for left knee inflammation (as well as a 15-game suspension for violating the league’s domestic violence policy) and missed the entire postseason, is similarly less than enamored with the results of his cartilage restoration surgery, which took place on May 8, 2017, 5 1/2 months prior to Pedroia’s. “It’s been tougher mentally than physically,” he told reporters last week. “It’s a game where it’s tough on your body, but then when you’ve got a flat tire going out there trying to pitch, it makes it a little bit tougher.” Wright later said that his knee currently feels like it’s been from a flat tire to a spare.

In anticipation of Pedroia’s early-2018 absence, the Red Sox retained July 2017 acquisition Eduardo Núñez in free agency, but both he and utilityman Brock Holt struggled in the role, to the point of topping my Replacement Level Killers list in late July. To that point, Sox second sackers had combined for a sad-sack -0.7 WAR. On July 30, they traded for the Tigers’ Ian Kinsler, who was replacement-level as well, though he did represent a defensive upgrade. In all, the eight players Boston used at second base — a cast rounded out by Brandon Phillips, Tzu-Wei Lin, Mookie Betts (!), and Blake Swihart — combined to hit .252/.308/.350 for a 75 wRC+ and -0.3 WAR. Only the Giants and Tigers were worse in the last two categories.

All of the participants save for Phillips and Kinser (who signed with the Padres) remain on the Red Sox’s 40-man roster. President of baseball operations Dave Dombrowski told reporters last week that he’s not expecting Pedroia to play 150 games. “We’re hopeful that he’s a 125-game player at this point,” said Dombrowsk. “We do feel we have some people who are solid and can fill in. To fill in if [Pedroia] plays 120, guys like Holt, a guy like Nunez coming over there. Even some depth with Tzu-Wei Lin in our organization we like a great deal.”

Our current depth charts projections, which feature Pedroia getting 74% of the team’s plate appearances at second base, forecast him for a .272/.344/.390 line with 2.0 WAR, and for the group as a whole to return 2.5 WAR, good for 12th in the majors. While the Red Sox would probably be overjoyed if things unfold that way, history suggests there may be less reason for optimism. Along the lines of what I did for Troy Tulowitzki last August, I asked Dan Szymborski to put together a list of infielders who are over 30 — shortstops and third basemen as well as second basemen, in order to start with a larger pool than the Tulo study — and missed all or nearly all of a season, then returned to the majors. I used 100 plate appearances as the upper limit for the “missed” season (Pedroia had 13), and confirmed with a variety of resources that the players’ absences were due to injuries, not detours to foreign leagues or mere attrition. As before, I considered only players from 1980 onward. The resulting list of 18 players was less than encouraging, to say the least:

One can wave off individual results here and there on various grounds. Sports medicine has improved vastly over the past four decades, and in considering Pedroia, it’s debatable how much we can extrapolate from players’ failures to recover from injuries to other body parts, such as Jeter’s ankle, the shoulders of Burleson, Glaus, and Perry, or the elbows of Fernandez, Howe (who also underwent ankle surgery), and a quartet of Tommy John surgery recipients (Furcal, Izturis, Santana, and Velarde). We can also debate the propriety of comparing the fates of players who were 30 or 38 at the start of these sequences to that of a 34-year-old.

Even so, within the timeframe, I couldn’t find any examples of an over-30 infielder who was significantly above replacement level both before missing most or all of a season and then again in his return year. Gonzalez (the “Sea Bass” one) underwent microfracture surgery in his left knee in March 2008 and did post a big 3.0 WAR season in 2010; he’s in this table twice thanks to a 2012 torn ACL. And Velarde, who underwent Tommy John surgery, totaled about 10 WAR from 1999-2001, but both needed at least one more year to recover form.

That said, three players who had dreadful seasons pre-injury did rebound substantially upon returning. Lansford, the regular third baseman for the A’s from 1983-1990 (including their three pennant-winning teams from 1988-1990) was pretty lousy in the last of those seasons, then suffered serious right shoulder and left knee injuries while snowmobiling in Oregon in January 1991. A’s vice president Sandy Alderson said at the time, “Carney’s contract doesn’t say he can snowmobile and it doesn’t say he can’t,” adding that the team declined to pursue terminating his employment. Lansford played just five games in the 1991 season, but returned for a solid 1992 season before retiring.

Fernandez, who racked up 43.5 WAR in a career that spanned from 1983-2001, endured a career-worst season as the Yankees’ starting shortstop in 1995, then fractured his right elbow in March 1996. The Yankees turned to the 22-year-old Jeter, who went on to win AL Rookie of the Year honors while helping the team win its first championship in 18 years. Fernandez missed the entire season, then resurfaced with the Indians in 1997, hitting for a 93 wRC+ with 2.1 WAR; he spent three more years in the majors and accumulated 4.9 WAR.

Spezio, a useful infielder for the Angels (6.6 WAR from 2000-2003), struggled in the first year of a three-year, $9.15 million deal with the Mariners (67 wRC+ and -0.1 WAR in 415 PA) in 2004. Between being relegated to a backup role by the free agent signings of Adrian Beltre and Richie Sexson, missing 10 weeks due to an oblique strain, and then drawing his release in late August, he was limited to 51 PA in 2005. Divorce, depression, and substance abuse — the last two of which would further derail his career years later — were contributing factors to his 2005 struggles. Nonetheless, he played a significant role for the Cardinals’ 2006 championship team, posting a 123 wRC+ and 1.3 WAR in 321 PA.

Even with that trio in my quick-and-dirty study, the group of players I identified averaged a meager 83 wRC+ and 0.2 WAR in their returns. Limiting the player pool to those whose pre-injury seasons were worth at least 1.0 WAR, which clears out some light hitters whose careers were essentially on their last legs anyway, the averages are 81 wRC+ (down from 105 pre-injury) and 0.0 WAR (down from 2.4) — an even steeper drop that probably owed something to more complete players getting longer leashes.

All of which is to say that if history is a guide, the 2019 Red Sox may not get anything close to the Pedroia they’re used to, the one who has never really had a below-average season. Even in his most injury-wracked campaigns, when his bat has been about average (98 wRC+ in 2014, 101 in 2017), he’s played at a three-win pace, solidly above average if not up to the standard that helped him make four All-Star teams and win four Gold Gloves from 2008-14. That said, even a slog of a season is unlikely to mean the end of the line for Pedroia, given that he’s still owed $40 million through 2021 via the eight-year, $110 million extension he signed in July 2013.

If Pedroia can recover some semblance of form, the stakes are high; he still has a shot at a Hall of Fame berth. He’s got the hardware (AL MVP, AL Rookie of the Year, three World Series rings, two of them as starter), but with 1,803 hits, he needs two seasons worth of playing time to reach 2,000, a number that has functioned as a bright-line test for Hall voters when it comes to post-1960 players. Likewise he’s still short in JAWS. He’s 19th among second basemen at 47.3, 9.6 points below the standard, but he’s only 2.0 WAR below the seven-year peak standard (42.4 versus 44.4). Given Chase Utley’s retirement short of 2,000 hits, Kinsler’s fade, and the PED suspension of Robinson Cano, none of this era’s aging second baseman have clear paths to Cooperstown, but a functioning if not fully resurgent Pedrioa would be best positioned to change that.

Of course, talk of any resurgence is premature when discussing a player whose availability can’t yet be taken for granted. If he’s batting leadoff on Opening Day, per manager Alex Cora’s promise for when he’ll return, Pedroia will have hit the first big mark.

Each year, we strive to improve our process for ranking prospects. Part of that is reviewing our past rankings to identify any trends or biases that may have emerged. This post doesn’t represent a comprehensive effort to do so. Rather, think of it more as a first step to finding any low-hanging fruit we may have missed.

The 2018 version of our Picks to Click was less formal than this year’s edition. Last year, we listed 62 players, representing a mix of those in striking distance of the top 100, guys we thought would improve their rankings in 2019, and players we just liked. This year, we narrowed it down further to players who have never been a 50 FV before and who hadn’t appeared on the 2018 Picks to Click list. The aim is to have a leaner list with a higher percentage of players who will become 50 FV prospects for the first time next year.

We identified 15 of the 33 players who were new to the 2019 Top 100 in the 2018 Picks to Click (this excludes Yusei Kikuchi and the players drafted in 2018, as none of them were eligible for last year’s list). A handful of the 18 we didn’t include in our Picks to Click were close to last year’s Top 100 and wouldn’t have been much of a stretch to rank; a few were late cuts from both lists. All told, we picked about half of the new Top 100 talents. The 62 players on last year’s Picks to Click essentially broke down into three equal groups: stock up, steady, and stock down. Our goal is that this year’s leaner and more-focused group of 55 prospects will beat that mark. We also found the Picks to Click group of performers with lesser tools had our worst outcome (the success was Brandon Lowe, who would’ve been shoehorned into another group if this one didn’t exist), so we axed that subset from this year’s incarnation.

On last year’s Top 100, we included honorable mentions for the rest of the 50 FV prospects, which got us to 139 total players. I’ve manually graded whether the stock of those 139 went up, down, or held steady in 2019, and added a “slight” modifier for the in-between cases. Read the rest of this entry »

Prospects “graduate” from prospect lists when they exceed the playing time/roster days necessary to retain rookie eligibility. But of course, that doesn’t mean they’re all in the big leagues for good. Several are up for a while but end up getting bounced back and forth from Triple-A for an extended period of time. Others get hurt at an inopportune moment and virtually disappear for years.

Nobody really covers these players in a meaningful way; they slip through the cracks, and exist in a limbo between prospectdom and any kind of relevant big-league sample. Adalberto Mondesi, Jurickson Profar, A.J. Reed, and Tyler Glasnow are recent examples of this. To address this blind spot in coverage, I’ve cherry-picked some of the more interesting players who fall under this umbrella who we didn’t see much of last year, but who we may in 2019. Read the rest of this entry »

After having typically appeared in the hallowed pages of Baseball Think Factory, Dan Szymborski’s ZiPS projections have now been released at FanGraphs for more than half a decade. The exercise continues this offseason. Below are the projections for the New York Yankees.

Batters

After I posted the Yankees’ WARtoon on twitter, lots of people responded by picking the overs, particularly when it came to Luke Voit. Don’t be greedy. This is essentially a mean projection of 96-99 or so wins. Didn’t you learn the meaning of A Christmas Carol? If you hoard all the WAR, you will be visited by George Steinbrenner’s ghost. Wait, that actually sounds pretty cool, unless he brings along a ghastly manifestation of the Ken Phelps trade. (What is a ghost’s favorite player? Jay BOO-hner.)

ZiPS may be low on Voit, but 2.4 WAR per 600 plate appearances is an above-average first baseman, and Voit doesn’t have a long track record of excellence. It was only in 2017 that ZiPS started being interested in him, seeing his combined major league time and minor league translation as a .283/.352/.481 line. In 2018, the combined line was .284/.361/.516, not all that different in the big picture. Voit’s not a young player, and while he has hit very well in the majors, a player’s minor league career doesn’t just evaporate the second they hit the bigs and rock the league in a cup of coffee (see: Chris Singleton or Shane Greene).

The Yankees appear to have an almost Dodgers-like level of flexibility, with a number of infielders who can theoretically play every other position on the diamond. It’s a bit of over-engineering, but ideally, top teams should accumulate depth in this manner. I’d argue that the Phillies ought to have tried to sign Mike Moustakas, though he may not have been all that interested upon realizing that he’d basically be the team’s backup in the event Manny Machado cares to sign.

ZiPS now projects Aaron Hicks to be the fifth best center fielder in baseball. It’s a weird timeline, but we’re in it. Hicks is a borderline star and the fluke talk should be more or less behind us. Read the rest of this entry »

Editor’s Note: In the run up to the January 22 Hall of Fame announcement, we were fortunate to feature a few pieces from Anthony Calamis and Adam Dore, members of Ryan Thibodaux’s excellent team that tracks public Hall of Fame ballot. This is the final such piece. Be sure to check out the ballot tracker, which is an indispensable tool for any Hall of Fame enthusiast.

Last month, the National Baseball Hall of Fame and Museum announced their latest group of inductees. Mariano Rivera received a vote on 100% of the ballots cast, while Edgar Martinez, Roy Halladay and Mike Mussina also cleared the 75% minimum.

Of the 35 players on the ballot, 11 got zero votes: Rick Ankiel, Jason Bay, Freddy Garcia, Jon Garland, Travis Hafner, Ted Lilly, Derek Lowe, Darren Oliver, Juan Pierre, Vernon Wells and Kevin Youkilis.

The results for the other 24 players are summarized below:

With 425 ballots cast, the average ballot contained 8.01 names, down from 2018’s 8.46. 42.8% of voters used all 10 available votes, down from 50.0% last year.

Nearly half of the 3,404 total votes cast were awarded to candidates who are no longer on the ballot. 1,671 votes (49.1%) went to inductees Rivera, Martinez, Halladay, and Mussina, final-year candidate McGriff, and the quintet of players who failed to received the minimum 5% to remain on the ballot.

The result of having a four-man induction class – to go along with a swan song candidate who received a vote on nearly two fifths of all ballots – is that an average of 3.93 votes per ballot will be freed up come next year. Given 2019’s 8.01 votes-per-ballot overall average, the 2020 election cycle begins with just 4.08 returning candidates per ballot.

With nearly six open spots on the average ballot heading into next year, the 2020 results are bound to look quite a bit different than those we saw a month ago. For starters, there is a strong possibility of only a single new Hall of Famer earning election via the BBWAA.

Derek Jeter is expected to draw a vote share near 100%, but none of the other first-year eligible players are expected to garner significant support. Bobby Abreu should have a decent chance to clear the 5% minimum to be on the 2021 ballot, but it’s possible none of the rest of Jason Giambi, Cliff Lee, Alfonso Soriano, Paul Konerko, Josh Beckett, Adam Dunn, Rafael Furcal, and Eric Chavez do. That will leave a lot of room for returning candidates to either begin gathering momentum or continue building on it.

Full ballot voters may add multiple new candidates next year, since many who regularly check 10 boxes have indicated they would vote for 11 or more if the bylaws permitted it.

Following the BBWAA’s publishing of voluntary public ballots, we have collected 168 of the overall 182 10-player ballots cast in the Tracker. Looking specifically at the full ballots, the average number of votes for players making their final ballot appearance was 4.46, more than half a vote higher than the overall ballot average of 3.93.

If we assume that Jeter collects 100% of the vote (he will probably receive 99% – 100% of the vote anyway), then voters who submitted a ballot with the full allotment of 10 names have an average of only 6.54 spots filled in 2020, even after including Jeter.

Since the BBWAA returned to annual voting in 1966, the record for the largest year-over-year decrease in votes per ballot is 1.23, from 2007 to 2008; only two other times has votes per ballot decreased by more than 1.0. The 1.23 vote per ballot decrease (from 6.58 to 5.35) came following the elections of Cal Ripken Jr. and Tony Gwynn, both of whom received over 97.5% of the vote. In 2008, Tim Raines (24.3%) was the only first-time eligible candidate to receive more than two votes, and his vote share was only slightly above 15th-year candidate Steve Garvey’s 21.1% showing the year prior.

The 14 players who got between 5% and 75% and will return to the 2020 ballot received an average of 4.08 votes per ballot. If voters tie the record for the largest decrease, there is still more than 2.5 votes per ballot to be spread out among newcomers and returnees next year.

There is a strong possibility that many of the 14 returning candidates receive at least a double-digit percentage increase from last year.

Overall Trends

When we last checked in on public voting data the afternoon of the announcement, there were 227 ballots logged in the Tracker. A few more ballots trickled in over the ensuing hours, increasing the total number of tracked ballots to 232 when Jeff Idelson stepped up the podium to announce the results.

Since the announcement, another 125 ballots have been made public. The 357 known ballots account for about 84% of the ballots cast, a Tracker-era high water mark.

The latest results in the Tracker are summarized in the following table, which shows how each player has done on the total public ballots, ballots released before and after the announcement, and those ballots that have not yet been released.

Among the interesting trends to emerge from the post-results data is Omar Vizquel receiving the sixth-most votes behind only the four inductees and Curt Schilling, Mussina receiving more than 75% of the vote on post-results public ballots, and Andy Pettitte tying Helton and outpolling Sheffield and others.

This marks the second consecutive cycle in which Vizquel finished markedly better in the final results than he was on track to. In 2018, he jumped from 33.5% to 37.0%, and this year he went from 38.4% to 42.8%.

The recent history of players getting 40-45% in their first five years has been a mixed bag, but Vizquel seems better positioned to make a slow climb than some of the others who have been in this range.

Jeff Bagwell began in a similar place to Vizquel but needed until 2017, his seventh year of eligibility, to gain the necessary 75%. It took Mussina a total of six years, with three coming after he entered the 40% range. Schilling and Barry Bonds, on the other hand, are still waiting and conceivably may never get there.

Encouragingly, 32 voters who did not vote for Vizquel in 2018 added him this year, the fourth-most among all returning candidates and the most among those with multiple years of eligibility remaining.

Nine voters withdrew their support for his case, but some of that could be temporary. One voter who “dropped” Vizquel cited ballot management, preferring to support Sheffield, while another added back Bonds after a temporary hiatus. Both of those voters chose 10 players. Nonetheless, the nine lost votes were tied with Manny Ramirez for the most in the cycle, and could be a sign of things to come for Vizquel.

Mussina got 76.7% of the votes overall, and was the only interesting bubble candidate heading into the announcement. The other 34 players were essentially guaranteed to either exceed 75% or fall short.

Moose was at 81.5% at the time of the announcement, and saw a drop of 6.5% from his pre-announcement public ballots to final percentage in 2018. That drop off shrank this year, however, leading to his induction. It is interesting to update data originally included in our piece from the day of the announcement.

The year before they were elected, Jeff Bagwell, Mike Piazza, and Tim Raines saw their pre-announcement public percentages fall 6.7%, 6.3%, and 5.6%, respectively. When they cleared 75%, those figures dropped to 1.4%, 3.3%, and 2.8%, respectively. Mussina benefitted from this phenomenon as well; the aforementioned 6.5% turned into a 4.8% decrease this year.

Are Walker and/or Schilling, who each saw their percentage drop by around 10% this year, be next in line? It is entirely possible both spend much of the 2020 cycle above 75% on public ballots.

Besides Mussina and the question of Rivera’s unanimity, the most interesting topic leading into the announcement was probably Pettitte and how the post-results ballots would treat him. After getting under 7% support on pre-announcement public ballots, Pettitte received a vote on about 13% of the post-results public ballots, and is on more than 14% of the private ballots. Helton had a better overall debut, but the two were similarly well-received on these later ballots.

Pettitte could stand to be a huge beneficiary of the ballot logjam easing in 2020 and 2021, as he is on just 16 of the 168 known 10-player ballots.

Inductees

The only drama concerning Rivera was whether or not he’d become the first unanimous Hall of Fame choice by the BBWAA. Now that one player has been elected unanimously, will more follow? Derek Jeter could do it next year, and perhaps Albert Pujols, Miguel Cabrera, Ichiro Suzuki, and others could in the years that follow.

Rivera’s induction breaks one of the more bizarre baseball traditions and permanently puts an end to the notion that no candidate can be a unanimous baseball Hall of Famer since nobody before him was. That a relief pitcher became the first player ever to receive every single vote speaks volumes about Rivera’s stature in the game and quiets complaints of an “anti-closer” bias.

Martinez, as expected, easily cleared the 75% threshold in his final year of eligibility after falling just 20 votes short in 2018.

In so doing, he became only the seventh candidate in the annual voting era to miss induction in each of their first two years and then subsequently be elected with a vote share of greater than 85%, following Raines (86.0% in 2017), Bagwell (86.2% in 2017), Barry Larkin (86.4% in 2012), Rich Gossage (85.8% in 2008), Billy Williams (85.7% in 1987), and Duke Snider (86.5% in 1980).

He just barely missed breaking Raines’ record for highest percentage of votes received in a player’s last year of eligibility, but did break Paul Molitor’s record for the highest percentage for any player who played a plurality of their games at DH has gotten. Molitor got 85.2% in his first year of eligibility in 2004, a hair behind Martinez’ 85.4% share.

Halladay missed out on a top 10 share for a starting pitcher, but still finished with the 16th highest percentage of any starter ever.

The inductions of both Halladay and Mussina could bode well for Schilling and make him a down-to-the-wire case next year along with Walker. Two starting pitching contemporaries making it to Cooperstown may increase consideration Schilling receives among his current holdouts. With Schilling, however, there will always be the matter of his off-field persona to contend with.

Among public 2019 voters, 27 did not vote for Schilling after having done so at least once previously. A handful of voters have been on record stating they will no longer vote for him due to his vitriolic comments, but it is unclear how many of the aforementioned 27 withdrew their support for these reasons.

Considering that Schilling lost just one vote he had received in 2018 – on a ballot that included just three names this past year – it’s fair to speculate that Schilling won’t be getting many of those 27 votes back.

Mussina’s induction continues the recent trend of players receiving under 25% of the vote before subsequently earning induction. He is the fifth player to be elected by the BBWAA after getting under 25% of the vote at some point, following Tim Raines, Bert Blyleven, Bruce Sutter, and Duke Snider.

That Mussina received 24.6% just four years ago and is now in Cooperstown offers hope to a number of candidates who have struggled to earn support in the last number of years. Multiple down-ballot candidates could be primed for a huge breakthrough in 2020 following 11 inductees in the last three elections. Any number of candidates currently in the 10% – 20% range could begin the long climb to 75% à la Martinez or Mussina

Larry Walker

The most intriguing situation on next year’s ballot will be Walker’s push for 75% in his final year of eligibility. If he were to get there, he’d be the third player in the last four years to cross 75% in their last chance.

Walker received 232 votes on the 425 ballots cast this year, 87 votes away from the magic threshold. With 193 “no” votes, he needs a bit less than half of the people who didn’t vote for him to change their minds.

Walker has at least 33 chances to gain a vote on ballots containing 10 names, and at least 42 additional chances to gain a vote on ballots containing eight or nine names with Bonds or Clemens, or seven, eight or nine names without them.

Those three groups were all ones with which Mussina excelled this year as he made the climb from below 64% to above 76%, and will be key for Walker as well.

Mussina gaining entrance this year also makes it more likely that some of the 69 ballots not included above could add him as they lose anywhere from 1 to 5 players on their ballots.

Others

The cat is already out of the bag that Walker will be the most fascinating story to follow nine months from now. Schilling has an outside chance at election as well. Beyond those top two returning vote-getters from last year, there is plenty of room for growth for numerous other candidates who will appear on the 2020 ballot.

Bonds and Clemens received the next highest vote shares of any returning candidates after Walker and Schilling, but are once again unlikely to gain much ground in 2020. After Clemens reeled in 54.1% and Bonds 53.8% in 2017, the two have seen only marginal vote increases in the two years since. Clemens increased 3.2% in ‘18 and 2.2% in ‘19, while Bonds gained 2.6% followed by 2.7% in the same timeframe.

Both will surpass 60% next year but probably not by more than a couple percentage points. It is possible that they gain some new support from voters who are typically stingier with awarding votes. Perhaps after seeing the ballot logjam rapidly clearing, voters would prefer to vote for this pair rather than return a ballot with only two or three names checked.

Expanding on the earlier exploration of his candidacy, Vizquel finds himself in a favorable position with a 42.8% share in 2019. Already more than 5% past the halfway point to election, Vizquel has another eight years to pick up an additional 32.2% – which would require right around a 4.0% annual increase. Most candidates in a similar position have historically been elected, often sooner rather than later, as the rate of increase tends to increase rapidly once a candidate holds a majority of the votes.

Overall, there are 168 public 10-player ballots. Of them, eight voters will have at most four holdovers, 71 will have five holdovers, 80 have six holdovers, and nine will have seven or eight holdovers.

Each of these candidates has at least 100 add opportunities on full ballots, and all except Ramirez and Scott Rolen have at least 95 chances to gain a vote on full ballots that already have Bonds, Clemens, and two or more of Schilling, Walker and Vizquel.

From 2018 to 2019, the percentage of voters using all 10 spots decreased from 50% to 42.8%, and a similar decrease this year would still leave a lot of space for these candidates to pick up considerable ground.

In that vein, it is worthwhile to look at how holdovers did in the following year recently.

The lower total of gained votes per ballot for returning candidates in 2018 can be explained by a very strong showing from first-time candidates that same year. Chipper Jones and Jim Thome both sailed into Cooperstown on their first try, with 97.2% and 89.8% of the vote, respectively. Both of those marks were higher than all three 2017 inductees – Bagwell, Raines, and Ivan Rodriguez.

Vizquel’s healthy 37.0% debut accounted for a higher share than Lee Smith’s final year showing of 34.2% the prior year. Rolen and Jones each received enough support to remain on the ballot.

The average votes gained jumped to more than one full vote per ballot in 2019, despite two first-ballot inductees. The big difference was that this past election followed a four-player induction class, whereas 2018 followed up a three-man class.

It is pretty much a guarantee that the 1.06 mark will improve substantially in 2020 after another four inductees and McGriff are off the ballot, and with Jeter the only newcomer likely to earn much support.

With ballots for 168 of the 182 10-player ballots, and 357 of the 425 total ballots, it is possible to get a good sense for how voters will react to the downturn in first-ballot locks over the next three cycles and better spot trends. We’re excited to pick it up again next year.

Over the course of yet another slow offseason, we’ve talked about labor relations and the free agent freezeout. But what we haven’t talked about is the opposite scenario.

So let’s take a look at answering this question: can players (and their agents) legally collude with each other? Teams colluding is somewhat straightforward: clubs make a collective decision to refuse to employ a player, or to offer a player more than a certain amount. We don’t have to go too far back in history to see what that looks like; the NFL, for instance, recently paid almost $80 million to settle claims that they did just that against quarterback Colin Kaepernick. For a baseball example, one need only look to the collusion cases of the 1980s, which ultimately resulted in ownership paying players a $280 million settlement; more recently, Barry Bonds filed (and lost) a grievance for collusion after the 2007 season.

Collusion by players wouldn’t be as simple. Players could, I suppose, all agree to not sign with one or more teams, but that would be inherently self-defeating if it restricted their own markets. More interesting would be if the players decided to coordinate on salary demands.

So let’s say that Manny Machado and Bryce Harper get frustrated at the slow-moving free agent market and tell their agents, Dan Lozano and Scott Boras, to coordinate their negotiations and agree that neither will sign for less than $300 million. Would that be collusion?

The preeminent legal definition of collusion is from Darren Heitner and Jillian Postal, who wrote a particularly excellent note on the subject for Harvard Law School’s Journal of Sports and Entertainment Law.

Collusion at its core is collective action that restricts competition. Under federal law, particularly the Sherman Anti-Trust Act (the “Sherman Act”), collusion is prohibited; however, because of labor exemptions, what constitutes collusive, prohibited behavior in specific sports leagues varies based on the league’s negotiated collective bargaining agreement (“CBA”).

Now, in case you were wondering, the word “collusion” doesn’t appear in the Major League Rules, and it doesn’t appear in the Collective Bargaining Agreement either. However, the Collective Bargaining Agreement does say in Article XX – governing the Reserve System – that rights under the CBA are individual, not collective.

The utilization or non-utilization of rights under Article XIX(A)(2) and Article XX is an individual matter to be determined solely by each Player and each Club for his or its own benefit. Players shall not act in concert with other Players and Clubs shall not act in concert with other Clubs.

That’s the language that bars collusion. As Marc Edelman explained for Forbes:

Although collusion under Baseball’s collective bargaining agreement is not identical to collusion under U.S. antitrust laws, the language and case precedence track similarly. Under antitrust law, mere parallel behavior among competitors is not enough to trigger a violation. But, parallel behavior along with a plus factor is sufficient.

That’s just a fancy way of saying that the mere fact that everyone is acting in the same way isn’t enough on its own to trigger a violation of the CBA’s collusion language. That’s why the current talk of collusion based on teams possessing similar player valuations, like the recent comments from reliever Brad Brach, is probably misguided; unless the valuations were based on a universal metric or algorithm all teams share, it’s probably insufficient to constitute collusive action.

So what does constitute collusion? Unfortunately, the CBA never actually specifies the necessary evidentiary showing. Per Heitner and Postal, “The Basic Agreement does not provide what burden needs to be met in order to prevail in this type of grievance.”

Nevertheless, we know that for players to collude in violation of Article XX(E), they’d have to not only have the same salary demands – which would be parallel behavior – but also coordinate their salary demands with each other. Now, to do that, they’d probably use their agents – and while agents aren’t mentioned in Article XX’s prohibition on collusion, they are, well, agents. Legally, under something called agency law, the authorized actions of an agent are considered, legally speaking, the actions of the principal. So if Manny Machado and Bryce Harper tell Lozano and Boras to coordinate their negotiations, the players aren’t exempt from violating Article XX(E) simply because their agents are the ones doing the coordinating.

Despite that, Major League Baseball can’t actually enforce those collusion rules against Machado and Harper or their agents. That’s because of a small oversight in the CBA. Again, from Heitner and Postal:

Article XX(E) prohibits concerted action from both Clubs and players, but the Basic Agreement fails to specify any redress if players violate the provision. The remaining sections of Article XX(E) outline the damages players can collect if they show a violation of Section E(1). Sections E(2) and E(3) provide that in addition to awarding attorney’s fees and costs, an arbitrator can award an aggrieved player treble damages, calculated from lost baseball income if the injury was the product of two or more clubs.106 Further, if five (5) or more clubs are shown to have violated Section E(1), the MLB Players Association (“MLBPA”) is entitled to reopen the Basic Agreement for renegotiation.

In other words, the CBA says that players can’t collide, but provides no punishment if they do so. That means that even if players did collude, MLB probably couldn’t do anything about it.

So why don’t players and their agents collude all the time? That’s actually a question we’ve answered already, back when we discussed agents and conflicts of interest. Remember, the MLBPA has some pretty strict rules governing agents and conflicts of interest:

§5(B)(12) – Actual or Potential Conflicts of Interest – No Player Agent, Expert Agent Advisor or Applicant shall engage in any conduct which, in the MLBPA’s reasonable judgment, may create an actual or potential conflict of interest with the effective representation of players, or the appearance of such a conflict, provided that the simultaneous representation of two or more players on any one Club shall not, standing alone, constitute a per se violation of this provision.

Lozano and Boras coordinating their negotiations would absolutely be a violation of that provision. That’s because coordination of negotiations would mean that agents are doing something other than trying to extract the best deal possible for their own clients. It might not be actionable collusion, but it might be something worse for the agents: breach of fiduciary duty, resulting in the loss of their agency certifications. And that’s why players and their representatives don’t collude – in addition to potentially getting worse deals, if they tried, their agents would likely lose their jobs.

12:02

Dan Szymborski: Powered by bourbon and a few packs of Smarties, it’s the Dan Szymborski Chat!

12:02

Dan Szymborski: Acuña projects as more valuable as a centerfielder, he can’t leverage all his talents in a corner!

12:03

Dan Szymborski: If Babe Ruth can share with like a dozen guys, Moose can share.

12:04

Dan Szymborski: It’s not even Mussina’s to hoard: Moose Haas, Moose Skowron, and Randy Milligan shared it with him with nary a protest!

Read the rest of this entry »

After having typically appeared in the hallowed pages of Baseball Think Factory, Dan Szymborski’s ZiPS projections have now been released at FanGraphs for more than half a decade. The exercise continues this offseason. Below are the projections for the Minnesota Twins.

Batters

In a league that is largely dominated by teams at the extremes of winning and rebuilding, the Minnesota Twins represent an increasingly rare breed: the excruciatingly, unbearably, average team. If you look at the lineup from top to bottom, there just aren’t any horrible weaknesses, with the bottom of the ZiPS ranks (Jorge Polanco, and the catching and first base timeshares) still living in the neighborhood of league-average; no scary negatives or zero-point-somethings lurk in there. But the highs aren’t really that high, either; Nelson Cruz tops the projections at 2.6 WAR, thanks to a .266/.348/.500 line. That’s the lowest top projected offensive WAR for any team that hopes to be a realistic contender in 2019.

This situation is not entirely the team’s fault, of course. ZiPS, like the Twins and most prospect-watchers, thought that Byron Buxton and Miguel Sano would, by this point, be players who formed the core of Minnesota’s lineup. There are a number of valid criticisms of how the Twins have managed Buxton and Sano, especially the former. Buxton’s inconsistent performance has been maddening, but so has the team’s tendency to go back-and-forth on whether he needed to be in the majors or the minors at any given time. His regular injuries add another frustrating complication. ZiPS has generally been optimistic about Buxton’s future, even when he’s struggled in the majors, but at this point, it’s only buying him as a league-average overall performer thanks to his excellent glove. It’s a good thing I don’t believe in curses; several years ago, I noted that ZiPS was rather cruel in assigning Carlos Gomez as Buxton’s top offensive comp, but it’s been more on-target than a computer can understand.

One of the problems with having a team full of average players is that it becomes damned hard to upgrade. There’s nowhere in the lineup where the Twins can be improved by any significant margin by adding a two-win player; really, even a three-win player would be an underwhelming addition at most positions. To upgrade the team’s offense, they need to add a legitimate star. And if that’s less likely to come by way of giant breakouts from Sano or Buxton than it once appeared, money becomes the best option in the short-term. After chasing Yu Darvish last year, I thought the Twins might sneakily be real players in the free agent market this offseason, but they really haven’t been. In a division with three rebuilding teams and a Cleveland squad whose plan in the outfield includes a heavy dose of Jordan Luplow, it feels like Minnesota should be more carpe diem and less dimitte diem.

Pitchers

The theme of being largely average continues with the pitching staff. Only Jose Berrios (at 14-10, 3.93, and 3.2 WAR) really pushes out of that middling range in the rotation. While his 2018 looked vaguely similar to his 2017, Berrios bumped up his strikes by a batter a game, and it still feels like that he has some ceiling left, even if ZiPS sees him continuing to pitch at about the same level. Berrios’s changeup has always looked nastier than it’s actually played on the field. I can’t speak for Twins fans, but to me, it’s looked like a pitch that should have struck out more than a mere 15 batters (against 11 home runs) in the right-hander’s brief career.

ZiPS remains unexcited about Blake Parker — one of the Twin’s “big” offseason moves — getting the highest-leverage innings, and instead sees Taylor Rogers or Trevor May (higher upside) as being better suited for the role. The projections are also bullish on Andrew Vasquez, a big lefty who serves up more sliders than White Castle and who rose from the Florida State League to the American League in ten weeks. But overall, the back of the bullpen looks shallower than other contenders’ and the highs are not quite as high. Of the teams that look to be around .500 or better, ZiPS projects Minnesota’s bullpen to be around the bottom, along with the Washington Nationals.

Bench and Prospects

Nick Gordon‘s offensive development has been a bit spotty, but ZiPS still sees him as a league-average regular in his prime, on the level of a slightly-better Polanco. That’s fine, but Royce Lewis’s projection is more exciting, with ZiPS seeing him peak as a 20-25 home run hitter, with projected lines in the .260/.330/.460 range, which will still work if he ends up playing center field rather than short in the majors. He even projects as useful if he were to play in the bigs this year, which is an impressive feat for a player with two months of Hi-A experience. As for Alex Kirilloff, while ZiPS doesn’t see him ever being a high-OBP player when his batting average inevitably comes down, his prime projections check in with just over a .280 batting average, with slugging percentages peeking just over the .500 mark.

One pedantic note for 2019: for the WAR graphic, I’m using FanGraphs’ depth chart playing time, not the playing time ZiPS spits out, so there will be occasional differences in WAR totals.

Ballpark graphic courtesy Eephus League. Depth charts constructed by way of those listed here at site.

Disclaimer: ZiPS projections are computer-based projections of performance. Performances have not been allocated to predicted playing time in the majors — many of the players listed above are unlikely to play in the majors at all in 2019. ZiPS is projecting equivalent production — a .240 ZiPS projection may end up being .280 in AAA or .300 in AA, for example. Whether or not a player will play is one of many non-statistical factors one has to take into account when predicting the future.

Players are listed with their most recent teams, unless I have made a mistake. This is very possible, as a lot of minor-league signings go generally unreported in the offseason.

ZiPS’ projections are based on the American League having a 4.29 ERA and the National League having a 4.15 ERA.

Players who are expected to be out due to injury are still projected. More information is always better than less information, and a computer isn’t the tool that should project the injury status of, for example, a pitcher who has had Tommy John surgery.

Both hitters and pitchers are ranked by projected zWAR — which is to say, WAR values as calculated by me, Dan Szymborski, whose surname is spelled with a z. WAR values might differ slightly from those which appear in full release of ZiPS. Finally, I will advise anyone against — and might karate chop anyone guilty of — merely adding up WAR totals on a depth chart to produce projected team WAR.

Free agency was supposed to be different for Mike Moustakas this time around. Last winter, coming off his second All-Star appearance and a career-high 38 home runs, the longtime Royals third baseman turned down a $17.4 million qualifying offer in anticipation of a multiyear contract, only to receive a frosty reception on the open market. Ultimately, he made a belated return to Kansas City on a one-year deal, was traded to the contending Brewers in late July, and hit free agency again this fall — this time without a qualifying offer and associated draft pick compensation. Yet once again, he only landed a considerably discounted one-year contract. He’ll return to the Brewers on a $10 million deal that includes a mutual option for 2020.

Moustakas, who turned 30 on September 11, hit a combined .251/.315/.459 with 28 homers, a 105 wRC+, and 2.4 WAR in 2018. His offense wasn’t quite what it was in 2017 (.272/.314/.521, 114 wRC+) but his defense improved significantly in terms of both DRS (from -8 in 2017 to 2 in 2018) and UZR (from -4.5 to 1.0). Given that he was finally more than a year removed from the right ACL tear that limited him to 27 games in 2016, that shouldn’t be tremendously surprising; according to Statcast, his average sprint speed also increased substantially, from 24.0 feet per second to 25.7, though even that puts him somewhere around the 15th percentile of major league regulars.

Moustakas’ defensive rebound helped offset his slight offensive downturn; his overall WAR was a bit ahead of his 2.1 from 2017, if still shy of his career-best 3.8 in 2015. He’s not what you’d call the most disciplined hitter; his 37.0% rate of swinging at pitches outside the zone places him in the 15th percentile among qualifiers, while his 7.7% walk rate puts him in the 35th percentile. He doesn’t strike out much (16.2%, 72nd percentile), but even so, 2018 saw his highest K rate since 2012, his first full major league season. A pull-happy hitter, he had 372 plate appearances against a traditional shift, the majors’ third-highest total. He hit for just a .243 average with a .311 slugging percentage and a 43 wRC+ in those plate appearances, which is 14th percentile stuff. Combine that with his non-blazing speed and you get a .259 BABIP, the majors’ 15th lowest among qualified hitters.

Read the rest of this entry »