Maybe Greg Holland Made An Inspired Choice
Yesterday, Greg Holland reportedly agreed to sign with the Colorado Rockies. As a guy coming off arm surgery, looking to re-establish himself as a premier reliever and rebuild his value, going to Denver seems to be an odd choice. As Travis Sawchik noted this morning, the recent history of pitchers escaping Colorado and finding significant paychecks are not great, and of course, because of how the park plays, Holland’s numbers are likely to be worse this year than if he had agreed to sign closer to sea level.
Generally, we’re used to players looking for big contracts next year signing in venues that fit their skills, and these type of one year deals are often called “pillow contracts”, but there’s nothing soft and comfortable about pitching in Coors Field. Maybe we need a new name for Holland’s choice? “A bed of nails contract” doesn’t roll off the tongue quite so easily, but more accurately portrays the situation Holland seemingly placed himself in.
But in thinking about why Holland would go to Colorado, I think we need to acknowledge that the game has evolved, and the methodologies for evaluating player performances have changed dramatically. And given those changes, maybe Holland didn’t just take a short-term cash-grab that puts him in a worse position for next year; maybe he made a choice that could actually be beneficial to his future earnings.
The conventional wisdom around a pitcher going to a place like Seattle, San Diego, or Miami to rebuild their value is based on the idea that bigger ballparks will suppress home run totals, and perhaps pitching in a more friendly environment will enable the pitcher to be more aggressive in the strike zone, challenging hitters in a way that they are less willing to do in a park that penalizes mistakes a bit more. The more friendly environment helps you put up better numbers, and then with those better numbers, you land a nice big paycheck the next winter.
Except that’s not how baseball works anymore.
For instance, Mark Trumbo moved from Seattle to Baltimore, hit 47 home runs in a hitter-friendly ballpark, and was mostly rejected during free agency, eventually returning to the Orioles for half of what he reportedly wanted at the start of the winter. No player in this class better exemplified the kind of context-based improvement that pillow contracts are supposed to provide than Trumbo, even though he didn’t intentionally choose to go to Baltimore, but the market simply didn’t respond to the change in results.
And the reality is that, while Holland’s ERA would probably be better in some other park this year, there isn’t a team left out there that is paying pitchers based on their raw ERA numbers. Every organization in the sport is maintaining an internal database of performance metrics, and all of them are park and league adjusting those numbers. The days of a team getting fooled by a shiny ERA in a pitcher’s environment are over. If you run a 3.25 ERA in San Francisco or a 4.00 ERA in Colorado, every decision maker in baseball is aware of the fact that those are equally impressive performances.
So I don’t think Holland’s next paycheck is going to be all that harmed by the fact that his numbers will be superficially worse because he’s pitching in Colorado this year. If a team wants to sign Holland next winter, they’re going to look at whether the stuff came back, whether he was missing bats like he used to, and whether he was able to pitch on back to back days. They’re going to look spin rate way before they look at ERA, and they’re going to account for the fact that breaking balls don’t work as well in Colorado as they might elsewhere. Any team that wants to sign Holland next winter will be asking what he might do for them in 2018, and paying him based on the answer to that question, not based on what his 2017 numbers looked like at altitude.
And there are reasons to think that joining the Rockies might set him for a better paycheck than if he had signed with, say, the Washington Nationals. The Nationals are one of the best teams in baseball, and barring some kind of disaster, they’ll be looking to upgrade their roster all year long; it would take a lot of things going wrong to imagine Mike Rizzo acting as a seller at the trade deadline. If Holland joined the Nationals, he was probably going to pitch for them all year, and if he had a successful comeback, the Nationals probably would make him a qualifying offer after the season.
Now, the qualifying offer is changing next year, but it still has a real chance of impacting Holland if he has the kind of rebound season he’s clearly hoping for. Under the new system, a player who receives a qualifying offer will still cost a signing team some future assets; depending on the team’s market size and luxury tax status, they’ll either forfeit a second and fifth round pick plus $1 million of their international spending pool, a second round pick and $500,000 of their international money, or a third round pick. And because of the new hard-slot international pools, that loss of spending money could end serving as a pretty significant deterrent, and it’s aimed directly at teams that would be most likely to spend big on relief pitchers; the Dodgers and Yankees specifically.
By joining the Rockies, Holland likely made it far more likely that he’ll be a true unrestricted free agent next winter. Even though the Rockies consider themselves contenders, the projections don’t really see it that way, putting them at 78-84 at this point. The Rockies aren’t a guaranteed also-ran like the Padres, but there’s a pretty good chance that the Rockies will be sellers at the trade deadline. And if Holland has had a strong first few months of the year, there won’t be many trade chips in baseball more attractive than a rental relief ace.
So either the Rockies win, which Holland would probably enjoy, or they don’t win and Holland gets traded to a contender mid-season, which would mean he cannot receive a qualifying offer next winter. Either of those seem like good outcomes. There is, of course, the scenario where Holland doesn’t pitch well or stay healthy, but if that happens, well, he probably took the most guaranteed money on the table, so even then, joining the Rockies was probably the best idea, since he maximized his earnings this winter.
For Holland’s future earnings, this very well might have been the best choice he could make. While his breaking ball might not drop as much in Denver as it would elsewhere, and his park-unadjusted numbers will probably appear worse than if he went to some pitcher-friendly ballpark, future employers will account for those factors and understand that they are out of his control. And by joining the Rockies, Holland made himself less likely to get saddled with the QO tax next year, setting himself up to be bid on by teams like LA and NY, which can only help his potential 2018 earnings.
Going to Colorado in your rebuild value year might not be conventional, but in this case, I think Holland might and his agent might have made a really smart decision. And we should probably stop thinking that players should try to go fool their future employers by putting up superficially decent numbers by choosing friendly ballparks to rebuild their value. Baseball just doesn’t work like that anymore.
Dave is the Managing Editor of FanGraphs.
Any team looking to sign Holland next offseason is going to have a problem, given that the deal includes a vesting option for 2018.
This would be a great comment if:
1. it wasn’t so snarky AND
2. it mentioned that, if the contract vests, Holland makes $14 million next year, a pretty good scenario for him
3. it also mentioned that, if he gets traded, it probably doesn’t vest at all since the vesting option is based on number of games finished.
Keep up the diligent work Fangraphs’ policeman.
“Every Article, Every Day” (the Fangraphs’ police dept. motto for those that didn’t know)
Was this one of those old “salute to/Real American Hero” bits from those 90’s Budweiser commercials?
This is a question asked in all ignorance: is there any reason to believe that, because of the affect Coors Field has on curveballs in particular, Holland might not recover as effectively? That is to say, for a pitch so dependent on feel, will playing 82 games in an environment where he will most likely rely more heavily on his slider than his curveball, dampen his ability to throw the curveball when away from Coors?
Or would side sessions and the such provide ample opportunity for Holland to maintain confidence in his curveball?
I was going to ask this question. I understand that pitching in a nondesireable location and its results on ERA are not something MLB Execs use to make decisions. But, 3 things, doesnt that altitude really screw up breaking pitches?
And, why not just go to Cincinnati?
and lastly, while execs understand ERAs depend on context, the reality of pitching in Coors is that a few unlucky bunching of HRs or runs could really rob Holland of high leverage situations, right? He needs to pitch in high leverage and pitch well for 2018, but if he’s removed from a role because some HRs went out in Colorado that wouldnt have happned in Cincy or SD, isnt that a legit concern? There’s far more of a safety net in every single other ballpark….
Coors is only the 3rd worst HR park (http://www.espn.com/mlb/stats/parkfactor/_/sort/HRFactor). Cincinnati is 4th worst. That’s an old Coors bias
He’s thrown curveballs less than 4% of his pitches over his career, and less than 1% in ’13-’14 when he was pretty good.
During his best years, Holland essentially never threw a pitch that was tracked as a curveball. The few he threw graded incredibly well, but I suspect it’s because those are sliders with unexpected movement that trackers got wrong. He went back to the curve a bit in 2015, but that was also the only year of his career that his slider graded out lower than 8. He might have been mixing in the curve more to give hitters another look, or his elbow issues might have started making his slider look like a curve to pitch trackers far more often.
This makes me wonder if you’re asking the question backward. Holland could use Denver as a reason to focus on just getting his two best pitches right and not worry about anything else. Put another way, if Holland’s fastball and slider are working, does he need a good third pitch?
My understanding is it’s harder to perform at altitude, period. Takes more out of you, thus makes you more susceptible to injury. I wouldn’t worry about GMs misreading my numbers, or any ‘lack of leverage’, or whatnot. I’d sure research the health aspects of it best I could, tho’. And go elsewhere unless that research really made me feel secure about pitching there.
I think there is also a psychological toll on pitchers. I have heard about this from minor leaguers in the California league, where there are a bunch of parks in the high desert. If one of your pitches doesn’t work as well as you expect, or if pitches that used to get weak contact end up getting flared for hits, then some pitchers react badly and start to change their approach. When a pitcher stops pitching to his strengths, or starts to lose his location because he’s trying to be too fine or something, then the result is usually that he simply pitches sub-optimally, which is another drag on performance on top of the Coors penalty. I think it takes a mature pitcher to take bad outings in stride and know the difference between a bad game and a good game in a bad park.
As one example, Matt Bowman made this exact point in an interview with David Laurila – http://www.fangraphs.com/blogs/matt-bowman-on-leaving-las-vegas-for-st-louis/
I *think* I was vaguely thinking, in part, of this interview!
If you’re a mature pitcher, your ERA- might benefit by moving to altitude. You’ll get hurt some because pitches don’t move, other pitchers in the park will get hurt more because their pitches don’t move *and* they start throwing unwisely.
“The days of a team getting fooled by a shiny ERA in a pitcher’s environment are over.”
As a Dbacks fan, it woulda been real nice if someone woulda told Stewart this last winter. Ya know before he destroyed the team.
I thought this was an excellent line, as well. After teams bought into the superficial numbers, I would often ask myself why there weren’t digging further into the park-adjusted metrics that have been available. Announcers and casual fans don’t ever pick up on the advanced metrics, but I would think that MLB front offices would. And, as Dave says, now they seem to be acting more rationally.
I think there are a few problems here. First, results do matter. Front offices certainly adjust their expectations for a player based on context, however, players adjust their approach based on results, and their confidence is also dependent on results. Successful pitching requires repeatability of mechanics, and slight changes in mechanics can lead to big changes in pitch outcome. Poor results in Colorado might lead to a change in approach, and consequently to unlearning all of the good things that made a pitcher successful elsewhere. Do pitchers have to relearn how to pitch in normal atmospheres? They certainly have to rebuild their confidence after pitching there!
Second, the extent to which park factors correctly adjust for context for any individual player is a function of how different that individual player is in being helped or hurt by the park to the average player. If Holland is hurt by Coors more than the average player then the adjustments the GMs make for his performance may not credit him enough. On the other hand, if Coors hurts Holland less than other players he might trick the GMs into believing he was better than he actually was!
1. This what I see as the real risk here. It seems like for some pitchers, the Coors penalty is actually worse than you would expect just based on park adjustment. I also feel it is partly why the Rockies haven’t developed any good pitchers, even adjusting for park. It is that much harder to develop an optimal approach and the confidence that comes with that when the altitude causes even the right approach to produce some stretches of misery. For this to work, Holland has to know himself well enough to know when to stick with his process through a bad stretch and when to change it.
2. Considering he is a hard-fastball, two-pitch guy, I think it is not likely he would be disproportionately affected. Velocity is the one thing that definitely plays at Coors, and if he doesn’t have that back, then his comeback is in trouble no matter where he signs. His slider may be less effective, but it should still serve as a change in speeds, another thing that works fine there. Looking at the movement, it is not particularly large in any direction, so it shouldn’t change too much. I think some of the pitch’s effectiveness in the past has come from the fact that there is a difference of more than 10 inches of vertical movement between his fastball and his slider, meaning it’s basically not possible for a batter to see one and adjust his swing to hit the other. If his fastball rises less in the thin air, that differential may decrease, and that may hurt him. However, as a reliever, I don’t think it will hurt him as much as it would if he had to face the same batters multiple times a game.
Colorado pitchers have said you never get a chance to adjust to the altitude because of travel. I could see that impacting more than just results.
Dave,
I don’t expect you to be same caliber writer as Clayton Kershaw is a pitcher but this was Dillon Gee level. Let’s get back in the lab and at least up your game to Nolasco level with Spring approaching.
Could also be simply a case of who would give him a better chance to be a closer. Not like the Rockies are overflowing with top quality relievers, so you gotta take the opportunity to be a closer while you have a chance. And if they end up better than projections, maybe having a “proven closer” will be a big reason for that.
I certainly wouldn’t say the Rockies are “overflowing” with bullpen talent, but Ottavino (who the Denver Post suggested is still the “clear favorite” to start the season as closer, fwiw) is the real deal. He didn’t come back from TJS until July, but he was a top-10 pitcher by xFIP in the second half last year.
http://www.fangraphs.com/leaders.aspx?pos=all&stats=pit&lg=all&qual=20&type=8&season=2016&month=31&season1=2016&ind=0&team=0&rost=0&age=0&filter=&players=0&sort=18,a
Now it’s possible Bud Black could be more creative with reliever usage than Weiss was- it would be hard not to be- but there were certainly better options than the Rockies if closing was so important to Holland.
I think most of you guys are thinking too hard about this, at least harder than Greg Hilland probably did. It may be that Colorado simply offered him the best deal, and he took it. His agent would understand the benefit of going some place where Holland was unlikely to incur a QO penalty. Dave’s logic is sound.
I think you need to consider the mentality of dominant closers when you look at the decisions they make. They pitch in high leverage situations are the time. There is plenty that can go wrong every time they take the mound. But they block it out. There is no room for self-doubt. So arguments that begin with, “Well…if things don’t go well for him in Colorado…” may not hold much sway with them.
That’s why he has an agent.
“there’s a pretty good chance that the Rockies will be sellers at the trade deadline.”
Well, a pretty good chance they *should*, at least.
Has any research been done on park factors for pitch movement? Specifically, by individual pitch rather than the general knowledge that everything moves a bit less. I’m interested to know what type of pitch would be affected most.
I believe the physics is pretty straightforward: it’s the air density. Low density, low Magnus force deflection from the same spin. Also lower drag.
That reminds me I’ve always wondered, doesn’t a fastball in Coors reach the batter some milliseconds faster than usual? Doesn’t that help, like releasing the ball forward helps, cutting into batter decision time? It seems like velocity should play up in Coors.
I enjoyed the article and I think it asks a worthwhile question, but I also think this part is the only one that really matters: “he probably took the most guaranteed money on the table”