The M’s 2015: The Risks
With opening day tantalizingly close, and the M’s set to open their most anticipated season since 2010, er, 2008, in a long time, the M’s remain the favorite in a revamped AL West. As we know, of course, projections don’t win actual games, and everything from trades to injury to breakout years and collapses mean that the season won’t play out the way PECOTA/ZiPS/Steamer/baseball writers see it. The M’s could be much, much better than their projection, and we will retroactively identify and scoff at the deficiencies in the projections. Alternatively, the M’s could underperform significantly, as anyone who’s followed this team for a while knows. So today and tomorrow we’ll take a look at some of the areas that could cause the M’s actual win total to diverge from the projection consensus of somewhere around 87-90 wins. We’ll start today with some of the reasons the M’s might underperform, because pessimism feels like an old comfortable old jacket at this point.. Ahhhhh.
1: Mike Zunino’s grasp of the strike zone remains tenuous
Mike Zunino did a brilliant job behind the plate last year, and the more we learn about catcher framing, the more we understand that Safeco field isn’t the only thing making the M’s pitchers look better than they otherwise would. He’s a leader, has the tools to become a perennial all-star, and just turned 24 last week. He also put up a .254 on base percentage last year, knocking 22 HRs but costing the M’s runs on offense thanks to his all-or-nothing approach. Zunino is both a catcher, and a catcher who hits freakishly few ground balls, so his BABIP is never going to be very high, but however many HRs you hit, striking out in a third of your plate appearances AND putting up a low BABIP is a recipe to getting reaaallly familiar with the Mendoza line.
There are two distinct issues here. The first, and the one that’s working very hard at correcting, is his strike zone judgment. Last year, he swung at nearly 40% of out-of-the-strike-zone pitches, up 10 percentage points from his call-up in 2013. As a result of this ecumenical approach to swinging, his K:BB ratio tanked – his 0.11 BB:K ratio was tied for the worst in MLB for anyone with at least 300 PAs. If this number is below the 0.16 put up by Javier Baez last year, you have a serious problem. What’s worrying is that it doesn’t appear that opposing pitchers have fully plumbed the depths of Zunino’s obsession with swinging. They threw him a slightly below-average number of strikes last year, but there’s some room for that number to drop. Pitchers understand that Pablo Sandoval *wants* to swing, and thus he gets far fewer strikes than average (or Zunino). As long as his O-swing stays around 40%, he’ll see his strike percentage drop, as pitchers learn there’s no need to risk strikes.
Second, it’s obviously a problem if Zunino swings and misses at breaking balls diving out of the zone. His other problem, though, is hitting them. Scott Lindholm had a great article on hitting pitches out of the zone a few months back at Beyond the Box Score. Within that is a great Tableau visualization/table comparing each batter’s BA and SLG% for in-zone pitches and out-of-zone pitches. It’s well worth diving into and playing with. What it’s showing you is that there are a few different ways to be a good hitter. For those hitters with massive, massive gaps in in-zone vs. out-of-zone production, you need to do serious damage on strikes. So, Jose Abreu, Devin Mesoraco, Giancarlo Stanton, Steve Pearce, Adam Dunn are all clustered near the top. Freakish, 70-80 grade power helps, but so does plate discipline – that’s why Dunn and Stanton are here.
That’s not the only approach, however. Sandoval has a predictably small gap. If he had his approach and wasn’t successful, he’d have washed out of the league years ago. As it is, he’s actually a pretty decent bad-ball hitter, and while his stats are worse, they’re not atrocious when he hits a ball. Same thing with Adam Jones, another guy with a poor BB:K ratio and some pop. He doesn’t really recognize the strike zone all that well, but he can hit bad balls well enough that the discipline issue isn’t critical. The final approach is just being good at everything, and that’s where we find Mike Trout, who destroys strikes and is suprisingly decent on balls, especially given his well-known struggles with high fastballs. In any event, just as you can be effective with a really small gap between strikes and balls, or with a really huge gap, you start to see the flip side of the coin, the evil Spock approach that could produce similar results. For example, if you’re not strong enough to do serious damage on strikes, then you’re going to have a Sandovalian gap but without Sandovalian production – this is the sad state that, say, Endy Chavez and Justin Smoak find themselves in. With Smoak in particular, it’s not that he didn’t recognize the strike zone – he was pretty good at that. It’s that even when he got his pitch, he couldn’t do much with it. And what the gap doesn’t tell you is the rate at which you put balls and strikes into play. Thus, having a very large gap isn’t helpful if you put too many bad pitches in play. Unfortunately, this is the situation Zunino finds himself in. He’s not yet good enough to SLG *enough* on balls that 2-strike counts aren’t a kind of death sentence. But unless he tightens the strike zone, he’ll keep seeing more and more balls.
The good news here is that Zunino and the M’s are very aware of the problem, and Zunino’s spent the spring giving us reasons to believe he’s made serious strides in plate discipline. It’s tempting to look at his spring and his birthdate and assume that this is something he’s rapidly growing out of, and that he’ll improve his BB:K over time like most hitters. But the cautionary tale here is that of another young catcher whose power helped him overcome a terrible BB:K ratio for a while, but who never improved on his first full season in the big leagues. Look on the works of JP Arencibia and despair, M’s fans. Arencibia blew through the low minors and then had a bit of a hiccup in his first taste of AAA. Returning to the level and the deliciously hitter-friendly Cashman Field, Arencibia uncorked a slash line of .301/.359/.626 in 2010. In 2011, he was the Blue Jays starting C, hit 23 HRs, and though the .282 OBP wasn’t ideal, Arencibia could improve on his plate discipline with time, right? Instead, he’s been unable to make adjustments, and his BB:K ratio got worse. His o-swing% started high, but climbed a bit more in the years that followed, culminating in an abysmal 2013, with a BB:K ratio of 0.12, a .194 average and a .227 OBP. He was a 27 year old catcher who hit 21 HRs, and the Blue Jays didn’t make him an arbitration offer, effectively cutting him. He’s now in the Orioles’ minor league system. Oh yeah – the spring before that career-nuking 2013 season? The one where his approach cratered? That spring, Arencibia hit .439/.477/.902.
That’s a worst-case scenario, clearly, but it’s a bit terrifying that Zunino’s o-swing and BB:K is ALREADY where Arencibia’s ended up in 2013. He’s younger, and honestly, his spring has produced some reasons to think he can get a lot better at the plate in 2015. It’s great that the coaching staff have worked on this so much with Zunino, because as we saw with Zunino from 2013 to 2014, or Arencibia from 2011 to 2013, it’s not enough to say that experience will take care of the problem. Fixing the problem takes care of the problem.
2: The bullpen acts like a bullpen, and puts up very different numbers this year
OK, you could write this about any team, in any year. High variance isn’t the point of a bullpen, but it seems to be the price we pay for the highly specialized and generally highly effective modern reliever corps. In 2014, the M’s had the lowest bullpen ERA in baseball. They had the best strand rate, even platoon splits, and a very high strikeout rate. While FIP didn’t love them for a few reasons, they were clearly an asset for the club, and you see that clearly in their 4-in-the-AL win probability added. The bullpen is largely unchanged this year, so can we chalk up another sub-3.00 ERA? The problem is that in 2013, the M’s had the *worst* strand rate in baseball and a terrible ERA *despite* an even better K rate than they had last year. Some of this is just sequencing – the difference in the bullpen’s FIP from 2013 to 2014 isn’t *that* big, but the M’s were unlucky on balls in play and sequencing in 2013, but very lucky on balls in play and sequencing in 2014.
So far, so orthodox. Regression to the mean, people! But this isn’t purely about luck. The specific relievers the M’s have come with a specific set of skills, and a specific set of concerns. The M’s got a brilliant rookie season from Dominic Leone last year, as the righty used a 95mph fastball to rack up a 25% K rate and an ERA hovering around 2. The issue isn’t so much that he “beat” his FIP, it’s that he was never a huge swing-and-miss guy in the minors. In 2013, he had one of the lowest swinging strike rates for the Jackson Generals, and his K% in the A+ and AA didn’t approach the mark he put last year. Some times pitchers figure something out and become far more hittable than they were in the minors (or in college), but it’s equally likely that an out-of-nowhere reliever can put up great numbers and then struggle to sustain them – Mark Lowe in 2006, Julio Mateo in 2003, etc. Dominic Leone has the tools to be good for a while, but it was just odd that big leaguers found him harder to square up than the California League. Spring training generally means nothing, but it makes me squeamish to see Leone give up 17 hits in just seven spring innings. Obviously, if your fourth or fifth righty in the pen is your problem, you’re living right. If Leone tanks, the M’s can swap in Carson Smith and not miss a beat. If it was *just* Leone, that’d be one thing.
It’s not, though. Not while Danny Farquhar’s velocity seemed to be down this spring. It’s always impossible to know what to do with that – maybe he was working on something, maybe he was tired one day and that screwed up the overall averages. Still, let’s remember that Farquhar was a frequent guest of the waiver wire, and struggled to latch on until he started throwing 95. Sure, the big change was a mechanical overhaul, and it’s not like Farquhar’s throwing sidearm again, but this is a guy who threw 89mph for years, and then induced his small frame to throw 95. His velocity dropped by over 1mph last year, and he proved he could still be effective at 94. However, Farquhar averaged 92.6 in Arizona. If he’s around 92 in Seattle, can he still avoid HRs the way he’s done the past two years? Farquhar’s smart and knows several pitches. He ditched his sinker in 2014, and though his four-seamer wasn’t quite as good in 2014, his cutter could compensate. Changing his approach – switching back to the four-seam and curve, say – can help him accommodate lower velo, but it’s not a fix. If he comes out throwing 95-96 again, then the point is moot. For the M’s sake, let’s hope he does.
That brings us to Tom Wilhelmsen, the most enigmatic/important of the M’s relievers. The Bartender led the bullpen in WPA last year, and he paired the biggest workload with the lowest ERA. We all know what he’s capable of, and that’s part of the reason his 2013 was so disappointing. Wilhelmsen struggled with his control all year, and that made his top-shelf stuff play down. He didn’t get as many strikeouts in part because he was behind in counts. His overall results were much better in 2014, but it’s worth remembering that he didn’t entirely solve the problem. His walk rate was still over 11%, and his K rate was good, but not in line with his excellent FB/CU combo. For a variety of reasons, Wilhelmsen’s been inconsistent, trading months with off-the-charts performances with months where he couldn’t find the zone. What helped him tremendously last year was his insane .200 BABIP. A pitcher with great velo can run lower-than-average BABIPs, but .200 just isn’t sustainable. To compensate, he needs to get back to the form he showed in 2012, with a 27% K rate and a walk rate safely under 10%.
So, finally, we’ve got the Fernando Rodney experience. Rodney is 38, and coming off a year in which his fastball lost 1.5MPH. It’s still a good pitch, and the change-up is mind-altering, but this is still the guy who was cut in 2011, and whose grasp of the closer job seemed shaky in April/May. His dominance the rest of the way was a huge part of the M’s run to contention, and if they want to get past the Angels/A’s, another year from the archer would really help. I’m wary of all non-Felix pitchers, though, and 38-year old closers are rare enough that it’s tough to feel absolutely safe about this. As Fangraphs and others have noted, the M’s could have the best bullpen in baseball this year. If one or two of the guys listed here struggles, they could still have an excellent group that adds value to the team overall. But the potential – however remote – is there for a more systemic collapse, or at least regression, that spells the difference between the division and a one game crapshoot, or the playoffs and another frustrating season. We need those arrows to be real.
3: The *good* Austin Jackson never returns
In the aggregate, we can say something meaningful about player aging curves, but that doesn’t help you understand an individual player. It’s nice to know when the median player peaks, or how gradually the average player (who is allowed to decline on a major league roster) declines. In the face of all of these averaged, combined numbers, we have the actual records of thousands of individual cases – and while that can help us understand just how uncommon a particular career arc is, it by definition can’t help identify the statistical anomalies, the great players who defy aging and those whose aging seems to run in fast-forward.
Since his 2010 debut, everyone has *known* what would doom Austin Jackson’s batting stats. After posting a sky-high .396 BABIP in 2010, it’s been easy to assume that his stats would regress. The problem is that Jackson continued to post sky-high (if not quite as high as .396) BABIPs, while also making each ball in play do a bit more damage. Not only that, but Jackson transformed himself from an extremely high K% hitter to someone with essentially average K rates. All of this speaks volumes about Jackson’s ability to adapt and change his approach. That’s good, because it’s crucial that he make further adjustments.
As we all saw in 2014, the version of Austin Jackson that the M’s acquired from Detroit turned out to be a terrible hitter. Jackson’s K% crept back up to 25% after the move, but that’s still below his K% in Detroit in 2010 and 2011, when he was a moderately effective hitter. His BABIP was low for him, but not extraordinarily low for the league. Instead, Jackson’s problem was that his contact produced shockingly weak results. I mean, NL pitcher-esque results. With the M’s, over 236 plate appearances, Jackson produced an ISO of .031, or exactly the same as NL pitchers managed in 2014. As many have pointed out, his production on fly balls was particularly poor. But anything abnormally poor should regress to the mean as surely as the M’s bullpen’s strand rate, right? Wellll, this is where the difference between a population and a specific player becomes pretty important.
Jackson’s ISO on fastballs dropped from .189 in 2012 to .073 in 2014. It is essentially impossible to remain a legitimate MLB starter at that level. James Jones’ SLG% on hard pitches was .080, for reference, and James Jones is not, as of this date, a legitimate MLB CF. Because Jackson’s been effective in 2012-2013, the projection systems forecast something of a bounce-back for 2015. It makes sense – if you average his production over the previous three years, you come out with a pretty good CF.
Only that ignores the fact that his ISO and BABIP have been moving in tandem. Jackson’s sort of the flip side of Mike Zunino, in that his plate discipline has actually gotten better over the years. He’s swinging at fewer balls, swinging at fewer pitches overall, and making more contact. This is the blueprint of a hitter who gets more selective in order to do more damage on each ball in play, only Jackson has left out the all-important do-more-damage-on-each-ball-in-play part. Like we talked about with Zunino, pitchers are going to notice they aren’t getting punished for leaving pitches up or in the zone, and that’s going to put even more pressure on Jackson’s walk rate. Jackson’s walk rate was as low as it had been since 2010 last year, but pitchers started to throw him a few more strikes in 2014. If they continue to do so, and if Jackson can’t run the .350+ BABIPs that sustained his first few years in the league, then he’s in trouble. He’s either got to do much more damage on balls in play, or he’s got to be extraordinarily selective (and great at defense). Unfortunately, he’s already exceptionally selective, so there just isn’t a whole lot of room for improvement there.
There IS, obviously, room to improve his ISO and his production on balls in play. Angel Pagan looked like he was winding down at age 29 before leading the Giants to a World Series title at age 30. Marlon Byrd wasn’t really a decent player *until* he was 29-30. Brian McRae’s career arc followed Jackson’s, with some early struggles, some success around age 27, some bad years after that, but then a nice bounce-back at 30. Clearly it’s possible to be productive after a year as barren as the one Jackson just suffered through. But not everyone manages it, and the problem is that after shipping Michael Saunders to Toronto, the M’s just don’t have a plan B.
Last year, the M’s turned to James Jones when Abe Almonte proved himself unable to handle the CF job. When his lack of pop and/or discipline proved too great an obstacle, the M’s traded for Jackson. That Jackson proceeded to put up a half year even worse than Jones’ line puts the M’s CF depth into question. At this point, the M’s have Jackson as the unquestioned starter, with Justin Ruggiano and Jones behind him. Ruggiano is a competent back-up, but he’s generally a platoon player hitting from the right side. Jones can hit lefty, but hasn’t proven he can hit at all, at this point. The M’s lack of a half-decent CF spot doomed their 2014, and their plan for 2015 is essentially, “What are the odds THAT happens again, right?”
Again, none of this is to suggest that Jackson is doomed, or that a bad year from Jackson (or Zunino, or Leone) will doom the M’s. It’s just that there are certain positions where you can kind of see the weak points in the armor. One problem the M’s have had in recent years has been unbalanced line-ups, with some decent players, some so-so guys, and then some automatic outs. Each year, the M’s have attempted to rid their line-up of Brendan-Ryan-style out machines, and they’ve done a decent job of it. Having one up-the-middle-defender put up an awful batting line won’t harm the M’s, but two might. One plus a bullpen decline might prove critical. Obviously, the M’s have some upside risk as well, and all of the risks we’ve talked about have equivalents in their divisional rivals. But if the M’s underperform, I’m betting these three issues will prove critical.
What do you think? What are you worried about for 2015?
Erasmo Ramirez to Tampa in Minimally Surprising Trade
You always wanted to blame a physical injury. Something small enough to go undiagnosed, but that changed his mechanics somehow. He’d always shown good control, but he was putting up average-to-mediocre walk numbers. Nothing looked wrong with the velocity, and every so often, a perfectly disguised change up would flutter in and befuddle some talented hitter. It – the combination of stuff, approach, confidence – was always on the verge of returning. After a few years of this, it’s someone else’s turn to wait and wonder if “it” was ever really there, and if it is ever coming back.
The above could apply to either Erasmo Ramirez or Mike Montgomery, two pitchers who’ve frustrated fans with promise, development and then years of stagnation. Tonight, the M’s have traded Ramirez to the Rays in exchange for the LHP Montgomery. Ramirez will get a chance to start for the decimated Rays while Montgomery figures to transition to the bullpen, first in Tacoma and later, hopefully, in Seattle.
Erasmo Ramirez was the ultimate underdog story – a kid from a small town in Nicaragua whose first pitching coach was his grandmother, and who bumped into a tourist who set him on a path to a boarding school in a neighboring country. Always undersized, he was doubted at every step, but who kept getting better until he couldn’t be kept down anymore. He first drew attention in M’s blog-land for his absurd K:BB ratio in the Venezuelan league in 2009. That earned him a trip to the US, and an aggressive promotion to Clinton, where the diminutive righty put together a great statistical year. People looked at his lack of pedigree, his stature, and so-so velocity and assumed his change-of-speed game would get crushed in the high minors – all this despite the fact that he continued to add velo along the way.
In 2011, his results were mixed enough that the debate just intensified. The walk rate was great, but high-minors hitters punished strikes much more, and he was getting hit hard. A trip to Tacoma didn’t solve those issues, but saw him touch 95 on the gun. Instead of the no-stuff, right-handed Nicaraguan Jamie Moyer, we had something both more common and more intriguing: a very young prospect who threw hard and had another great pitch, but who struggled when exposed to elite hitters in the PCL. In 2012, Ramirez adjusted. After a brilliant spring, he made the opening day roster as the long reliever, and showed flashes of promise admixed with a selection of mistake pitches. After some additional seasoning in Tacoma to stretch him out, Erasmo returned and looked like a sure-fire #3 starter: his control was back where it was in Clinton, and his change-up was one of the better swing-and-miss pitches in the league. On June 25th, in his 3rd big league start, Ramirez made a statement with a brilliant 10K, 1R start against Oakland. Because Mariners, he lost it 1-0 to his American, lefty doppelganger Tommy Milone, but he’d proven his game *could* work in the bigs. He followed it up with several solid starts, and while he picked up a nagging injury, he looked like a key part of the M’s rotation for years to come.
Mike Montgomery was a first round pick out of HS, and took to the pro game immediately. After laying waste to the low minors, the lefty with a big fastball and devastating change seemed like a #1 starter for the perpetually rebuilding Royals. He was their #1 prospect in 2010, and one of the top guys on their 2011 list – the one that was infamously and with horror-movie-like-foreshadowing called the best collection of propsects ever assembled. It’s easy to forget how ill-fated that collection seemed a few years ago, before Moustakas and Hosmer underproduced for a pennant winner and not just for another so-so 90-loss team. LHP Jake Lamb retired for a while, then came back. Danny Duffy kept getting hurt. Mike Moustakas has essentially been a poor-man’s Dustin Ackley. Jake Odorizzi seemingly hit a wall in AAA. And then there was Mike Montgomery.
Through 2010, Montgomery was a dominant force – combining low walk rates with solid K rates and stuff that seemed extremely difficult to square up. He paired a fastball at 92-95 with a change-up at 84 and a developing curve that could help him dominate lefties. He started the 2011 campaign in AAA Omaha, seemingly on the brink of a call-up. Instead, he tanked. His walk rate crept up over 3 per 9. His strikeouts dropped, while his hits allowed climbed. He looked like the same pitcher, but, depending on who saw him, he either lacked confidence, the right approach or a decent third pitch.
In what was increasingly looking like a make or break 2012, Montgomery regressed further, culminating in a demotion to AA, where his numbers cratered. That off-season, he was a change-of-scenery throw in to the Wil Myers for James Shields mega-deal. It seemed easy to see him rebounding with Tampa, but while Jake Odorizzi fashioned himself into a better-than-average starter, Montgomery was merely so-so for AAA Durham. He wasn’t getting blown off the mound the way he had been in 2012, but the K:BB ratio was as bad as it had ever been. Whatever the reason, Montgomery just wasn’t the same guy any more. He improved a bit last year with Durham, trimming his walk rate and showing some signs of an ability to induce weak contact, but he still didn’t look like a big league starter. The Rays injury woes leave them in desperate need of a starter, but they’d toyed with moving him to the pen and seeing if they could turn him into a lights-out reliever the way they had with another disappointing uber-prospect, Wade Davis (who, coincidentally, moved to Kansas City in the trade that brought Montgomery to Tampa). After Alex Cobb went down, though, the Rays need for someone to man the back of the rotation outweighed their desire to see what Montgomery could become.
The M’s simply couldn’t keep Erasmo Ramirez. We’ve all known it since camp kicked off. Out of options, but on a team with at least five better starters, Erasmo’s spring (and his stellar VWL season this winter) was an extended audition for someone else. It’s fitting, in a way, that he heads to Tampa in exchange for Montgomery. At this point, Ramirez seems like a smart play for a team looking for a bit of upside and the need to get something like replacement-level-or-better performance. The M’s have no need for that, so exchange Ramirez for a lottery ticket – a guy with tons of unrealized talent, who could potentially play a role in the bullpen once he fully commits to the role in Tacoma. Everyone’s stuff is better in short stints, but for certain players, the change in role unlocks something more. Wade Davis was a starter who couldn’t miss bats, until he became a dominant reliever. I’m not saying Montgomery will follow in Davis’ footsteps, but while the M’s picked up a pitcher who’s disappointed for four solid years, he’s a hell of a lot more talented than the average waiver claim.
Erasmo’s move to the AL East doesn’t look great on paper; home runs have hurt his results for years now, and he’s leaving the spacious, marine-layered parks of the West coast for Toronto, new Yankee Stadium and Camden Yards. And Mike Montgomery’s an odd-looking lefty specialist given his change-up – he’s actually got a better FIP against righties than lefties over the last four years. This isn’t a blockbuster, and we may never look back on this trade at all, let alone ruefully. But the M’s bargaining position here was compromised by the fact that everyone knew Ramirez had to go. If they held out for more in trade, teams could wait the M’s out and snatch Erasmo on waivers. For a variety of reasons, not the least because Erasmo already WAS a successful big leaguer for a few months, I think it’s much more likely that the Rays get some positive WAR out of this deal than the M’s. However, given the circumstances, that’s not the way to evaluate this trade. The M’s will – and should – accept a hell of a lot more variance, more uncertainty, in the hopes of unlocking Montgomery’s talent. Given that Erasmo was clearly behind Roenis Elias and perhaps even Jordan Pries in SP-depth-in-AAA, his superior ZiPS projection meant less to the M’s than any other team.
I’ll close by saying that I’ve always been an Erasmo-Optimist, and that the team’s tough love approach to his struggles produced shockingly little in the way of growth, adjustment or improvement. I always thought the M’s would fritter away a perfectly solid, cost-controlled starter, and I worried that they’d reach for one of Zduriencik’s all-tools, no-results guys like Mark Rogers last year. Now that they have, I’m OK with it. What’s changed isn’t my assessment of Ramirez (ok, actually, that’s a bit of it) or my blood-alcohol level. It’s the team Ramirez was a part of. The M’s got a lot better a lot quicker, and their starting pitching got much, much better than any of us would’ve expected. Just because you have depth doesn’t mean you give it away for pennies on the dollar, but the M’s aren’t in a position to wait around and see if Erasmo can stop giving up dingers anymore. That’s not a slam on Erasmo, really, it’s a testament to a rotation that gelled after Ramirez blew his last shot. I wish him well, and probably *still* overrate the odds he can pull off an Odorizzi-like improvement, but good on the M’s for getting something interesting for him.
Cactus League Game 27, Mariners at Padres
JA Happ vs. Andrew Cashner again, 1:05pm, Root TV
These two faced off five days ago, and while neither was sharp, Happ was concerningly bad. Earlier this spring, I said he needed to focus much more on his four-seamer and throw a lot fewer of the sinkers that righties have enjoyed since Happ learned the pitch in 2010. Well, he tried that last game and it didn’t go terribly well. It’s a truism that the same approach in Arizona may not produce the same results as it would in Seattle, but I’m sure it’d help Happ’s confidence to have a solid game today. Seth Smith was a late scratch today, and that’ll afford us the chance to get a look at Alex Jackson, the M’s #1 or #2 prospect, who gets the start in Smith’s place. Jackson even gets Smith’s line-up slot.
1: Ruggiano, CF
2: Jackson, RF
3: Cano, 2B
4: Seager, 3B
5: Weeks, DH
6: Morrison, 1B
7: Ackley, LF
8: Zunino, C
9: Ascanio, SS
SP: Happ
The M’s optioned lefty Roenis Elias to Tacoma today, formally ending the battle for the 5th spot in the rotation (a battle that seemed all but over a few weeks ago). By sending him to AAA, they can keep Elias stretched out; keeping him around as another situational lefty or swingman always seemed like a waste, and I’m glad the M’s agreed. Bob Dutton speculates that they could use Elias as a kind of sixth starter to manage innings for Taijuan Walker and James Paxton, both of whom had injury issues in 2014.
As many of you know, M’s pitching prospect Victor Sanchez died last night in Venezuela, roughly a month after he suffered a head injury while swimming in Carupano. The 20-year old suffered skull fractures and brain injuries, but early reports were – given the context here – somewhat positive. Perhaps it was his age, perhaps it was just optimism, but I somehow thought he’d survive this. My thoughts are with his family, including his new wife. The M’s statement mentioned that he was quiet, but popular amongst his teammates. It’s been amazing to see all of the comments from those teammates on Twitter. Sanchez famously threw a no-hitter for Clinton the day his mother arrived from Venezuela to see him pitch professionally for the first time. His size (6′ or a bit less, 260-280 lbs.) made him something of an anomaly, and many ranked hurlers with “projectable” frames above Sanchez, but he’d fought AA to a draw last year at the age of 19. But beyond the stats or age-relative-to-league, Sanchez seemed to make the organization a bit better simply by being a part of it. RIP.
Cactus League Game 22, Cubs at Mariners
Taijuan Walker vs. Travis Wood, 7:05pm Root TV
Travis Wood is a lefty with a rising fastball and an assortment of breaking pitches behind it – the cutter gets the most use, but he makes several stops along the cutter-to-curve route: a slider about 6mph slower than the cutter, and then a curve another 8mph slower than the slider. He’s got a change-up, as he’s faced right-handed line-ups frequently, and while it’s not a plus pitch, it’s not a disaster, either. As an extreme fly-ball guy, Wood’s been up and down depending on his control (last year’s poor results stem in large part from his nearly 10% walk rate) and how many of those fly balls leave the yard. When both of those factors align, he’s a solidly above-average pitcher; he put up nearly 3 fWAR in 200 innings in 2013, for example. The problem is that it’s apparently quite difficult for him to maintain that alignment. His HR/FB ratio crept up last year, and combined with poor control and a BABIP spike, and Wood became a replacement-level hurler by fielding-dependent metrics (his ERA started with a 5), and only so-so by FIP/xFIP. So is he an intriguing 4th starter with upside, or someone you simply can’t count on in the rotation if you want to compete for a divisional title?
Probably both, but despite so-so velocity, there’s enough raw stuff here that I wouldn’t be comfortable writing off his 2013 as a fluke. The extreme vertical rise on his fastball generates quite a few infield pop-ups, and that should – SHOULD – help him beat his FIP. It hasn’t for a few reasons. One is probably some bad luck, including the bad luck to play in a division that includes two very homer-friendly parks (MIL and CIN). The other has to do with his pitch mix. Like a lot of pitchers we’ve talked about recently, from JA Happ to Jered Weaver, Wood’s picked up a sinker to complement his four-seam fastball. Unfortunately, it’s atrocious, and he’s exacerbated that by throwing it mostly to opposite-handed hitters. Righties have feasted on the offering, batting .339 with a .535 SLG% over the course of 300+ at-bats. Meanwhile, they’re struggling against his four-seamer, which makes sense given all of that platoon-split-killing rise. He’s throwing a worse pitch more often.
When the Cubs acquired him from the Reds, he was coming off a sub-par season, and he placed some of the blame for that on an over-reliance on his cutter. The Cubs evidently didn’t agree with that assessment, as Wood has gone from throwing it a bit less than 1/5th of his pitches to over 1/3 in 2012 and 2013. Wood’s results haven’t been outstanding with it, though it’s a pitch he uses more frequently when behind in the count, so that’s to be expected. His results aren’t altogether bad with it, and he’s never quite got the hang of his slider. With his solid four-seamer and passable cutter and change-up, it might help to simplify things a bit. Some guys can throw five pitches for strikes, and some guys put up 10% walk rates.
Meanwhile, Taijuan Walker’s march to the opening day rotation continues. Today, he’ll miss uber-prospect Kris Bryant and get to face Mike Olt instead. He *will* face six lefty-hitting Cubs to start with, though.
1: Weeks, DH
2: Jackson, CF
3: Cano, 2B
4: Cruz, RF
5: Seager, 3B
6: Ruggiano, LF
7: Morrison, 1B
8: Zunino, C
9: Miller, SS
SP: Walker
The story of the day may be word about an investigation into gambling by former Astro current Marlin righty Jarred Cosart. Cosart DM’ed a sports handicapper on twitter, and then the handicapper posted (edited) screenshots of the conversation. Twitter, ladies and gentlemen. I should say that the screenshots at the center of this do not indicate or imply that he bet on baseball; as a product of Texas high schooling, it doesn’t seem like a big leap to assume he was betting on football, for example. But then, a pro athlete tweeting at an internet betting tipster does not earn one a whole lot of benefit-of-the-doubt points when it comes to doing something as stupid and career-threatening as betting on baseball while being employed by baseball.
Cactus League Game 21, Mariners at Padres
JA Happ vs. Andrew Cashner, 1:05pm
We’ve now settled into that period of the Cactus League where the novelty that people are actually playing baseball has worn off, and the routine has set in. JA Happ has made the rotation, and has been fairly solid thus far. We’ve already seen Andrew Cashner as well; if you like, you can brush up on the fireballing pitch-to-contact enigma here. So if we can’t talk about Happ/Cashner, what can we talk about? I’d *like* to talk about the improbability of yesteday’s opponent, Matt Shoemaker, but we’ll have plenty of opportunities to go into the black arts that led a PCL journeyman, sub-AAAA player to post a well-above average season last year and shut down the M’s again on Monday.
Instead, let’s focus on essentially the only remaining question involving the Mariners opening day line-up – who’s in the Bullpen? I know McClendon has said that he’s not really looking at the numbers when deciding between Roenis Elias and Taijuan Walker for the fifth starter job, but there’s essentially no argument here that I can see – not even Kris Bryant-style service time shenanigans. Keeping Elias as a swingman, sort of like the A’s are considering with Barry Zito, could work, but it would deprive him of regular work and would also make it much harder for the M’s to keep LHP David Rollins. Rollins was the M’s Rule 5 pick from Houston, and has opened some eyes this spring, as we’ve said. Rollins’ velocity has been better than Elias, Tyler Olson, and some of the righties like Danny Farquhar – it’s essentially matched Wilhelmsen and Leone’s this spring. He’s walked no one in 8 IP, and while he’s a bit less of a pure lefty specialist than Olson, Furbush or even Elias could be, he’s got to be kept on the M’s 25 man roster or be returned to the Astros organization.
Losing a Rule 5 pick is not the end of the world. Losing Jose Flores in the spring or Kanekoa Teixeira didn’t much matter to the M’s. Still, Elias would seem to have more value to the M’s as starting pitching depth than as a long man out of the pen. And while Tyler Olson’s impressed in the spring, it wouldn’t hurt to have him start in Tacoma or make the move to the bullpen once and for all in the PCL and see how he handles the new role. With Rollins, the team doesn’t really have options. And while the M’s try and figure out how many lefties is enough, they need to take some steps to bolster a bullpen that was brilliant in 2014 but may be in line for some regression. Farquhar’s velocity is down substantially this spring, and while he’s typically been a slow starter (he averaged 92-93 in mid-March games last year), his four-seam velocity is down 3-4mph from 2013, and his fastball and cutter are down 2-3mph from last season. I’m not suggesting that Rollins is in line for a bullpen spot ahead of the M’s 2014 set-up man, but I am suggesting that the M’s need to keep as much talent within the organization as possible. They need to be able to shuffle people in and out as needed, and they need contributions from guys who won’t be on the opening day roster. It’s in that context that Rollins’ roster situation becomes more of a thumb on the scale, even just to see what they have for a month or two (as the M’s did with Teixeira).
1: Jackson, CF
2: Smith, RF
3: Cano, 2B
4: Cruz, DH
5: Seager, 3B
6: Morrison, 1B
7: Zunino, C
8: Ackley, LF
9: Miller, SS
SP: Happ
Cactus League Game 19, Rangers at Mariners
James Paxton vs. Yovani Gallardo, 1:05
I felt bad about missing yesterday’s Felix day until I saw the box score. Hopefully Paxton has an easier time of things today against the Rangers. He’ll face righty Yovani Gallardo who came over from Milwaukee in a trade for prospects. In one sense, the move was a curious one for Texas; they’re in a rebuild despite the presence of well compensated vets like Shin-Soo Choo (struggling with a forearm injury) and Prince Fielder (recovering from season-ending surgery last year). In another, he’s a desperately needed mid-rotation bridge between Yu Darvish at the top and a continually churning sea of AAA arms the Rangers will sort through in 2015.
Gallardo was supposed to be the ace of the Brewers rotation for years, but freak injuries and an inability to move from “promising” to “star” allowed the Rangers to acquire him for middling prospects. Gallardo’s strikeout rate has declined every year since peaking in 2009, and he’s now a below-average K guy…and that was in the NL. He throws a rising four-seamer at about 92, and also throws a sinker, a curveball and slider. He had a change-up that he’d use to lefties occasionally, but he essentially shelved that last year.
Gallardo’s straight-over-the-top delivery is noteworthy, and as many have noted, seems to be something that the Brewers consciously select for and/or teach. One of the presumed benefits of the over-the-the-top delivery is minimal platoon splits, and while Gallardo’s have bounced around, his career numbers bear this out – he’s got splits, but they’re on the low end of normal. That’ll be important as he moves from the division that plays match-ups the *least* (the NL Central) to the one that figures to do so the most. Gallardo has had the platoon advantage for the majority of his batters faced in each year since 2010. Yu Darvish faced lefties over 63% of the time last year. It’s going to be an adjustment for Gallardo.
The rising fastball should produce more fly balls (hopefully lowering BABIP), and Gallardo (and the other Brewers hurlers) have had problems with homers off and on. Oddly, though, Gallardo’s GB% has risen steadily as his K% dropped, and he’s now an above average grounder guy. Developing the sinker certainly helped (he was strictly a four-seam guy early on), but it’s also a testament to the fact that batters tend to put his breaking balls in play on the ground. At one point, they were swing-and-miss pitches, but that day’s long gone; when he’s on, Gallardo can minimize damage by getting batters to top curve balls to the infielders.
Line-up:
1: Jackson, CF
2: Smith, RF
3: Cano, 2B
4: Cruz, DH
5: Seager, 3B
6: Morrison, 1B
7: Ackley, LF
8: Miller, SS
9: Sucre, C
SP: Paxton
Six lefties. Welcome to the AL West, Yovani.
Cactus League Game 17, Mariners at Rangers
Roenis Elias vs. Derek Holland, 1:05pm, 710am Radio, no TV
Welp, Roenis, I suppose you could make the case that the league needs to get back to having a good old fashioned long reliever, and that all this specialization is less effective in the AL West, where every team has the depth to mix and match pinch hitters to LOOGYs and ROOGYs. I don’t think it’ll work, but hey, Tacoma’s not so bad. Hit me up for bar and restaurant recommendations, or just ask Curto. Last night, Taijuan Walker seemingly wrapped up the #5 starter job with another scoreless outing, and he did so by working through some very early command issues. He’s now gone 12 innings with 13 Ks and just 3 BBs, and has yet to give up a run. Spring stats themselves don’t matter, but Walker has earned these stats not just through a high-octane fastball, but because his split-change is much improved from last season. There was so much talk about his cutter/slider transition, but the mechanical changes he’s made seem to have given him much better control. If he can throw the split for strikes as he did last night, then he’s less liable to have serious platoon split issues, and that was a bit of a concern if he was more of a FB/SL/CU pitcher (a BIT of a concern, not a huge one). Some times you just have to tip your cap, Roenis.
Derek Holland’s appearance today is good news/bad news for the Rangers. Holland came down with the always-good-to-hear shoulder soreness as camp opened, so the fact that he’s getting some game action in is a good sign that it wasn’t so severe that he’d need a DL stint or worse to correct it. On the other hand, it’s March 20th and Holland is pitching his first game, and is a couple weeks off of recuperation from *shoulder soreness*. It’s doubtful he can make the opening day start at this time, and while he’ll be ready in early April, the Rangers rotation which looked good at the top (Darvish, Holland, Gallardo) now looks extremely shaky. Holland’s been plagued by injuries throughout his career, and the Rangers desperately need someone behind Gallardo if they’re going to put up a respectably bad season instead of one of those 2013-Astros-style disasters.
Holland seemed to put everything together in 2013, topping 200 IP for the first time, and developing his change-up to the point where it wasn’t just an afterthought. As a FB/SL/CU pitcher, he had some platoon split issues earlier, and given that he can expect to face righties about 3/4 of the time, that limited his upside. The fact that he’s a fly ball pitcher in Arlington didn’t help either, but as a lefty with a fastball at 94-95 and a good slider that he can throw to righties and lefties, he always seemed like a breakout candidate. The change-up helped him control righties and the home run, at least in 2013. Unfortunately for the Rangers, he missed nearly all of 2014 with injuries, along with basically every other good Rangers player. It’ll be interesting to see if he’s still working on the change, or if he tries to keep things simple today and throw fastballs and sliders. This is also a good game for the M’s righties to show what they can do against a lefty with above-average stuff. Go, Justin Ruggiano.
1: Jackson, CF
2: Bloomquist, 1B
3: Ruggiano, LF
4: Cruz, RF
5: Zunino, DH
6: Romero, 3B
7: Tyler Smith, SS
8: Sucre, C
9: O’Malley, 2B
SP: Elias
I checked this like three times, but it’s the RANGERS who are playing split squad games today, not the M’s. Hmmm. Well, good for Smith, O’Malley and Bloomquist. It’s no Tinker-to-Evers-to-Chance, but it’ll do for today.
Cactus League Game 16, Indians at Mariners
Taijuan Walker vs. Carlos Carrasco, 7:05pm, Root TV, 710am Radio
The M’s play their first night game of the spring, and for those of you who can tear yourself away from March Madness, the game’s on TV. C’mon, Kentucky against the 16-seed can’t be more entertaining than the match-up we’ve got here – the M’s Taijuan Walker against the Tribe’s Carlos Carrasco. No one in baseball was better down the stretch in 2014 than Carrasco, whose final 10 starts were Kershaw-esque. Carrasco’s teammate, Corey Kluber, produced the most fWAR in the second half of anyone in the majors, and his 4.1 was higher than anyone since Randy Johnson in 2004, but once Carrasco moved in from the pen, he was essentially matching Kluber start for start. Carrasco had always been a high GB%, high velocity guy, but without the secondary stuff to keep in the rotation full time.
This wasn’t just a BABIP fluke – Carrasco’s K rate was up over 10 percentage points from where it was in 2011, and doubled from his mediocre bullpen season in 2013. His walk rate fell. His velocity was up significantly, and suddenly, no one could touch his slider or his change-up. In 171+ innings in 2011 and 2013, Carrasco gave up 8 HRs on his slider/change. Lefties feasted on his sub-par cambio, slugging .490 off of it. Righties struggled overall against his slider, but a FB/SL guy without a true weapon is going to be vulnerable to opposite-handed hitters, and that was Carrasco’s problem in a nutshell. It’s why he kept getting sent to the pen, and it’s why his stints in the rotation weren’t much to write home about. Suddenly, in 2014, his change-up was a devastating pitch – batters knocked one XBH, a double, on it. His whiff rate jumped, and when anyone did hit it, they beat it into the ground over 70% of the time. It’s important to note that it moves in essentially the exact same fashion – this isn’t a new or different pitch, and he uses it broadly the same way. He’s better at keeping the ball down, which is another example of his improved command (his walk rate is another). It’s all so simple, that it starts to seem implausible.
Carrasco’s stretch run brings to mind the 2012 finish from the Braves’ Kris Medlen. Medlen had been a perfectly serviceable swing man for a couple of seasons before closing 2012 by going 9-0 in 83 IP, riding an unhittable splitter to stardom. He was solid in 2013 before succumbing to TJ surgery, though he couldnt’ quite recapture the form he had in late 2012. The Indians know well that stretch runs aren’t always predictive, even when the core metrics rule out flukes. In 2013, the pitcher who racked up the most 2nd-half WAR was the Tribe’s Ubaldo Jimenez. A little ways back was his teammate, Scott Kazmir. The Indians let both leave in free agency, and while Kazmir impressed with Oakland, he wasn’t quite the dominant force he had been late in 2013. Jimenez was a mess for the Orioles last year, as his control left him again and he struggled to keep his ERA/FIP under 5. This isn’t to say that it’s *always* a fluke – Clayton Kershaw shifted into overdrive at the end of 2012 and he hasn’t stopped since. Still, Carrasco’s amazing results and extremely short track record of achieving them is a key reason no one really knows what to expect from the Indians rotation, and thus from the Indians.
Line-up:
1: Jackson, CF
2: Smith, RF
3: Cano, 2B
4: Cruz, DH
5: Seager, 3B
6: Morrison, 1B
7: Ackley, LF
8: Miller, SS
9: Sucre, C
SP: Walker
Substitute Zunino for Miller and that’s essentially your opening day line-up. This is pretty much how the M’s are going to look against righties.
Was going to wait until the M’s faced the A’s again, but this piece by Jason Wojciechowski about Oakland’s crazy offseason is worth your time. Every time people think they’ve figured out the A’s strategy (OBP! Fly ball hitters! Short pitchers!), they’ve moved on.
Cactus League Game 15, Athletics at Mariners
Erasmo Ramirez vs. Drew Pomeranz, 1:05pm
Drew Pomeranz was once the 5th pick in the draft, a consensus top prospect, and the centerpiece of a trade for an ace (Ubaldo Jimenez). It was one of those trades that everyone thinks is going to be gigantic, a trade that can transform the fortunes of both clubs – sort of like the Pineda/Montero swap the M’s and Yankees made. And, like that trade, I think it ended up looking a lot more important than it actually was. Jimenez struggled down the stretch in 2011 and spent 2012 as a replacement-level pitcher. He bounced back in 2013, sort of like Pineda in 2014, but that just earned him a contract somewhere else. Meanwhile, Pomeranz struggled to find a third pitch behind his fastball and curve. The two pitches had been enough in the minors, but his attempts at developing a change-up hadn’t produced another MLB-level pitch. Ultimately, the Rockies sent him out – as a complementary piece this time – for Brett Anderson.
The A’s haven’t figured out how to turn Pomeranz’s change into a good pitch either. In fact, the best thing they’ve done with it is convince him to stop throwing it. Instead, they built confidence in his sinker, a pitch he threw sparingly in Colorado, but which became a key part of his arsenal last season. While it actually didn’t improve his GB% overall, it’s given him a GB pitch to throw to righties. Lefties get a steady diet of four-seamers, while righties get the sinker. This means that he’s got sharply different batted ball results depending on the handedness of the hitter, but in a park like Oakland, fly balls don’t matter. Filling in for injured starters, Pomeranz put together a surprisingly effective 2014, throwing 69 innings of solid, league-average-to-better baseball. Sure, an 80%+ strand rate made the ERA even better than that, but Pomeranz improved his K% while lowering his walk rate all while facing lefties the majority of the time. The A’s constant rotation churn means he still doesn’t have a spot lined up, but he’s a key part of the A’s enviable pitching depth.
1: Weeks, DH
2: Jackson, CF
3: Cano, 2B
4: Cruz, RF
5: Seager, 3B
6: Ruggiano, LF
7: Morrison, 1B
8: Zunino, C
9: Miller, SS
SP: Erasmooooo
Weeks starts at DH so he can get to a dentist’s appointment (he’ll get 3 ABs quicker this way).
Today’s the day the M’s 2015 commercials come out. They should be up at Mariners.com or twitter – I’d link ’em, but I’m writing this before release. Good? Bad? Better than last years?
Good article from Bob Dutton on David Rollins, the hard-throwing Rule 5 LOOGY candidate. Tyler Olson’s spring has opened a lot of eyes, but Rollins’ pure stuff (including a FB that sits 94-94) looks impressive. Olson’s frisbee slider looks perfect for the LOOGY role, while Rollins admits his slider hasn’t been “on” this spring. That said, Rollins rising FB can be effective even when facing a righty pinch hitter. The second lefty in the pen is not going to decide the AL West, and the M’s look like they have several solid candidates. But if I had to guess, I’d say it’s Rollins job to lose.
On cFIP
I’ll start out with a warning. This is a post about posts about math. If that’s not your cup of tea, that’s great – you sound well adjusted. If trying to figure out what new pitching metrics are trying to tell us in baseball terms, read on.
On Wednesday, Jonathan Judge released a new public pitching measure, Context-FIP or cFIP for short. There’ve been many run estimators over the years, including the mostly descriptive ERA, RA/9 and FIP, and a series that regress past results in order to capture a pitcher’s elusive “True talent” and thus help predict his future results. For a number of reasons, ERA is exceptionally poor as a predictive measure. It conflates what a pitcher does with the contributions of his team, it ignores parks and opponents, and its attempt to strip errors out causes some odd effects that make it even less predictive than plain old RA/9. More predictive metrics like xFIP strip out defense, and then heavily regress actual HRs allowed. It’s generally pretty close to FIP, but it’s generally more stable from year to year. Importantly, the fact that it’s more stable than FIP is either a feature or a bug, depending on what you’re using it for. What Judge’s statistic attempts to do is bridge the gap between predictive models and descriptive ones. It’s not necessarily the best at predicting future runs allowed, but given the noise involved in *that*, Judge argues that we need to evaluate not only how well it predicts runs out of sample, but how well it predicts *itself* in future years.
One of the tantalizing aspects of cFIP is Judge’s use of “mixed models” to calculate cFIP. Instead of ignoring everything from batter handedness to ballpark to umpire, the model incorporates them, while keeping them segregated from the fixed effects (things with only a set number of possibilities; no matter how many observations you make, pitchers will throw with either their left or right hands). The model can then examine the “random effects” to see how they effect runs, adding certainty as you add more and more observations/data. So while FIP treats all HRs the same, and xFIP strips out all actual HRs, cFIP is an early example of a cool hybrid. A HR to Troy Tulowtizski in Colorado is *different* than a HR to Brendan Ryan in Safeco, and it’d be cool to incorporate that information into a FIP-like statistic. Judge is a great writer, and the explanation of the approach is surprisingly readable – he outlined mixed models in this great post on catcher framing, and his description of their application to cFIP is surprisingly lucid to a non-gory math person like me.
In the spirit of embracing context instead of ignoring it, Judge’s tests of various metrics isn’t how well it predicts future runs, but rather RE24. This is a win-expectancy-based stat, that calculates changes in run expectancy (runners on base and outs) as well as actual runs scored. A 2-out bases loaded hit is more damaging to a pitcher’s RE24 than it is to his FIP, which is uninterested in the base-out state, and uninterested in the hit itself. That’s an interesting change, though it’s something to keep in mind when you look at Judge’s table of correlations – he compares in-season correlations to RE24/PA of a bunch of run estimators, like FIP, xFIP, SIERA, ERA, RA, etc. Correlating to runs is hard enough, but adding context makes it even tougher – cFIP isn’t particularly good at the task, though it squeaks past SIERA, xFIP and the like. That said, a contextual measure being better correlated with another contextual measure than those that explicitly and intentionally *ignore* context isn’t all that impressive. cFIP moves to last place when he runs a weighted correlation on all pitchers, not just those who meet a batters-faced floor. Unsurprisingly, the ones that do well in this test are the *most* context-dependent, namely RA – a measure that lugs around a pitcher’s defense with him.
That said, cFIP shines when predicting RE24 in the following year. Shines may not be a particularly good term here; its three year average correlation is under .4, leaving it all but tied with SIERA, xFIP and kwERA (a Tom Tango metric that uses strikeouts and walks *only*). It’s really, really hard to predict a measure with as much noise as RE24, given that it RE24 is so dependent on sequencing. That’s not a knock on cFIP (or kwERA), but it’s worth noting that we’re talking about very small differences in what is ultimately not-so-hot predictive power. cFIP also ranks #1 in how well the measure correlates with itself from year to year, again finishing ahead of kwERA and SIERA.
So…is it, you know, good? It’s promising, but I’m not sure that *this version* gets us all that far. As we’ve seen, it’s quite close to kwERA, an extremely simple measure that does a bit better as a descriptor if a tiny bit worse as a predictor (of RE24, mind you). That it’s stable *could* be an indication that it’s homed in on true talent, but it could be an artifact of all the regression the model is doing. A measure will reduce big errors by squishing everyone towards the mean, but that can obscure or underestimate the gaps between great pitchers and their not-so-great colleagues. Adding more regression may give you more stability, but it does so by ignoring what actually happened; it may not be getting you closer to true talent, it may just be minimizing the importance of true talent. A number of much smarter-than-I analysts have already pointed these concerns out – here’s Neil Weinberg on the former, and Peter Jensen on the latter.
For me, I find it somewhat odd that it’s *so close* to measures like xFIP and kwERA that ignore HRs entirely. It’s utilizing actual results, but I’m not seeing much of an effect from that. Let’s turn to a Mariner-centric example. James Paxton had an injury-shortened 2014, but he was effective due to a high GB% (and thus very few HRs allowed) and due to batting average on balls in play. ERA loved him, FIP loved the few-HRs thing, but wasn’t blown away by his K:BB ratio, while xFIP liked the GB%, but thought he got a bit lucky. Taking all of that into account and regressing the real results, what does cFIP think? Hmm, a touch *below* league average. Not even xFIP was willing to go that far. One thing that it could indicate is that it’s putting a lot of weight on parks and the specific batters Paxton faced – maybe he didn’t give up a lot of HRs because he faced a disproportionate share of Eric Sogards and few Mike Trouts. Baseball Prospectus tracks the quality of opposing hitters AND a pitcher-specific park factor, so that can help us explain this. Paxton did indeed benefit from a great run environment thanks to pitching in Seattle and Anaheim, but the quality of opponent metric is extreme – in the opposite direction. Paxton faced an extremely difficult slate of hitters; facing Anaheim four times, and then adding in Baltimore, Toronto and Oakland will do that. No one except Carson Smith faced an average hitter with so high an OPS, and Paxton was in the top few starters by this metric (relievers typically face tougher hitters for obvious reasons).
Another example is ex-Angels reliever Ernesto Frieri. Judge notes that Frieri is the player with the largest gap between his cFIP and FIP-, both of which are park-adjusted. Last year, Frieri had a lovely K:BB ratio, and well over a strikeout per inning. However, a barrage of HRs and an abysmal strand rate got him shipped out of town. Frieri ended the year with an ERA well over 7, and a FIP of around 5.5 – giving up 11 HRs in just over 40 innings will do that to a guy. His xFIP was better, but with such an extreme fly ball ratio, it’s still not great (it’s lower than Paxton’s, for example). cFIP sees past all that, giving him a 90, or 10% better than league average. Paxton’s slightly worse than average, Frieri – who faced a slightly *worse* set of hitters, and also enjoyed HR-suppressing park environments – was better. Batter handedness? Nah, Paxton faced *five times* as many righties as he did lefties. This is a very anecdotal way to analyze a statistic, but whatever the model is doing with actual HRs allowed an actual hitters, it can’t be much. If some set of circumstances completely outweigh the actual results, that’s fine, but then the complexity in adding in all of those actual results to the model doesn’t seem to have been worth it.
The model’s promise is the ability to bridge the gap between descriptive and predictive, but it’s not immediately clear what all of the “actual results” are doing. Maybe the model regresses them away, as they don’t have the stability of good old strikeouts and walks. That’s fine, that’s interesting, but if so, it doesn’t seem to offer a lot beyond kwERA/kwFIP. Instead of building a bridge between the two classes of metrics, it certainly *looks* like cFIP is setting up camp with the predictive models. It appears to be more stable, but again, if it’s more stable solely because the spread is much lower than it is for FIP, xFIP, etc. (to say nothing of ERA), then that limits cFIP’s utility. What would be interesting is to show the correlation between cFIP and kwERA, or cFIP and SIERA. My guess is that they’re going to be very, very high.
At this point, we’ve seen two innovative approaches to integrating actual results to predictive models, SIERA and now cFIP. Just as an outside observer, those actual results seem to get regressed away pretty quickly. Both seem, on paper, to take some pretty important things into account – velocity for SIERA and umpire for cFIP. And despite that, or rather *because* of that, they end up looking like a souped up xFIP. ERA is clearly and increasingly widely seen as inadequate, but every new pitching metric seems to train its guns on FIP. If you’re looking to better describe *actual* results, RA/9’s place in the pantheon isn’t imperiled by cFIP. To the degree that we learn something new about the game of baseball, and every new metric should attempt to illuminate some aspect of the game, what we learn (or re-learn) is the central insight into DIPS – that strikeouts and walks matter so, so much more hits. We’ve added tons of data to FIP, or rather xFIP, and we’ve moved the needle, but by frustratingly little. That’s interesting in itself, if frustrating. At this point, it seems like we’re not going to get a noticeably more predictive/descriptive model by adding a bit more data. Multiple smart people have added tons, and the gains are marginal. If we’re going to break actual new ground, it seems like we might need to add tons more data. Don’t just incorporate umpire or velo, but incorporate pitch type, location, of every pitch, and what pitches precede and follow each pitch. These models are already frightfully complicated, and I hope/fear they’re going to get exponentially more complicated.
Ultimately, I think the mixed model approach has so much potential, and my skepticism (or confusion!) about cFIP isn’t based on a low ranking of Paxton, but on the fact that I can’t immediately see how the model uses actual results, especially HRs. FIP is *so* HR-dependent, and that leads it to underestimate guys like Hisashi Iwakuma. Other measures drop them entirely. We need something in between, but it may be that there’s simply too much variability in them to do this effectively or neutrally. As Neil Weinberg says, the star of the show may be kwERA – that knowing a pitcher’s Ks and BBs can give you as much information as you’re likely to get about future runs allowed as metrics that are light years more complex. Still, cFIP is something to watch. I’m excited to see what Judge does with it, and how analysts might utilize it – I’m even more excited to see what Judge does next.