2015 Positional Power Rankings: Starting Rotations (#1-15)
What do we have here? For an explanation of this series, please read this introductory post. As noted in that introduction, the data below is a hybrid projection of the ZIPS and Steamer systems, with playing time determined through depth charts created by our team of authors. The rankings are based on aggregate projected WAR for each team at a given position.
Yes, we know WAR is imperfect and there is more to player value than is wrapped up in that single projection, but for the purposes of talking about a team’s strengths and weaknesses, it is a useful tool. Also, the author writing this post did not move your team down ten spots in order to make you angry. We don’t hate your team. I promise.
As you’ve no doubt read in our other wonderful Positional Power Rankings, what is important here isn’t so much the specific ranking for each team — the slightest of bumps in playing time distribution would put the #7 Cubs on par with the #3 Cardinals, or the #15 Marlins on par with the Tigers and Giants, who are tied at #10. (Not to mention that the ghastly spectre of Tommy John always always looms, ready to sweep the leg out from under any number of teams, as has already happened to the Texas Rangers and Toronto Blue Jays.) What’s most important here is the tiers that teams have separated themselves into, and what’s most obvious is that the Washington Nationals and Los Angeles Dodgers have rotations that match the rest of their rosters: in a class of their own, primed to claw their way to a World Series championship.
| Name | IP | K/9 | BB/9 | HR/9 | BABIP | LOB% | ERA | FIP | WAR |
|---|---|---|---|---|---|---|---|---|---|
| Max Scherzer | 211.0 | 10.4 | 2.4 | 0.7 | .301 | 76.8 % | 2.80 | 2.78 | 4.9 |
| Stephen Strasburg | 191.0 | 10.0 | 2.2 | 0.8 | .307 | 75.8 % | 3.02 | 2.96 | 3.9 |
| Jordan Zimmermann | 191.0 | 7.5 | 1.6 | 0.8 | .301 | 74.1 % | 3.19 | 3.28 | 3.5 |
| Doug Fister | 192.0 | 6.3 | 1.7 | 0.8 | .302 | 72.8 % | 3.47 | 3.64 | 2.4 |
| Gio Gonzalez | 162.0 | 8.9 | 3.4 | 0.7 | .301 | 73.7 % | 3.45 | 3.44 | 2.4 |
| Tanner Roark | 40.0 | 6.5 | 1.9 | 0.9 | .295 | 73.1 % | 3.47 | 3.67 | 0.5 |
| Blake Treinen | 9.0 | 6.5 | 2.2 | 0.7 | .308 | 71.4 % | 3.64 | 3.62 | 0.1 |
| Taylor Jordan | 9.0 | 6.2 | 2.2 | 0.8 | .307 | 71.0 % | 3.84 | 3.83 | 0.1 |
| Total | 1005.0 | 8.5 | 2.2 | 0.8 | .302 | 74.5 % | 3.19 | 3.23 | 17.9 |
Contrary to initial expectations back in mid-January, the Nationals’ signing of Scherzer does not appear to be a move to set up a move. The Major League team who received the most WAR from their starting rotation in 2014 really did go out and invest in the consensus marquee free agent starter of the 2014-15 offseason. The gauntlet has been thrown down.
Never mind, for now, that the Nationals will be paying Scherzer eight-figure sums until Julio Urias is 33 years old. This ranking is all about 2015, and in 2015 the Nationals have the basically unprecedented luxury of bumping a starter with the 30th-best qualifying FIP of the previous season — that would be Roark, pitching to virtually the same effectiveness as Oakland A’s ace Sonny Gray during 2014 — to their bullpen.
Their fifth starter — let’s say it’s Fister — posted a career-worst 3.93 FIP, but at the same time used a top-five walk rate and a top-five line-drives-allowed rate to assemble a 2.41 ERA, which missed the top five by a margin small enough to be a rounding error.
Oh, is Gonzales actually their fifth starter? Since his first full season in 2010, Gonzales has posted over 3 WAR every year, indicative of both consistency and good health. The only other pitchers to accomplish this feat over the same timeframe are: Clayton Kershaw, Felix Hernandez, David Price, Cole Hamels, Zack Greinke, and Anibal Sanchez. Their fifth. Starter.
The competition for fifth-starter-ship is supposed to be a toss-up between low-ceiling prospects and/or veteran retreads whose names fans learn halfway through Spring Training. It’s not supposed to be a fiery battle between Cy Young-vote-getters. This is the bold, brave reality that the Nationals have set up for themselves.
The last time a rotation was this bogglingly stacked was the 2011 Phillies, a starting staff that inspired tomes on its way to a 102-win season and the team’s best winning percentage in 132 years of existence. And the 2011 Phillies walked into the season with Joe Blanton as their fifth starter, and when Blanton managed only eight starts due to injury, they turned to rookie Vance Worley. While the 2015 Nationals maybe don’t have the eye-bulging ceiling of the 2011 Phillies — Roy Halladay contributed 8.4 WAR for that team, Cliff Lee had 6.8; Scherzer is projected for 4.9 — they do have those Phillies beat in their ability to trot out an ace-caliber pitcher every single game. The snowball effect of having elite pitching every single game — a phenomenally rested bullpen; a playoff bullpen full of starters allowed to unleash their most vicious Wade Davis impression — might result in a team season like nothing we’ve ever seen before.
How many of the Nationals’ five starters are top-25 in cumulative starter WAR over the last five full seasons? All of them.
The cost is high, and free agency looms for a number of these fine arms as soon as seven months from now. But flags do fly for forever.
#2 Dodgers
| Name | IP | K/9 | BB/9 | HR/9 | BABIP | LOB% | ERA | FIP | WAR |
|---|---|---|---|---|---|---|---|---|---|
| Clayton Kershaw | 221.0 | 10.2 | 1.9 | 0.6 | .293 | 79.5 % | 2.24 | 2.43 | 6.4 |
| Zack Greinke | 202.0 | 8.4 | 2.0 | 0.8 | .299 | 75.6 % | 3.04 | 3.16 | 3.8 |
| Brandon McCarthy | 179.0 | 7.6 | 1.6 | 0.9 | .313 | 71.8 % | 3.62 | 3.42 | 2.5 |
Hyun-Jin Ryu  |
147.0 | 7.9 | 2.1 | 0.8 | .303 | 73.7 % | 3.27 | 3.26 | 2.4 |
| Brett Anderson | 72.0 | 7.4 | 2.8 | 0.7 | .305 | 71.4 % | 3.58 | 3.48 | 1.0 |
| Joe Wieland | 74.0 | 7.6 | 2.5 | 1.2 | .300 | 70.9 % | 4.08 | 4.02 | 0.5 |
| Brandon Beachy |
47.0 | 7.0 | 3.3 | 1.2 | .290 | 71.7 % | 4.23 | 4.44 | 0.1 |
| Zach Lee | 37.0 | 6.4 | 2.9 | 1.1 | .296 | 69.9 % | 4.41 | 4.46 | 0.1 |
| Carlos Frias | 18.0 | 6.0 | 2.8 | 1.0 | .303 | 69.4 % | 4.35 | 4.26 | 0.1 |
| Total | 998.0 | 8.3 | 2.1 | 0.8 | .301 | 74.1 % | 3.25 | 3.27 | 16.9 |
As phenomenally impressive as the Nationals’ rotation is, it also wouldn’t be too surprising to see the Dodgers finish #1 in rotation WAR at season’s end. It also wouldn’t be too surprising to see the Dodgers finish somewhere around league average in rotation WAR. The Dodgers’ new braintrust of Andrew Freidman and Farhan Zaidi has set their rotation up for a highly volatile season — but by design, it would seem, which will go a long way in making the Dodgers’ season compelling viewing, no matter the result. If there is a single key to the Dodgers’ eventual success or demise, it will likely be the health — or lack of it — on this staff.
I don’t think there’s any single move that acts as a microcosm of the current state of the Dodgers other than their free agent signing of Brett Anderson. On a per-inning basis, Anderson has been a phenomenally productive pitcher for his entire career: conceivably just entering his prime years at age 27, Anderson has already averaged 3.7 WAR for every 200 innings he’s pitched, a pretty stellar rate. (Although, as Carson has pointed out previously, Anderson’s drop in fastball velocity has actually seen his per-inning productivity drop over time.) The problem is that Anderson has only pitched 206.1 innings total since the beginning of the 2011 season, an average of about 50 innings, or about eight starts a year.
Given this uncomfortably extensive injury history, there was not a large market for Anderson. The Colorado Rockies, his employer last season, turned down a $12M option for 2015, thus releasing Anderson into free agency. There, the Dodgers signed him for the hefty sum of $10M for one season, with a $4M performance bonus if Anderson pitches 200 major league innings this season — an accomplishment he has never accomplished before. If recent trends hold, Anderson will pitch 40-50 innings over 5-8 starts, producing nearly 1 WAR along the way. That’s not a bargain for $10M.
It’s worth remembering that Zaidi was in the Oakland A’s front office when Anderson was with the team, meaning Zaidi has as much first-hand experience as anybody at having to fill Anderson’s vacated spot in the rotation. (I.e., Travis Blackley being called into action for 15 starts for the 2012 A’s.) Zaidi did not have to make another gamble on Anderson’s health, and he probably didn’t have to do it for $10M. But for these mightily bankrolled Dodgers, it’s worth the roll of the dice if this is the year that Anderson can make 30 starts.
The injury concerns do not end there. Hyun-Jin Ryu managed just 26 starts last year, and is not projected by the fans (172 innings, 28 starts) nor Steamer (143/24) nor ZiPS (169/27) to make it through the year wholly intact. Since 2011, the first time McCarthy spent the full season in the majors, he has averaged about 150 innings each year. Beachy has not pitched a competitive game since August of 2013.
There are enough innings being missed here that it looks like it will take six or seven men receiving significant starts to get the team through the season. Wieland — of 39 Major League innings and one Tommy John surgery — could be called into meaningful duty for this team with visions of World Series rings, a possibility that hardly looks enticing for the Dodgers considering the middling 40 Future Value that Kiley prescribed to Wieland.
So that’s one outcome. Another outcome is that the Dodgers have acquired all of these pitchers because they believe they have found, as an organization, a way to reliably prevent injuries. In which case, Anderson or McCarthy serves as this team’s fifth starter — enviable depth that would certainly challenge the Nationals at the top of the league.
#3 Cardinals
| Name | IP | K/9 | BB/9 | HR/9 | BABIP | LOB% | ERA | FIP | WAR |
|---|---|---|---|---|---|---|---|---|---|
| Adam Wainwright | 197.0 | 7.6 | 1.9 | 0.7 | .301 | 73.3 % | 3.12 | 3.13 | 3.5 |
| Lance Lynn | 183.0 | 8.4 | 3.1 | 0.8 | .303 | 73.7 % | 3.46 | 3.54 | 2.6 |
| John Lackey | 184.0 | 7.4 | 2.0 | 1.0 | .300 | 72.9 % | 3.60 | 3.65 | 2.4 |
| Michael Wacha | 149.0 | 8.2 | 2.7 | 0.8 | .298 | 73.8 % | 3.41 | 3.48 | 2.1 |
| Carlos Martinez | 128.0 | 7.7 | 3.2 | 0.7 | .309 | 71.2 % | 3.75 | 3.58 | 1.7 |
| Jaime Garcia |
57.0 | 7.5 | 2.2 | 0.9 | .307 | 72.1 % | 3.65 | 3.54 | 0.7 |
| Marco Gonzales | 28.0 | 7.8 | 3.0 | 0.9 | .298 | 74.1 % | 3.59 | 3.82 | 0.4 |
| Tyler Lyons | 28.0 | 7.2 | 2.4 | 0.9 | .301 | 71.5 % | 3.82 | 3.83 | 0.3 |
| Tim Cooney | 19.0 | 6.5 | 2.5 | 1.0 | .302 | 71.1 % | 3.99 | 4.05 | 0.1 |
| Total | 974.0 | 7.8 | 2.5 | 0.8 | .302 | 72.9 % | 3.48 | 3.50 | 13.9 |
Behold their boring brilliance. The Cardinals went from sixth in last year’s pre-season rankings to eleventh in last year’s end-of-season rankings, and up to #3 this year. A big reason that the Cardinals underperformed expectations in 2014 — but are expected to improve in 2015 — is that the Cardinals didn’t have a fifth starter, but now (barring, of course, injury) they do.
Joe Kelly was projected to serve as a reliable fifth starter for last year’s Cardinals, but found himself traded to the Boston Red Sox after injuries limited him to just seven first-half starts. In return on the trade, Lackey actually did serve as a reliable starter, providing consistency over ten late-season starts and two postseason starts. That consistency was pretty crucial for a team that uncharacteristically started 12 pitchers last year.
While the exact identity of this year’s fifth starter is not yet nailed down, the 2015 Cardinals at least have a variety of options available to them — unlike the 2014 Cardinals. While Garcia has (yet again) eliminated himself from the competition due to (yet again) injury, St. Louis has two appealing homegrown, cost-controlled options to fill out the staff. 2013 first-round pick Gonzales already has postseason innings (as a reliever) under his belt, and has posted a sterling 1.04 ERA over five appearances this spring. He is in competition with bullpen convert Martinez, who has struck out 16 batters in his 16 spring innings. Factor in the eventual 40-50 innings that Garcia will probably manage this season, and the Cardinals’ pitching duties look like they could be split amongst seven pitchers instead of last year’s twelve-man staff, which included desperate late-season scrambles like Masterson.
Obviously, this staff isn’t rated so high just because of the potential of its back end. While Lynn’s K/9 rate has fallen in each of the two years since his 2012 debut (from 9.2 to 8.84 to 8.0), his HR/9 rate has done the same (from 0.82 to 0.62 to 0.57). Lynn’s end-of-season FIP has ended up in a remarkably narrow window between his three full seasons — between 3.28 and 3.49 — meaning that his metronomic excellence is pretty darn similar to that of his franchise.
We’ve seen enough of Wacha that it’s easy to forget he’s 23 years old: this is a pretty stellar projection for a 23-year-old who has yet to pitch 20 starts in a season! He’s younger than Marcus Stroman, Danny Salazar, Eddie Butler, and many more pitchers who still have the exciting sheen of a prospect on them.
At the top, Wainwright remains one of the game’s few bona fide aces. The killer thing is, he’s not totally being paid like one. While his annual $19.5M price tag from here through 2018 is far from a team-friendly robbery, but that’s cheaper annual value than the much-less-efficient Justin Verlander ($28M in 2015), CC Sabathia ($23M in 2015), Matt Cain ($20.8M), or Mark Buehrle ($20M). It’s hard to get a dollar-per-win bargain at that level of the MLB tax bracket, but doggone it, the Cardinals sure have done it.
#4 Padres
| Name | IP | K/9 | BB/9 | HR/9 | BABIP | LOB% | ERA | FIP | WAR |
|---|---|---|---|---|---|---|---|---|---|
| James Shields | 217.0 | 8.1 | 2.1 | 0.7 | .304 | 73.5 % | 3.25 | 3.24 | 3.9 |
| Ian Kennedy | 200.0 | 8.8 | 3.0 | 0.9 | .305 | 72.9 % | 3.71 | 3.65 | 2.3 |
| Tyson Ross | 169.0 | 8.9 | 3.3 | 0.6 | .304 | 73.2 % | 3.33 | 3.29 | 2.6 |
| Andrew Cashner | 163.0 | 7.4 | 2.5 | 0.7 | .301 | 72.2 % | 3.50 | 3.43 | 2.3 |
| Odrisamer Despaigne | 93.0 | 7.4 | 3.3 | 0.8 | .299 | 71.0 % | 3.87 | 3.87 | 0.9 |
| Brandon Morrow |
73.0 | 8.5 | 3.2 | 0.8 | .305 | 72.4 % | 3.69 | 3.55 | 0.9 |
| Josh Johnson |
56.0 | 7.8 | 3.0 | 0.8 | .311 | 70.8 % | 3.92 | 3.66 | 0.7 |
| Robbie Erlin | 9.0 | 7.2 | 2.6 | 1.0 | .305 | 71.1 % | 3.98 | 3.88 | 0.1 |
| Matt Wisler | 9.0 | 7.6 | 2.8 | 1.0 | .303 | 71.1 % | 4.03 | 3.97 | 0.1 |
| Total | 990.0 | 8.2 | 2.8 | 0.8 | .304 | 72.6 % | 3.54 | 3.48 | 13.7 |
Perhaps lost among all the Padres’ famously prolific wheeling and dealing this winter: San Diego had an average staff and an upper-echelon bullpen last season, and, save for the addition of Shields, the arms on their roster have remained more or less intact. Additions like Derek Norris, Matt Kemp, and Justin Upton will not necessarily have to provide elite offense for this team to be successful — their staff, working in a pitcher’s park, will likely carry this team however far it can go.
While GM A.J. Preller did acquire injury-prone Brandon Morrow and Josh Johnson — who had previously been united on the 2013 Toronto Blue Jays, one of the most disappointing splashy-winter teams in recent memory — the difference is that Preller acquired them both for the combined cost of $3.5M, and the Padres also have five capable pitchers ahead of those two on the depth chart. Also to consider: the Padres also acquired Brandon Maurer for Seth Smith over the winter. While Maurer is being used as a reliever this spring, and thus is projected as a member of the Padres’ bullpen, he also has experience as a starter should San Diego ever need an emergency spot start.
In other words: viable major league starters will be on the outside looking in at this rotation, a position of luxury that even heavily bankrolled teams like the New York Yankees or Detroit Tigers can’t claim. Even more encouraging for the Padres: of the top five men in the depth chart, only Cashner has a history of injury. Shields and Kennedy routinely hurl 200 innings a year, while Ross and Despaigne have clean records of health in their young careers. It should be another year of low-scoring games in San Diego — just, this time, the Padres have the arsenal to end up on top more times than not.
#5 Indians
| Name | IP | K/9 | BB/9 | HR/9 | BABIP | LOB% | ERA | FIP | WAR |
|---|---|---|---|---|---|---|---|---|---|
| Corey Kluber | 212.0 | 9.6 | 2.0 | 0.7 | .312 | 74.2 % | 3.07 | 2.85 | 5.0 |
| Trevor Bauer | 171.0 | 8.5 | 3.9 | 1.0 | .310 | 70.6 % | 4.38 | 4.20 | 1.1 |
| Carlos Carrasco |
162.0 | 8.4 | 2.6 | 0.8 | .312 | 71.4 % | 3.64 | 3.33 | 2.6 |
| T.J. House | 130.0 | 6.3 | 2.9 | 0.9 | .316 | 68.9 % | 4.38 | 4.12 | 0.8 |
| Zach McAllister | 103.0 | 7.4 | 2.6 | 0.9 | .313 | 70.1 % | 4.08 | 3.77 | 1.3 |
| Danny Salazar | 118.0 | 9.8 | 3.0 | 1.0 | .309 | 73.7 % | 3.65 | 3.51 | 1.9 |
| Josh Tomlin | 47.0 | 7.2 | 1.7 | 1.2 | .313 | 69.4 % | 4.26 | 3.94 | 0.5 |
| Gavin Floyd |
18.0 | 7.2 | 2.8 | 1.1 | .312 | 70.6 % | 4.32 | 4.21 | 0.1 |
| Total | 962.0 | 8.4 | 2.7 | 0.9 | .312 | 71.4 % | 3.84 | 3.60 | 13.4 |
As August pointed out earlier this offseason, the Indians starting rotation is, in terms of $/WAR, unmatched across the entire league. That was true in November, when their top five starters were projected to create 11.0 WAR while earning a combined $3.5M, and it’s true now, when their top six starters are projected to create 12.8 WAR while earning a combined $5.35M. Here are other assets on the pitching market that teams will pay $5.35M for in 2015:
-Six or seven starts by Verlander.
-Seven or eight starts by Sabathia.
-Nine or ten starts by C.J. Wilson.
-A half-season of Bronson Arroyo’s rehab.
-One season of Kyle Kendrick.
Clearly, by ending up so high on this list of projections, Cleveland isn’t pulling any cheap small-market Scrooge-ery here, either. Salazar and his career 10.28 K/9 rate just got sent down to Triple-A last week, unable to crack the rotation ahead of McAllister. FanGraphs has the team projected to win the AL Central thanks to the firepower this rotation provides. This rotation had the best K/9 rate in the majors last year, and by a significant margin, and they’re projected to be (slightly) outpaced by only the Nationals this year.
With so comparatively little invested in this rotation, the Cleveland front office took a very calculated risk in signing injury-prone Floyd to a 1-year/$4M show-me contract. Since the team does not need Floyd to fill any particular role, he will be allowed the luxury of unrushed rehab. While it’s impossible for Floyd to match the dollar-per-dollar efficiency of the rest of the staff, this still feels like a deft move for the front office to pull off, given the depth and youth the team has throughout the rest of the staff.
What’s exciting about this projection for Indians fans is that it already has built in a considerable amount of regression for Kluber, coming out of his breathtaking breakout season. Where Kluber produced at a rate of 6.1 WAR per 200 innings last season, the projection above has him going for 4.7 WAR per 200 innings, a more-than-20% drop-off in production per inning. Kluber does not have to win another Cy Young for this rotation to be the best in the American League. And if Kluber does put together another Cy Young-caliber season? Well, shoot.
#6 Mets
| Name | IP | K/9 | BB/9 | HR/9 | BABIP | LOB% | ERA | FIP | WAR |
|---|---|---|---|---|---|---|---|---|---|
| Matt Harvey | 173.0 | 9.1 | 2.6 | 0.7 | .296 | 74.1 % | 3.07 | 3.08 | 3.5 |
| Bartolo Colon | 176.0 | 6.2 | 1.5 | 1.0 | .308 | 70.3 % | 4.00 | 3.80 | 2.0 |
| Jon Niese | 178.0 | 6.7 | 2.4 | 0.9 | .307 | 71.1 % | 3.91 | 3.83 | 1.8 |
| Jacob deGrom | 165.0 | 8.2 | 2.7 | 0.8 | .306 | 72.6 % | 3.53 | 3.40 | 2.5 |
| Dillon Gee | 98.0 | 6.7 | 2.5 | 1.1 | .300 | 70.5 % | 4.23 | 4.24 | 0.6 |
| Noah Syndergaard | 93.0 | 9.2 | 2.7 | 0.9 | .310 | 72.7 % | 3.54 | 3.37 | 1.5 |
| Rafael Montero | 56.0 | 8.7 | 3.0 | 0.9 | .301 | 73.8 % | 3.62 | 3.68 | 0.9 |
| Steven Matz | 46.0 | 7.9 | 2.9 | 0.8 | .306 | 72.2 % | 3.69 | 3.65 | 0.6 |
| Total | 985.0 | 7.7 | 2.4 | 0.9 | .304 | 72.0 % | 3.68 | 3.60 | 13.4 |
What is this? Is this evidence of the Mets’ future looking…bright? Even with Zack Wheeler and his 2.3 ZiPS-projected WAR wiped out due to T.J., this team is right in line with the analytics-savvy Indians and the grandiose-spending Cubs. Last season, this staff provided comparable season numbers to the Cardinals and A’s, and that was with Harvey quite sadly sitting out the whole season with T.J. of his own. Not only is Harvey healthy, but both his Steamer and ZiPS projections have baked in a considerable amount of regression from his 2013 opus — ZiPS is predicting that Harvey will pitch in 2015 with only half of his per-inning effectiveness. And still the Mets’ rotation ranks this well. If Harvey doesn’t post a career-worst K/9 rate or a career-worst strand rate — which ZiPS projects him to do both — then look out.
Between ownership and the front office, from Bernie Madoff to Bobby Bonilla, the Mets have not exactly been known for shrewd financial stewardship, but this pitching staff is a refreshing exception to that rule, with this whole crew coming in at below the cost of one Verlander in 2015. The lion’s share of the money is going to Colon, who is, thrillingly, almost matching his prime earning wages from a decade ago here in his age-42 season. While Colon is definitely the one thing that is not like the others on this staff, he could end up being worth all of it: this rotation is the team’s foremost strength, and the Mets are projected to have a reasonable shot at a Wild Card spot, which would be their first playoff entry since 2006 — not to mention their first winning season since 2008.
Since a thundercloud of pessimism and disaster seems to follow the Mets through the years, however, it seems hard right now, before Opening Day, to imagine a scenario where the Mets’ 2015 season breaks in the right direction. It’s okay: Colon aside, this is the pitching staff of a youthful rebuilding team, with a bounty of future years under team control for reasonable cost. Gee will be the first starter to hit the free agent market, which won’t happen until 2017. Niese is around until 2018, with club options for the last two years of his deal. Harvey isn’t a free agent until 2019; Wheeler until 2020; deGrom and Montero until 2021; Syndergaard until 2022 at the earliest. The gang is bound to not be decimated by season-ending surgery one of those years, right? Right?
The only other question is: if the Mets can routinely draft and trade for pitching diamonds in the rough like these guys, why haven’t they been doing the same with position players?
#7 Cubs
| Name | IP | K/9 | BB/9 | HR/9 | BABIP | LOB% | ERA | FIP | WAR |
|---|---|---|---|---|---|---|---|---|---|
| Jon Lester | 214.0 | 8.2 | 2.3 | 0.8 | .303 | 74.7 % | 3.23 | 3.31 | 4.2 |
| Jake Arrieta | 173.0 | 8.7 | 3.3 | 0.8 | .301 | 72.8 % | 3.62 | 3.61 | 2.6 |
| Jason Hammel | 161.0 | 7.7 | 2.6 | 1.0 | .301 | 71.8 % | 3.88 | 3.86 | 1.8 |
| Kyle Hendricks | 161.0 | 6.4 | 2.1 | 0.8 | .303 | 71.1 % | 3.73 | 3.73 | 2.3 |
| Travis Wood | 131.0 | 7.2 | 3.4 | 1.1 | .296 | 71.8 % | 4.27 | 4.44 | 0.9 |
| Edwin Jackson | 66.0 | 7.7 | 3.1 | 1.0 | .315 | 69.6 % | 4.36 | 3.98 | 0.7 |
| Tsuyoshi Wada |
38.0 | 7.9 | 2.6 | 1.0 | .303 | 73.6 % | 3.75 | 3.84 | 0.5 |
| Jacob Turner |
28.0 | 6.4 | 2.9 | 1.0 | .312 | 69.0 % | 4.56 | 4.28 | 0.2 |
| Total | 973.0 | 7.7 | 2.7 | 0.9 | .302 | 72.3 % | 3.77 | 3.77 | 13.1 |
So this is kind of crazy, but: after securing the most high-profile free agent of the winter, the Cubs’ rotation is projected to produce…fewer WAR than they produced last year. Huh?
Last year’s Cubs starters came in at 14.3 WAR, tied with the Rays for fifth-best in the majors. That staff was led by Jeff Samardzija, who is gone, Hammel, who left and came back, Arrieta, and Hendricks — all of whom are projected for notable regressions here in 2015. That might seem harsh of the projection systems — but, then again, here at FanGraphs we’re projecting Hammel and Arrieta to well surpass their career averages in 2015. Just, they’re projected to not surpass those career averages by as much as they did in 2014.
For his career, Hammel has had a 4.23 FIP as a starter, which includes his 17 starts as a Cub in the first half of 2014, when he carried a 3.19 FIP. His 3.86 projected FIP feels like a pretty reasonable midpoint. The same goes for Arrieta: 4.06 starter’s FIP for his career (including 2014); 2.26 last year; 3.61 projection.
Put it all together and, combined with Samardzija’s not insignificant absence, the Cubs needed to acquire a pitcher as talented as Lester just to keep pace with last year’s surprisingly excellent staff.
What will be fascinating to watch, though, is to see if the Cubs as an organization have found a way to ensure that pitchers succeed projected performance. The staff that produced a 5th-best 14.3 WAR last year was only projected to contribute a 21st-best 9.7 WAR, and that was with assumed full-season contributions from Samardzija and Hammels. What in tarnation is Theo building over there?
#8 Mariners
| Name | IP | K/9 | BB/9 | HR/9 | BABIP | LOB% | ERA | FIP | WAR |
|---|---|---|---|---|---|---|---|---|---|
| Felix Hernandez | 217.0 | 9.3 | 1.9 | 0.6 | .302 | 75.3 % | 2.79 | 2.74 | 5.1 |
| Hisashi Iwakuma |
204.0 | 7.4 | 1.6 | 1.0 | .295 | 73.3 % | 3.40 | 3.48 | 2.9 |
| James Paxton | 170.0 | 7.2 | 3.4 | 0.8 | .301 | 71.3 % | 3.92 | 3.91 | 1.5 |
| J.A. Happ | 158.0 | 7.6 | 3.2 | 1.1 | .300 | 72.0 % | 4.13 | 4.14 | 1.0 |
| Taijuan Walker | 130.0 | 8.0 | 3.3 | 0.9 | .298 | 71.2 % | 3.98 | 3.98 | 1.2 |
| Roenis Elias | 73.0 | 7.4 | 3.5 | 1.0 | .300 | 71.1 % | 4.17 | 4.24 | 0.4 |
| Danny Hultzen |
19.0 | 8.2 | 4.4 | 0.8 | .301 | 72.5 % | 3.97 | 4.09 | 0.1 |
| Mike Kickham | 9.0 | 7.3 | 3.8 | 0.9 | .306 | 70.8 % | 4.26 | 4.20 | 0.0 |
| Total | 981.0 | 7.9 | 2.7 | 0.9 | .299 | 72.6 % | 3.63 | 3.63 | 12.3 |
How long can an elite prospect continually be called an elite prospect? Whatever the answer, Paxton and Walker are sure to be bestowed with new labels — whether complementary or no — by the end of the season. Paxton is only two years younger than Hernandez — he’s also pitched nearly 2,000 fewer major league innings than Hernandez as well. Walker has pitched 53 major league innings. They’ve been good innings, but other top prospects of 2012 like Gerrit Cole, Julio Teheran, or Jose Fernandez have all already compiled more than 200 innings in the big show.
The Mariners really need both young pitchers to pitch and pitch well for the full season, and contributing full seasons isn’t something either player has done much of lately. Paxton has only pitched more than 125 innings once in his four seasons in the Mariner organization (during 2013, when he made his big-league debut after spending most of the season in Triple-A) and was healthy for less than half of the schedule in 2014. Walker hasn’t pitched in the big leagues without a short leash — he’s only seen the seventh inning in one of his eight career starts.
What’s scary about this Mariners season is that this is when their championship window might be open the widest: now into their third season of having Iwakuma as one of the game’s greatest free agency bargains, the stoic pitcher will hit the open market next winter. Unless Paxton or Walker take big leaps forward, the Mariners will go into 2016 without a solid #2 behind Hernandez — as, alas, has been the case for most of Felix’s years with the team.
As for Hernandez himself, his accomplishments are beginning to inch further and further up the record books. Amongst active players, Hernandez is seventh in complete games, sixth in strikeouts, and fourth in (Baseball-Reference) WAR. He turns 29 next week.
#9 Yankees
| Name | IP | K/9 | BB/9 | HR/9 | BABIP | LOB% | ERA | FIP | WAR |
|---|---|---|---|---|---|---|---|---|---|
| CC Sabathia |
189.0 | 7.6 | 2.3 | 1.1 | .306 | 70.4 % | 4.14 | 3.96 | 2.2 |
| Masahiro Tanaka | 158.0 | 8.3 | 1.8 | 0.9 | .300 | 72.8 % | 3.39 | 3.37 | 2.9 |
| Michael Pineda | 142.0 | 7.4 | 2.1 | 1.1 | .298 | 72.9 % | 3.73 | 3.79 | 2.1 |
| Nathan Eovaldi | 159.0 | 6.5 | 2.6 | 1.1 | .307 | 69.8 % | 4.39 | 4.25 | 1.3 |
| Adam Warren | 141.0 | 8.0 | 3.1 | 0.9 | .298 | 72.9 % | 3.76 | 3.86 | 1.8 |
| Ivan Nova |
56.0 | 7.3 | 2.8 | 1.0 | .308 | 70.9 % | 4.10 | 4.07 | 0.5 |
| Chris Capuano |
37.0 | 7.3 | 2.9 | 1.2 | .306 | 70.7 % | 4.36 | 4.26 | 0.3 |
| Chase Whitley | 38.0 | 7.3 | 3.0 | 1.1 | .307 | 70.2 % | 4.41 | 4.30 | 0.3 |
| Bryan Mitchell | 28.0 | 6.6 | 4.4 | 1.3 | .305 | 68.7 % | 5.19 | 5.10 | 0.0 |
| Esmil Rogers | 28.0 | 7.3 | 3.1 | 1.1 | .307 | 70.0 % | 4.44 | 4.34 | 0.2 |
| Total | 976.0 | 7.5 | 2.5 | 1.1 | .303 | 71.3 % | 4.00 | 3.95 | 11.6 |
Last year, thirteen different pitchers started for the Yankees. Nine of them started at least ten games; only two of them pitched over 100 innings; only one of them pitched over 150 innings; and that guy (Hiroki Kuroda) is gone now. But Brian Cashman fought and he scratched and he clawed and the Yankees finished a respectable seventh overall in pitcher WAR in 2014.
On the one hand it seems like the invisible forces of regression would prevent the Yankees from experiencing anything close to that sheer tonnage of injuries here in 2015. On the other hand, there are a lot of pitchers here who, fairly or not, have to be called injury-prone. Pineda’s 13 starts in 2014 were his his major-league appearance since 2011. Nova has hit the DL in each year since 2012, culminating in last year’s four-start mulligan. Even though Tanaka has started twenty games each year going back to his age-18 season in 2007, his 2+ months missed in his first year stateside is anything but a positive omen. Last year, Sabathia might have taken a Bartolo-esque dip that he’ll soon recover from — or his wheels might have fallen off, Roy Halladay-style.
What Yankees fans can all but be assured of, though, is that their team will proactively maneuver to solve problems should this depth chart go awry. Last year’s trade-deadline acquisition of Brandon McCarthy went a long way in helping the team get as close to a playoff seed as they eventually did. Minor-league call-ups Greene and Whitley were supplemented with the mid-season addition of Capuano — who’s about as good a pitcher as one could hope to find available just for cash in the middle of the season. 2014’s injury-riddled debacle just might have represented a worst-case scenario for this staff — and really, things weren’t all that bad.
#10 Tigers
| Name | IP | K/9 | BB/9 | HR/9 | BABIP | LOB% | ERA | FIP | WAR |
|---|---|---|---|---|---|---|---|---|---|
| David Price | 220.0 | 8.6 | 1.7 | 0.9 | .313 | 73.2 % | 3.45 | 3.24 | 4.1 |
| Justin Verlander |
209.0 | 7.7 | 2.7 | 0.9 | .306 | 71.9 % | 3.83 | 3.72 | 2.9 |
| Anibal Sanchez | 163.0 | 8.0 | 2.3 | 0.8 | .308 | 71.6 % | 3.55 | 3.31 | 2.8 |
| Alfredo Simon | 146.0 | 5.4 | 2.8 | 1.1 | .306 | 68.5 % | 4.71 | 4.59 | 0.4 |
| Shane Greene | 130.0 | 7.0 | 3.3 | 0.9 | .315 | 69.4 % | 4.45 | 4.19 | 0.8 |
| Buck Farmer | 47.0 | 6.8 | 3.2 | 1.4 | .310 | 68.2 % | 5.06 | 4.85 | -0.1 |
| Kyle Lobstein | 38.0 | 6.3 | 3.2 | 1.1 | .312 | 68.4 % | 4.76 | 4.47 | 0.2 |
| Kyle Ryan | 28.0 | 4.7 | 2.7 | 1.3 | .308 | 66.8 % | 5.22 | 5.01 | -0.1 |
| Drew VerHagen | 19.0 | 4.6 | 3.0 | 1.1 | .308 | 67.0 % | 5.01 | 4.76 | 0.0 |
| Total | 999.0 | 7.3 | 2.6 | 1.0 | .310 | 70.6 % | 4.07 | 3.87 | 11.0 |
Perspective is key here: while there are twenty other teams, presumably, who would prefer to be at this spot on the projection hierarchy instead of at their own, for the Tigers this is a fall from grace, a crumbling of the empire.
In 2014 the Tigers were projected to be the majors’ best staff in this pre-season exercise, and ended up with the second-best WAR total. And then in the 2013 season, the Tigers were projected to have the best rotation, and that year’s staff lapped the field in any category you could want. Oh: and they had the most productive staff of 2012 as well. Dropping down to 10th would be something of a free-fall for this team and, alas, there is no obvious and/or cheap way to reverse course in the near future.
Unfortunately, the Tigers have nowhere to point the finger but in the mirror. They simply extended the wrong dude — it was all the way back in Spring Training 2010, but the ripples from Verlander’s $180M deal, which runs from 2013-19, have not stopped spreading. We’re entering the third year of the seven-year pact, and already Dave has listed the Verlander contract as the second-most-burdensome deal in all of baseball.
Perhaps financial concerns stemming from Verlander’s mega-deal went a long way in motivating the Tigers’ trade of Doug Fister to the Nationals in exchange for Robbie Ray, Shane Greene, and Ian Krol — a trade that you might recall Dave panning as the very worst deal of the 2013-14 offseason. What Dave didn’t even know then is that, less than 18 months later, only Krol would be left with the Tigers: Lombardozzi was traded before ever appearing in a Tigers uniform, and Ray was dealt to the Diamondbacks in a three-way trade that netted Greene. While Greene appears to be a fine young pitcher, only the most aggressive of rebuilding teams would swap him for Fister — an exchange that the Tigers have effectively pulled off. This is not a rebuilding team.
From the look of the tea leaves, the Tigers will slide even further down this list at the same time next year. Price, the best pitcher on this staff, is due to hit the free agent market — a market that the Tigers might very well be priced out of thanks to their $410M in commitments for 2016 and beyond to Verlander, Sanchez, Miguel Cabrera, Ian Kinsler, Victor Martinez, and Prince Fielder. (Yep — the Tigers are paying $30M of Fielder’s salary from 2016-2020.)
But hey: Sanchez is still really good!
#11 Giants
| Name | IP | K/9 | BB/9 | HR/9 | BABIP | LOB% | ERA | FIP | WAR |
|---|---|---|---|---|---|---|---|---|---|
| Madison Bumgarner | 211.0 | 8.9 | 2.1 | 0.8 | .294 | 76.0 % | 2.89 | 3.07 | 3.9 |
| Matt Cain | 180.0 | 7.6 | 2.6 | 1.0 | .290 | 73.0 % | 3.68 | 3.88 | 1.4 |
| Tim Hudson | 170.0 | 6.0 | 2.1 | 0.7 | .305 | 70.5 % | 3.75 | 3.65 | 1.7 |
| Jake Peavy | 171.0 | 7.2 | 2.4 | 0.9 | .293 | 73.6 % | 3.53 | 3.75 | 1.8 |
| Tim Lincecum | 122.0 | 8.0 | 3.5 | 0.9 | .305 | 70.6 % | 4.14 | 3.96 | 0.7 |
| Ryan Vogelsong | 85.0 | 6.9 | 2.9 | 0.9 | .300 | 71.0 % | 4.03 | 4.05 | 0.5 |
| Yusmeiro Petit | 56.0 | 8.2 | 2.0 | 1.0 | .300 | 73.2 % | 3.47 | 3.45 | 0.8 |
| Total | 994.0 | 7.5 | 2.5 | 0.9 | .297 | 72.7 % | 3.57 | 3.65 | 10.8 |
If there was ever a team that rendered the exercise of projections meaningless, here they are. Last year’s staff was projected to be 14th-best in the league, ended up at 20th-best, and then the World Series behind the strength of, among other things, their starting pitching.
Does this rotation look intimidating, even though it returns all eight guys who started for last year’s champions? Well, no. Actually, it looks especially not-intimidating now that Cain busted his personal streak of eight straight seasons with at least 30 starts and has to be considered a question mark going forward. Just when you think how disastrous a salary albatross Cain’s deal could be going forward (he is owed $70M over the next four seasons if the Giants choose to buy him out in 2018), you remember that the Giants won two World Series with Barry Zito as their highest-paid player. So, here are the Giants’ projections, but who knows what the heck they mean.
#12 Angels
| Name | IP | K/9 | BB/9 | HR/9 | BABIP | LOB% | ERA | FIP | WAR |
|---|---|---|---|---|---|---|---|---|---|
| Jered Weaver | 204.0 | 6.6 | 2.5 | 1.2 | .288 | 72.9 % | 3.98 | 4.30 | 1.5 |
| Matt Shoemaker | 198.0 | 7.3 | 1.9 | 1.0 | .299 | 72.2 % | 3.67 | 3.67 | 2.5 |
| C.J. Wilson | 193.0 | 7.7 | 3.9 | 0.9 | .303 | 70.9 % | 4.14 | 4.08 | 1.6 |
| Garrett Richards |
176.0 | 8.2 | 2.9 | 0.6 | .301 | 72.7 % | 3.27 | 3.24 | 3.2 |
| Hector Santiago | 113.0 | 7.5 | 3.6 | 1.0 | .295 | 73.1 % | 4.02 | 4.29 | 0.9 |
| Andrew Heaney | 55.0 | 7.6 | 2.7 | 1.0 | .302 | 72.2 % | 3.89 | 3.98 | 0.5 |
| Nicholas Tropeano | 38.0 | 7.6 | 2.8 | 0.9 | .302 | 70.8 % | 3.98 | 3.87 | 0.4 |
| Total | 977.0 | 7.4 | 2.9 | 0.9 | .298 | 72.2 % | 3.82 | 3.90 | 10.6 |
My apologies if the tone of this post is getting more and more pessimistic as we venture deeper down the ranks. The projected gap in production between these 12th-ranked Angels and the 30th-ranked Kansas City Royals (a difference of 5.2 WAR) is actually smaller than the gap between the 12th-ranked Angels and the 2nd-ranked Dodgers (a difference of 6.3 WAR). Plenty of the game’s brightest young individual talents can be found further down both this list and the accompanying post looking at staffs #’s 16-30. If nothing else, though, putting this post together is showing me just how rare and special it is for an entire staff of talents to be put together.
So, the Angels: they’re a very bizarre team in that, despite being unabashedly a big-market squad, heaping portions of their recent success has been generated from players that the team is paying at Tampa Bay-bargain prices. This goes for their offense — you mighta heard of this good-hittin’ fella named Mike Trout — and it goes for their pitching staff as well.
The Angels have a young nucleus of a pitching staff that would be the envy of any cost-conscious contender. Richards is earning $3.2M in 2015 (his first year in arbitration). Shoemaker, Tyler Skaggs (who will pitch in 2016 but who will miss the season due to a Tommy John surgery sustained late last summer), and Heaney all make the minimum, and all of them are under team control until after Josh Hamilton’s contract is done. Combined, these four pitchers will earn about what Weaver ($18.2M in 2015) or Wilson ($18.5) will make by Memorial Day. Hm.
And it is the uninspiring — or at least cost-ineffective — results from Weaver and Wilson that makes this staff feel old and bloated instead of young and on the rise. I can’t help but blame the two veterans for guilt by association: in recent the Angels have flushed such prolific amounts of money away on other veterans like Hamilton and Albert Pujols and Vernon Wells and Scott Kazmir, it feels like they are primed to do the same with these pitchers while the young guys carry the team.
#13 Pirates
| Name | IP | K/9 | BB/9 | HR/9 | BABIP | LOB% | ERA | FIP | WAR |
|---|---|---|---|---|---|---|---|---|---|
| Francisco Liriano | 169.0 | 9.3 | 3.7 | 0.6 | .304 | 73.3 % | 3.40 | 3.29 | 2.6 |
| A.J. Burnett | 184.0 | 7.8 | 3.5 | 0.7 | .310 | 69.9 % | 3.98 | 3.71 | 1.8 |
| Gerrit Cole | 177.0 | 8.3 | 2.6 | 0.7 | .308 | 72.5 % | 3.44 | 3.32 | 2.8 |
| Jeff Locke | 158.0 | 6.7 | 3.4 | 0.9 | .303 | 71.0 % | 4.14 | 4.15 | 0.8 |
| Charlie Morton |
122.0 | 6.7 | 3.0 | 0.7 | .307 | 70.2 % | 3.89 | 3.89 | 1.0 |
| Vance Worley | 66.0 | 6.2 | 2.1 | 0.8 | .312 | 70.3 % | 3.92 | 3.75 | 0.7 |
| Jameson Taillon |
47.0 | 7.2 | 3.1 | 0.8 | .307 | 71.1 % | 3.91 | 3.81 | 0.5 |
| Nick Kingham | 37.0 | 6.4 | 3.1 | 0.9 | .303 | 69.6 % | 4.28 | 4.22 | 0.2 |
| Total | 961.0 | 7.7 | 3.1 | 0.7 | .307 | 71.2 % | 3.80 | 3.69 | 10.3 |
This is, at the very least, one of the most intentionally assembled rotations in the majors. Neal Huntington and his team of analysts have put together a group of ground-ball pitchers (both starters and relievers) who generate a bunch of ground balls into Pittsburgh’s frequent defensive shifts. Over the last two seasons, Pirates pitchers have produced ground balls more frequently than any other staff — and it isn’t particularly close. I suspect this is how the Pirates have outperformed their FIP by one of the widest gaps in the majors over the same time span: they’ve produced a 3.38 ERA over the last two seasons despite a 3.61 ERA.
Clearly, great pains were taken to get the whole gang back together. By signing Liriano to a 3-year/$39M deal this winter, the Pirates made him their biggest free agent investment in franchise history, an action that speaks louder than any complimentary press-conference-speak. Burnett also returns after a sojourn to the other side of Pennsylvania. His leaving the analytically savvy Pirates for the decidedly old-school Phillies, followed by a return to the same Pirates makes Burnett’s 2015 season one of the more interesting single-season experiments going on the majors, in my opinion. Burnett’s FIP ballooned from 2.80 in 2013 (Pirates) to 4.14 in 2014 (Phillies).
But also: Burnett’s ground ball rate with the Pirates was significantly higher than it was with the Phillies (56.5% to 50.9%). Probably more importantly: in high-leverage situations with the Phillies, Burnett drifted away from the ground ball, generating one per fly ball. In high-leverage situations with the Pirates in 2013, Burnett ramped up the ground balls even higher, generating an incredible 3.83 ground balls per fly ball. If Burnett has a bounce-back year — which I do expect will happen — then the Pirates really deserve further credit for their implementation of analytics throughout the organization.
#14 White Sox
| Name | IP | K/9 | BB/9 | HR/9 | BABIP | LOB% | ERA | FIP | WAR |
|---|---|---|---|---|---|---|---|---|---|
| Chris Sale |
216.0 | 10.1 | 2.2 | 0.9 | .304 | 76.7 % | 2.96 | 3.02 | 5.1 |
| Jeff Samardzija | 208.0 | 8.8 | 2.5 | 1.1 | .312 | 71.7 % | 3.91 | 3.68 | 3.0 |
| Jose Quintana | 184.0 | 7.5 | 2.6 | 0.9 | .309 | 71.7 % | 3.84 | 3.70 | 2.5 |
| John Danks | 133.0 | 5.8 | 3.2 | 1.4 | .304 | 68.0 % | 5.20 | 5.10 | -0.2 |
| Hector Noesi | 113.0 | 6.2 | 3.2 | 1.6 | .305 | 68.3 % | 5.37 | 5.23 | -0.3 |
| Carlos Rodon | 75.0 | 10.2 | 5.6 | 1.2 | .307 | 74.7 % | 4.39 | 4.72 | 0.2 |
| Erik Johnson |
38.0 | 6.1 | 4.2 | 1.3 | .309 | 67.7 % | 5.37 | 5.16 | 0.0 |
| Brad Penny | 18.0 | 5.8 | 3.5 | 1.2 | .314 | 67.6 % | 5.19 | 4.89 | 0.0 |
| Frankie Montas | 9.0 | 7.2 | 4.5 | 1.4 | .310 | 68.3 % | 5.37 | 5.16 | 0.0 |
| Total | 995.0 | 8.1 | 3.0 | 1.1 | .308 | 71.6 % | 4.16 | 4.08 | 10.2 |
It remains to be seen if the White Sox’s winter spree was an admirable exercise in proactively seizing the day, or if it will end up being a cautionary tale about the perils of not building through the draft. This staff epitomizes a “stars and scrubs” approach more than any other: none of the other 30 teams have two projected sub-replacement starters among the top five on their depth chart. If the White Sox meet their projection, it would be a continuation of how last year went for the team: Sale and Quintana produced 10.1 WAR, while all the rest of the starters managed 0.6 WAR between them.
One side effect that befell last year’s team was an inability to sustain winning streaks. When a replacement-level starter is given the ball two or three days each week, it’s hard to string any wins together. The White Sox managed two four-game winning streaks last year. That’s not a very high apex to the season: the Rockies had a six-gamer, and the Rangers, Astros, and Red Sox all strung together seven-game streaks. Is this a problem that is actually solved with all the new big boppers in the lineup?
#15 Marlins
| Name | IP | K/9 | BB/9 | HR/9 | BABIP | LOB% | ERA | FIP | WAR |
|---|---|---|---|---|---|---|---|---|---|
| Jose Fernandez |
99.0 | 9.9 | 3.1 | 0.6 | .298 | 75.4 % | 2.89 | 2.94 | 2.0 |
| Mat Latos | 192.0 | 6.9 | 2.5 | 0.8 | .303 | 72.0 % | 3.68 | 3.65 | 2.7 |
| Dan Haren | 158.0 | 7.0 | 2.0 | 1.1 | .306 | 71.0 % | 3.99 | 3.87 | 1.5 |
| Henderson Alvarez | 166.0 | 5.3 | 2.1 | 0.7 | .306 | 70.6 % | 3.79 | 3.80 | 1.4 |
| Jarred Cosart |
132.0 | 6.1 | 3.9 | 0.6 | .301 | 70.7 % | 4.04 | 4.06 | 0.8 |
| Tom Koehler | 103.0 | 6.9 | 3.6 | 0.9 | .303 | 70.7 % | 4.25 | 4.22 | 0.4 |
| David Phelps | 47.0 | 7.8 | 3.3 | 0.8 | .303 | 72.1 % | 3.77 | 3.80 | 0.5 |
| Andre Rienzo | 47.0 | 6.9 | 3.9 | 0.9 | .309 | 69.9 % | 4.45 | 4.30 | 0.1 |
| Brad Hand | 9.0 | 7.3 | 3.8 | 0.9 | .302 | 71.7 % | 4.11 | 4.16 | 0.0 |
| Justin Nicolino | 9.0 | 4.6 | 1.9 | 0.9 | .305 | 67.9 % | 4.35 | 4.20 | 0.0 |
| Total | 964.0 | 6.9 | 2.8 | 0.8 | .304 | 71.4 % | 3.83 | 3.81 | 9.5 |
Because Jeffrey Loria is around, things with the Marlins tend to look like mirrored-funhouse versions of conventional baseball teams. But, by jove, it looks like an honest-to-goodness small-market rebuilding process has been gathering momentum in Miami. As this team has inched further up the projected playoff odds — what’s this? they have the same playoff odds as the defending champion Giants? — they’ve also began to progressively shell out more money for players, including pitchers.
While Haren and his $10M price tag are covered by the Dodgers, Latos’ $9.4M salary counts as a fundamentally different approach for this team. Last year, their highest-paid pitcher was Steve Cishek at $3.8M, and their highest-paid starter was Turner at $1M. Proportionally, the Marlins have committed to a huge leap in salary from 2014, which seems like a show of good faith that bode wells for both the franchise’s relationship with Giancarlo Stanton and also the franchise’s karmic relationship to the universe as a whole. What remains to be seen is if any of these young arms — Fernandez aside, being a Scott Boras client and all — will be locked in long-term, a la Christian Yelich, or if they will be replaced to continue the rebuilding anew, a la Nathan Eovaldi.
Miles Wray contributes sports commentary to McSweeney's Internet Tendency, Ploughshares, The Classical and Hardwood Paroxysm. Follow him on Twitter @mileswray or email him here.
Thanks for this, but last time I check, Carlos Carrasco and Sabathia are not injured. Also, the links to those 2 pitchers are broken.
Not a fan of the new WAR. I get why it was changed, but now it’s skewing things for some teams. In terms of Mets, it’s not a pitcher park in terms of fip park factors, but in reality it actually is a pitchers park (been in bottom 9 in run park factors every season). Nothing in new calculation takes this into account so nothing within the new WAR incorporates the fact that they play in a stadium that suppresses runs. This is going to make all the Mets WARs higher than what they should be. They have now been giving a pass for fact that Citi is a run suppressing park.
Colon with a 4 era in a pitcher park has similar WARs of players with better era in a hitter park.
Oh jeez, not this again. If you want to know why the new calculation is correct, please read everybody’s response to your claims in this thread:
http://www.fangraphs.com/blogs/instagraphs/war-updates/
WAR is based on FIP, not ERA. The conclusion is very simple:
” The Mets’ FIPs are not helped by Citi Field. The end.”
” The Mets’ FIPs are not helped by Citi Field. The end.”
” The Mets’ FIPs are not helped by Citi Field. The end.”
No. The mets WARs are helped by fip park factors.
Not sure how u can value a player when not taking into account how good their park is at suppressing runs.
Purpose of WAR is to value players.
The Mets’ FIP’s are hurt by Citi Field. WAR offsets that.
You are not getting my point.
Mets park gives up a lot less runs than an avg park. This is a fact. This means if you take the Mets pitchers, kept all variables equal, and put them in a neutral ball prk they would likely give up more runs.
This is no longer incorporated into the calculation we now use to determine how valuable a player is.
I am not saying the calculation is incorrect, I am simply saying any value derived from a WAR that doesnt actually incorporate a ball parks ability to suppress runs is not actually giving you what I believe to be an accurate interpretation of that player.
If player X and player Y have all things equal, but player X does it in a park that suppresses runs than plyer Y should clearly be worth more wins to his team. Thats not the case anymore.
The point of WAR is to compare all players on the same scale, based on their actual skill. Ideally, we could directly measure a player’s skill units directly. These obviously are not impacted by ballpark, so if we measured these skill units, it wouldn’t matter that the Mets park suppresses offense. The park affects the results, and in this instance, we are not measuring the results.
This is kind of what FIP is. It’s a way of more directly measuring a player’s skill, by looking past the results. We want to measure every player’s skill score on the same scale, and if the Mets’ park doesn’t affect our measurement of that score, then it shouldn’t be adjusted for, as long as we’re calculating WAR based on that rating.
Put another way, we don’t have to adjust for park because our method of measurement has already adjusted for it, by looking at the inputs rather than the output.
Here’s the point in a real example. Because the Mets’ park suppresses runs, but doesn’t suppress FIP, then basically we should expect their FIPs to be worse than their ERA. Let’s say this year the Mets staff has a 3.60 ERA and a 3.75 FIP, and 3.75 was their true talent. That means our choice would be to use ERA and adjust for park, or just directly use FIP. Either way, we get to the same place.
Basically, in your example, player Y would have a better FIP because while player’s X’s park suppresses runs, it doesn’t suppress FIP, so unadjusted FIP would show both pitchers for what they are.
@Bip
The purpose of WAR is NOT to measure skill, but to attribute value.
This is my issue with the new changes (and also the old formula). On the offensive side we don’t attempt to measure how many hits a batter *should* have had, we are valuing the hits they did have.
With FIP we are guessing. In addition, by taking away the run based park factors we have completely removed the calculation from what ultimately happens on the field.
In what way does the current method for WAR calculation reconcile the theoretical, estimate run values with what ultimately happened on the field? That is the big question.
“If player X and player Y have all things equal, but player X does it in a park that suppresses runs than plyer Y should clearly be worth more wins to his team”
No. Here’s another way to think about it. War is WINS above replacement. If you’re pitching well simply because you’re doing it in a pitcher’s park, you’re not providing your team with any additional wins, because the team you’re playing against also benefits from pitching in said park on that day.
Minor point, but Preller was still in the Rangers organization when the Padres signed Josh Johnson.
That’s true if you’re talking about the contract they signed Johnson to for 2014 but Preller declined the option for 2015 under that contract. A few months later they signed him again for $1 million guaranteed instead of the $4 million guaranteed that he would have been owed under the option from the 2014 contract.
I’m just sitting here loving how underrated the Mets position players are, not to mention the prospects who haven’t come up yet. The Mets lead the majors with a +54 run differential and their bats are all red hot. Look out ya’ll
Agreed. There have been several studies proving that WAR severely underrates Mets pitchers. That rotation is much better than they’re getting credit for.
If that’s true, then that should be better now. The Mets pitchers should see their WAR go up, because their FIPs were being treated as if they were being suppressed, and now they are not.
That’s correct. I think they were ranked 20th a couple weeks ago. After the corrections, they’re 6th.
Lead the majors in run differential…in spring training?
File under “who cares” and “wake me up when the games count”.
I believe the Mets are actually in a 29-way tie for first in run differential with 0. The Phillies somehow are already at -11.
As a fan a analytics how can you not be a fan of Pittsburgh’s system? Win or lose they fully committed to this groundball+shift philosophy and it’s been fun watching em beat their projected numbers year after year.
Jose Quintana has been a 5.0+ WAR guy over the last few seasons and he projects half that. Meanwhile, Jake Arietta has had 2/3 of 1 good season for his ENTIRE CAREER…and he projects 0.1 higher than Quintana.
Is it because one pitches to a DH every 9 AB, while the other essentially gets an automatic out every 9 AB?
No, it’s because Fangraphs has had absolutely no clue what it’s doing when projecting Quintana since they started doing so. By their projections, Quintana should have a career WAR of 4.8 – his career WAR is 9.8. Pretty hilarious.
The number of people who think an objective projection system is actually a wild judgment call Mr. Fangraphs is giving me a migrane.
Clearly it’s not called Mr. Fangraphs; the fact is, fangraphs ties themselves to the system. For the vast majority of players, clearly, the projection system falls within 1 SD and is fair/reasonable. There are outliers though and instead of finding out why, fangraphs/ZIPS/steamer just allows the same players to consistently be undervalued. Mark Buehrle was under valued, I believe, 11 out of 12 years to start his career and there was never an adjustment made. It’s lazy.
How do you think these systems work? You think that they should go in there and throw in the “Mark Buehrle exception”?
if (player.lastname == “Buehrle”) {
player.projWAR = player.projWAR + 1;
}
It’s unfortunate that these projections take so much heat on here and the makers are not here to defend them at all.
These systems project thousands of players. First of all, with that many players, random chance should produce players who seem to beat the projections every year, just by chance. So there’s difficulty even knowing what is noise and what is signal.
And then, there’s the matter of actually discovering what Buehrle does. Let’s say that the only thing that he does is consistently hit corners and avoid the middle of the plate, but only with men on base. Then, let’s say the only place where that shows up is that his LOB% is better than one would expect given his other numbers. How would you incorporate that into a projection system? We don’t have that data in an easily usable form. So, sure, it’s a thing the system doesn’t account for, but that doesn’t mean it’s easy to incorporate.
And then, there’s the possibility that we simply don’t have a clue how some of these player beat their projections.
You’re just being unnecessarily dismissive and uncharitable. The fact that the system has weaknesses means that the makers are lazy? Please.
Falls within 1 SD of what? And furthermore, are you really suggesting that Mr. FanGraphs should just pick and choose certain players as “Oh, there’s no way these projections are right. Let’s fudge em a bit to make em good!”? That’s absurd.
The projections are just projections. They’re not predictions or being judgmental about your favorite player. Players beat and underperform projections all the time. Don’t take them so personally.
It’s the people who self-label, loudly, as “rational” that are typically anything but. (You’ll see that a LOT with politicians and political commentators also.)
Do I think projection systems should go back, and judge each player individually? Uh, yes. I didn’t realize that was a question.
Yes, they project thousands of players. I never said it was easy to incorporate; I understand these are free projection systems and they will have their flaws. That also means that me calling attention to said flaw is reasonable, and understandable…. so why it is such an outcry when I do such is laughable. Is it because I don’t feel the need to post my approach? Fair enough, but obviously it wouldn’t make any sense for me to do so in a comment section on fangraphs.
The fact remains if sportsbooks were offering odds on Quintana’s WAR and set the O/U at 2.5, the modeling market/community abound would be pounding the over 2.5.
Also, I’m pretty sure I’m being more than rational. Somehow it’s irrational because I’m calling to attention a poor projection that is probably widely regarded as a poor projection.
I’ll say I made the proper adjustment to Buehrle, and it wasn’t overly complex. I’ll also add that I don’t have the task of projecting every player, MiLB and MLB, that ZIPS/STEAMER does. Obviously this allows me more time to dig into certain inefficiencies individually. As I said, while the basis for every projection stems from a similar set, all players are not projected based on the exact same set of numbers. That wouldn’t make any sense.
Or that “he” makes decisions based on the identity of the player, not the information its fed.
Ok. Well then it’s obvious your projections and these projections are at cross-purposes. The purpose of these projection systems is to be objective. Putting in adjustments for particular players is the antithesis of this.
On the one hand, you’ll project existing major league players better. However, your ability to do this will increase only gradually as the players accumulate more time in the bigs, at which point the players are more or less known-quantities. However, this approach gives you no help whatsoever when projecting new players, or players who have had radical changes in their approach.
On the other hand, if you figure out why some pitchers outperform their FIPs or their projections, and instead of adding an exception for that player, you factor that player’s skills into the projection, then you have something that works for not only that player, but for all future players who do that same thing. Jered Weaver has always had low BABIPs. The system could either just have a hard-coded Weaver BABIP projection, or it could incorporate the way his strange angle, tendency to pitch high, and fastball rise lead to a lot of pop-ups which leads to low BABIP. The former will work for Weaver, but the latter works for everyone.
So, sure, they could add individual player projections, but that would defeat the whole point of the system.
Fair point BIP. I understand the differences between objectives; my goal is to find undervalued commodities based on grander projection systems, while there’s is to be the most efficient given the large sample size. Their goal is to be reasonably effective in predicting every player, as opposed to projecting every single player effectively every single season. I have the same beef with Steamer/Zips projections in terms of progression; the nature of their model will always weigh the greater sample as opposed to the most recent results. Deciphering the validity of recent adjustments and progressions a given player makes year to year is much more intricate than simply using projecting each player based on their collective data. Predicting a breakout goes well beyond what peripheral numbers may have told you the year prior; although I do believe they remain extremely valuable and I use probably 99.9% of the same data that ZIPS/Steamer use when projecting the next year.
Only point I’d refute is the following:
“However, this approach gives you no help whatsoever when projecting new players, or players who have had radical changes in their approach.”
With new players I’m using pretty much the exact same projections as ZIPS; the only differences you might see is adjustment for age, and/or more weight on your most recent MiLB performance in comparison to your early results. A lot of time I will throw out your first year if you are a High School player, and I’ll use adjusted college stats if you were a college player.
lastly, changes in approach are the one thing that I think I have an edge. For one example, I was much quicker to react to Jose Bautista than others; I weighted Bautista’s final two months in 2009 more heavily than I did his prior two seasons. His change in approach was very loud and clear. That’s one example, and there were others that I misjudged, as will always be the case in this industry, but in general I think it’s easier to get out in front of those adjustments when you give them more weight.
@Phillies:
If I can predict which players will under and over perform their projections, then there is something wrong with it. If I was incapable of doing so, there would be absolutely no issues with it. Luck exists but is not predictive, therefore there’s nothing wrong with discounting it in projection systems.
The reasons for why Buehrle and other weak-contact-inducing starters, who don’t generate a lot of swings and misses, are undervalued is quantifiable. I say that with confidence because I have shown a capability to quantify it; as have countless others. In fact, I would say that what I am doing doesn’t rank in the top 25% of baseball modeling. Guys I know that post here – they might even occasionally drop an article – make my work look like childs play.
ZIPs/Steamer and the likes are great PUBLICLY available projection systems. I would never refute that. That said, they have their problems and I’ll never understand why people take it so personally when it is pointed out.
Anyone can predict which players will under or over perform their projections. That’s the whole point; predictions are not the same thing as projections. If, however, you are telling me that you can predict this with 100% accuracy every time, then, yes, you’ve found a flaw with the projection systems, and corrections should be made to adjust those flaws.
The beef, really, is that you’re touting your own personal projection system as being much better than ZiPS and Steamer, yet you aren’t providing the data for anyone to review. No one is going to take you or your projection system seriously if all we have to go on that it is successful is your word.
If your system is that much better, then yes, I agree that posting it in the FanGraphs comment section is not ideal. However, there is a link right below the banner where you can submit articles to the Community section. There are quite a few interesting articles there. A lot of really smart people post really smart things there! Show us this projection system, because otherwise, everyone will just assume, probably correctly, that you’re just blowing smoke.
I wouldn’t have to do it 100% of the time. In reality, 3 out of 4 would prove an inefficiency; even less actually, but that’s the example I used above.
I am not saying anything near that; in fact, I state the contrary above. I model current MLB players only; I am not tasked with modeling every player in professional baseball. My sample size allows me to make individual adjustments to players, and the goal of what I do when projecting is drastically different than what Steamer/ZIPS is trying to accomplish. Can you use Steamer/ZIPS to churn a profit within the market somehow? No. So for that reason, I would say what I am doing is more valuable. It’s very similar to Ken Pomeroy in college basketball; his metrics are terrific, but you can’t make money using Pomeroy’s numbers. That doesn’t mean they aren’t important, or sophisticated and it certainly doesn’t mean their validity should be questioned. All it means is that it is now publicly available, therefore killing it’s value. As I also noted above, I would say 99.9% (probably closer to just 95%) of my inputs are the exact same as ZIPS.
@Phillies;
Lastly, there doesn’t seem to be any value in what I do for fangraphs. That isn’t the market or direction they are headed so submitting said work would make even less sense. I offered some advice last year during a daily fantasy baseball discussion on one of the daily recommendation articles. I simply asked the author why he did not account for the games total, or the team totals, that were available at pinnaclesports.com or bookmaker.eu; or any other sportsbook for that matter. He laughed and made some statement about it being irrelevant because sportsbooks don’t predict anything. Quants, who invest millions in the market every single day, beg to differ and their models account for game and team totals; obviously. The more runs scored in a baseball game, means the more RBI’s, runs, and HR’s; which are all extremely valuable in daily. Same goes for finding a starting pitcher; a pitcher that pitches for a team who is -200 to win that day (66.6%), has a higher likelihood to record a win than a pitcher who is -105 (51.2%); wins are valuable, so when deciding between the two you obviously go with the -200 if they are close. That was a very basic angle, that was easy to learn and take, and it was something that was discredited because of a lack of understanding. If that was difficult to comprehend, I can only imagine the difficulties I would have in going in depth.
This does not appear to be a site for serious daily or yearly fantasy discussion. Not that I would share something that makes me money anyway, but I would certainly be more inclined to if the site was more active in that area.
And I don’t think anybody projected Trout to have 30 career WAR at this point. If they projected him to have three 5 WAR seasons even that would have been high. Players can always outperform projections substantially (see 2014 Kluber).
Kluber is a great case because nothing told us this was coming. Steamer and ZIPS were a bit more bullish on Kluber than I was – I had him projected for a 3.3 WAR, I believe he was projected to be under 2.5 here; I am not looking at the numbers right now though.
Trout’s projections started out horrible – for everyone – but that is more of a product of the unknown. No one was going to project the greatest start for a player in MLB history because there’s no history to support that reality.
Gio Gonzalez’s name is misspelled in the body of the Nationals article
None tabled of the player links work.
Your sentence sense cannot be made.
The Quintana projection cracks me up every time. Fangraphs is projecting a player who is entering his prime, and has gotten better every single year of his career, to suddenly go back to being as effective and good as he was his first year in the league. The best in fangraphs doesn’t even point out nor admit that the Quintana projection is obviously low.
Both Steamer and ZiPS give Quintana a mediocre projection. I’m sure the makers of those systems are either brainless idiots who don’t understand such simple concepts that come so easily to you, or they both just personally hate Quintana. Or maybe fangraphs posted the objective projections, but saw Quintana’s, didn’t like it, and then cut it in half.
You and Shankbone are both insufferable.
“Scientists discovered today they may have miscalculated…”
HA! Look at how wrong those scientists are! What a joke. Remember that time scientists thought Newton’s laws accurately described the force of gravity? Hilarious!
So now you’re saying that baseball projectors are now comparable to Newton? Wow. Its no wonder why Bill James doesn’t like projections (or fWAR, for that matter).
That’s a very selective way to interpret that comment.
I’m saying that people who laugh at attempts to create useful models when those models show themselves to be imperfect are not contributing anything.
That isn’t even close to what he said. Holy shit some of you are stupid.
You are clueless. I never said it was personal; it is merely lazy modeling. When certain players consistently outperform their projection by a considerable amount, a normal modeler/programmer would go back and look to see if there is an inefficiency somewhere. Successful pitch to contact pitchers – Quintana is becoming less of this each year though – are consistently underprojected by Steamer/ZIPS and the likes; dare I say Fangraphs. I would post my data to support this claim, but why would I do that for free? I am not a lazy modeler, and I actually make countless adjustments each season, and don’t judge every single player based on the exact same numbers. The fact remains that pitchers who induce weak contact but don’t generate a lot of swings and misses are under projected every year; keeping a batter off balance is an ability, it’s just one that doesn’t fit under Steamer/Zips and the likes because the majority of pitchers that don’t miss bats, don’t generate weak contact.
“I would post my data to support this claim, but why would I do that for free?”
Yes, please, continue to laugh at our silly projections while having the data to improve them, and purposely not sharing it.
You come to this site only to brag about how your own personal projections are so much better. If that’s true, you have no reason to come here. Go away.
“I would post data to support this claim, but why would I do it for free” = “I am talking out of my ass and can’t imagine anyone would dare ask for some proof of validity to my ridiculous diatribe! Because, frankly, I have no proof whatsoever, which is quite painfully obvious.”
No, I come here to discuss metrics with others interested; I also come here to interact with people that I share information with off site. Whether you find that dickish or not is entirely up to you; I know it would be pretty stupid of me to share information like that for no reason. 8 years ago I probably post it just to prove a point in an internet argument; my wallet is more important than my ego these days though.
Jason B, you got me. For another example of a pitcher that I would almost guarantee, barring injury of course, outperforms his projection look no further than Henderson Alvarez; who falls into the exact same category despite being a hard thrower. He does not generate a lot of swing and misses (he and Buehrle both have contact rates right around 85%), but he generates a ton of weak contact. I could list 15 pitchers that will outperform their projection based on the exact same criteria I used/adjusted when projected Buerhle. Of those 15, I would be willing to bet that 12 of them beat projection. If that doesn’t show an inefficiency (if I am correct again), I don’t know what would. As I said though, no need for me to stroke the ego; rather pad the wallet.
Look for me on draftkings or fanduel this upcoming season. If you’re interested, I’ll post my name and you can follow along and become a believer for yourself.
FWIW I have Henderson Alvarez at a 2.3 WAR for this upcoming season.
“I’ll post my name and you can follow along and become a believer for yourself.”
Whether or not I believe your projections is so, so far from the point of this.
The science comparison is hilarious. For something to become a theory in science, it needs to survive millions upon billions of trials and it needs to be 100% correct every single time. There is no margin for error that allows the theory to remain accurate.
Fangraphs projection systems (ZIPS/STEAMER etc) end of years numbers fit on a bell curve, with countless outliers and aberrations.
“For something to become a theory in science, it needs to survive millions upon billions of trials and it needs to be 100% correct every single time.”
This is not even close to being true.
How is that not true? Give me one example of a scientific theory that allows room for a “but.”
Theory of relativity does not work if a single aspect of it is disproved; if space and time are proven to not be relative to each other, end. If the speed of light is not the same to all observers, then the theory of relativity is disproved. The basis of the theory may remain the same, but the theory itself would take on an entirely new basis.
You’re confusing science with trans-science.
For someone trying to make a point related to the importance of accuracy and precision in predicting outcomes, your “millions upon billions” comment makes you sound like a hippocrate. You just went from annoying internet fraud to raving jackass.
For the Angels, Drew Rucinski may be the front-runner for the 5th spot in the rotation. He starts today.
OK, I see Rucinski is projected to get 15 innings, none as a starter, on the Depth Charts page. That’s identical to his steamer projection. ZIPS has him making 18 starts. Nevertheless, there have been some rumblings he may be in the running for that 5th spot, and I’m pretty sure he’ll be in line as depth.
I look forward to watching the M’s beat this projection.
The Cubs are a tad too high
One of the biggest Giants fans here, but I think their rotation is ranked awfully high. I expect everyone except Bumgarner to be below 1 WAR, with a few in negative territory.
this. I’m a giants fan too, and I am worried about everyone except bumgarner and petit. I think timmy is going to suck and get moved to the pen. vogelsong has not been good for 3 yrs now, and hudson/peavy are washed up.
And of course the O’s are 27th despite being six deep and returning all starters from last season.
Didn’t you know that FanGraphs HATES the O’s? It’s not at all that they have mediocre starters, it’s a site-wide conspiracy, obviously.
Luckily you are the first to have noticed/commented on it!
And where did those exact same starters, the mediocre ones, end up when you look at RA in 2014?
Isn’t RS and RA where the meat is?
RS and RA were awesome and meaty last year, the Orioles enjoyed them. Problem is RS and RA in 2014 are not as predictive as different data (the inputs for ZiPS and Steamer) from last year.
Interestingly enough, I don’t think the 2015 rankings were composed based on an ordinal ranking of runs allowed in 2014.
There is no way that the Yankees rotation should be ranked 9th, and here’s why:
Masahiro Tanaka’s elbow still remains unrepaired, and platelet-rich plasma (PRP) therapy has had little success in fully repairing torn ligaments. The lack of PRP’s efficacy has been demonstrated and is subject of continuous debate among physicians. Tanaka will go under the knife at some point this season; it would be shocking if he lasts past the all-star break. There is no way he gets to 150 innings, and certainly not 2+ WAR.
CC Sabathia is projected for over 2 WAR, yet his velocity has declined substantially in each of the past 3 seasons, and his FIP has rocketed skyward in each of the last 3 seasons. He had a 0.1 WAR in 8 starts last year, and will be worth <1 WAR in 2015, and will likely be facing possible retirement halfway through 2016.
Pineda. The last time he pitched more than 76 innings in the majors was 4 years ago. He could be really awesome, or he could be really bad. Either way, we cannot say with great certainty how well or how bad he will be in 2015, based on his unpredictable on-the-field antics (ahem, pine tar), and the limited sample size of his recent performance.
The other fellas in the rotation are not exemplary enough to overcome the cited drawbacks of the Yanks top 3.
With the age of the core contributors in the lineup, and virtually no good defensive players outside of Headley (Gregorius had a -1.4 defensive WAR in half a season last year, I don't buy the 6 def-WAR in 2015 projection), this is going to be a very, very, bad Yankees team. Possibly one of the worst Yankee teams since the early 90s. Their bullpen has potential to be really good, but that's the lone bright spot on this team.
Ranking the rotation 9th, ahead of the Tigers, White Sox, Giants, and Angels seems foolish, given that the WAR projections of Sabathia and Tanaka are excessively optimistic.
I see a rotation with an 8.0 to 9.0 WAR total, and that would put them somewhere below the Marlins rotation.
The SFG won the World Series fair and square, but they did not do it on the strength of their starting pitchers. They did it on the strength of one starting pitcher. Collectively, the other starters performed as you would expect a 20th ranked staff to perform. Unless there’s been a change in the rules, MadBum will only be starting every fifth game during the regular season.
No rule change would be required for MadBum to start more frequently.
They would need to allow PEDs.
I guess WAR should take into account platelet-rich plasma therapy in future versions.
There is a precedent for pitchers being successful, for even years at a time, without having TJ. Given that every doctor that looked at Tanaka’s elbow did not recommend surgery, he would be in as good a position (and it is not a great one) as a pitcher could be to join that (admittedly small) group. Tanaka was able to pitch at the end of last year, he hasn’t pitched like an injured pitcher this spring, he is a boom or bust candidate. He could go at any minute, but until then he’ll likely be wonderful. Given that numerous medical professionals looked at his elbow and said, “give this a shot,” Tanaka is probably a good bet to get 2+ WAR, he certainly won’t need 150 innings to do it.
C.C. probably cannot finish a season, but he is throwing in the 90s again, so he should be able to be effective until the knee surrenders. Look at his player page, his peripherals have never really fallen below league average.
Pineda: Uncertainty in projecting a projecting a young pitcher? Oh my! Should they not project him because of high variability? If he’s healthy he could be the best pitcher in the AL East, IF. Nuff said.
Ivan Nova is similarly hard to project, but with far less upside.
Eovaldi could be massively underrated here, given his new split-change. If they can do to him what they did to McCarthy, they have a great young pitcher how can take on 200ish innings.
The profile of Warren on this very site was remarkably favorable, but indicates that his projection is hardly rosy.
Anyone could do what you did to 25 of these staffs. When the absolute worst case scenario nets your staff 9 WAR, that’s why the staff is ranked so highly.
Yeah, the Yankees’ rotation is incredibly high variance this year.
I won’t be surprised if Tanaka puts up 6 WAR, Pineda 5, Eovaldi 4, CC 3 and the fifth starters 2, and they coast on strong starting pitching, and an insane bullpen, to the playoffs.
I also won’t be surprised if none of them crack 2 WAR, due to injuries and ineffectiveness, and they win 82 games (or 70, but I’m pretty sure after the last few years that the baseball gods won’t let them finish below .500)
Indians staff is way over-rated. Kluber is in for a big regression. Kluber’s previous play gave no indication that he was going to have a season like last year. Let alone be an ace. Did you know he has only thrown 450 IP in the majors and it’s not like he is young He’ll already be 29 this season. He is only 8 months younger that David Price.
And their other ace… Carrasco is already 28. Carrasco’s last 10 starts of 2014 were amazing (1.30 ERA and 0.812 WHIP). His first 44 starts equaled a 5.66 ERA and 1.561 WHIP. That includes a horrendous 6 starts to at the beginning of 2014. They are taking the last 10 starts and saying that he has figured it out. It’s more likely to represent the inconsistency of a mediocre major league pitcher. I bet he is a lot closer to his career starting numbers of 4.70 ERA and a 1.4 WHIP.
Even if we were to discount any mechanical “change” in what Kluber or Carrasco does in the act of pitching… and Carracso certainly fits that bill with Tommy John and slider usage… the sheer probability of his final 10 starts being a product of variance (luck?) are astronomical.
Let’s be kind and say that a middling pitcher has a 50-50 chance of having a quality start outing. Last year in the MLB, this number was 54% for all starters. Carrasco had 9 quality starts out of 10 after his return to the rotation… and in the one that didn’t qualify (Sept. 12th @ DET), he went 6 1/3 innings and gave up 4 earned runs… one of them facilitated by his own throwing error.
What are the odds of flipping a coin 10 times and getting 9 heads?
Not good. Show me a so-called “mediocre pitcher” with 9 QS in 10 consecutive starts, and I might rethink your post’s classification as laughable.
He also had 78 strikeouts in 69 innings during that stretch.
I’m willing to bet his ‘regression’ still makes him a top 30 starter.
First off, using the stat QS to describe what Carrasco did his last 10 starts is really under selling how good he was for that period. He had a 1.30 ERA and a 0.812 WHIP. Ball players do get hot and they can lose it as quickly as the got it, especially if they are mediocre.
As for your 9 out of 10 QS? I only had to research one pitcher from last season. I am sure I could find a few more. Career mediocre pitcher Aaron Harang. The last 2 start from 2013 and first 8 from 2014 9 out of 10 QS’s and then later on in 2014, he when 10 out of 11.
Oddly enough, here are 37 yr old Harang’s career #’s vs 28 yr old Carrasco
4.21 – 4.30 ERA
1.357 – 1.335 WHIP
7.3 – 7.3 SO/9
2.8 – 2.7 BB/9
BTW… Officially Carrasco only had 7 QS’s in his last 10 starts. To qualify for this stat you have to go at least 6 innings. Even though the QS stat has some pretty week requirements (4.50 ERA or less), usually the best pitchers have the most and highest percentage of QS’s. Unfortunately, despite his great finish to the season 7 and 50% put him 142nd and 91st respectively. Even if you were to take his last 10 starts and go with 70%. That still only puts him as high as 28th percentage wise.
Again, I am not saying he hasn’t figured everything out and that he won’t be a quality #2, I am just saying he still has a long way to go to prove it. 10 starts is a SSS to base even 2.6 wins considering his highest WAR prior to last season was 0.7.
I’m disappointed by Miles Ray focusing on a narrative that’s 5+ years old now. Come on – Sandy’s been astute ever since he took over and yet FanGraphs continues to frame the organization as “LOLMets”. I’m disappoint.
Thanks for this. But! Your initial explanation of the Ray for Fister trade is a little muddled. Lombardozzi, not Greene.
Just pickin.
Rational Fan is not being an Internet jackass. Y’all got personal defending the systems like they’re your children. What I took from him was simply pointing out the fact that these systems consistently under-project a certain type of guy. They sure do a pretty good job, but there’s work that remains to be done.
Instead of simply writing it off as “these systems aren’t perfect, relax everybody”, he’s pointing out a specific area that actually shows consistent discrepancy. Patterned discrepancy should excite us and interest us. I’m looking forward to any new intel that somehow values speed-changing wizards.
It’s not at all a “hey relax this projection system is perfect DON’T TOUCH IT”
or even a “hey relax, all projection systems are pretty crappy so go easy”,
it’s more of a “let’s take your suggested tweak and apply it evenly across the entire universe of players (rather than a hand-picked example or two) and see if the accuracy of the system increases or decreases.”
thanks, Howie Porker
You’re welcome, Howie Porker. I mean, Rational Fan.
Damn, Fangraphs just got sealioned …