Division Preview: AL East
And now the final division preview, just in time for Opening Day. If you missed them, here are the first five:
NL West
AL West
NL Central
AL Central
NL East
Now, wrapping things up with the AL East.
The Projected Standings
| Team | Wins | Losses | Division | Wild Card | World Series |
|---|---|---|---|---|---|
| Red Sox | 87 | 75 | 45% | 18% | 8% |
| Blue Jays | 83 | 79 | 19% | 17% | 3% |
| Yankees | 83 | 79 | 19% | 16% | 3% |
| Rays | 80 | 82 | 11% | 12% | 2% |
| Orioles | 79 | 83 | 7% | 9% | 1% |
The only division in baseball where all five teams have a legitimate shot at winning; the projected spread between first and last place in the AL East is smaller than the gap between first and second place in the NL East. The forecasts have a favorite, but this division is wide open, and nearly any order of finish could be reasonable. On to the teams themselves.
Boston Red Sox
Maybe the Giants and Red Sox have some kind of weird co-deal with the devil, with San Francisco getting the even years and the Red Sox getting the odds? Boston is projected to go worst-to-first for the second consecutive opportunity, as the projections buy into the offensive makeover and a solid bullpen making up for a mediocre rotation. The team’s depth of position players has Rusney Castillo starting the year in the minors, and while that won’t help them on the field in April, the many backup plans the team have in place should keep them from playing too many scrubs this season. While the pre-season focus is usually on a team’s ceiling, avoiding a low floor can often be just as important, and the Red Sox have one of the highest position player floors in baseball.
This team’s fate is likely to be decided on the mound, however. Boston’s front office has made a bet that you don’t need an ace to win in the regular season, especially if you have solid mid-rotation starters from front-to-back. I think they’re right, and while having a dominant starter is important for the postseason, I don’t think you really need one in April, so long as the rest of your roster is strong enough to compensate. Come August, if the Sox are contenders, I’d speculate that Rick Porcello won’t be the team’s #1 starter heading into the stretch-run, but they’ll take their team figuring out what they have before they commit to paying the market price for a #1 starter.
This Red Sox team still isn’t quite a finished product, and I wouldn’t bet on this roster to do that well in October, but Boston has six months to realign things for the postseason tournament, and the current batch of players should be strong enough to keep them in the race for a playoff berth while they figure out what they have.
Toronto Blue Jays
The early winter’s most active team — only later overtaken by A.J. Preller after an apparent overdose of Red Bull — the Blue Jays are another classic stars-and-scrubs team that will likely go as far as their role players can take them. The top-end of the roster is tremendous, with Josh Donaldson joining Jose Bautista and Edwin Encarnacion in a line-up may cause many lefties to come down with flu-like symptoms the night before their starts. The Jays have stacked some very strong hitters together, but despite the star power, their line-up also features Justin Smoak, Dalton Pompey, and Devon Travis. These guys all project as slightly-above replacement, so while they aren’t complete black holes, the bottom of Toronto’s order isn’t as strong as you’d expect based on the guys at the top.
And then there’s the pitching. Without Marcus Stroman, the rotation is now leaning very heavily on young hurlers Daniel Norris and Aaron Sanchez, both of whom may struggle to throw strikes consistently in the majors. Toss in 20-year-olds Roberto Osuna and Miguel Castro in the bullpen, and this pitching staff is one enormous bet on stuff over experience. The projections look at these four young hurlers and see a bunch of guys who are probably not ready for the roles they are being asked to fill, but the reality is that if the forecasts are right about these kids, they probably won’t stay in these roles for very long; the upside here is that the Blue Jays are projected to win 83 games despite getting very little from two rotation spots and nothing from two bullpen spots.
If any of the kids are more big-league ready than the projections think, the Jays aren’t far behind the Red Sox, and even if the kids do struggle, it seems likely that Toronto won’t wait too terribly long to swap them out for big league upgrades. This is a team built to win now taking a gamble on some young guys, but they aren’t committed to these gambles for the entire season, and if they don’t work early, they’ll try something else. That’s why I think the forecasts are probably a win or two light here, and I wouldn’t be at all surprised to see Toronto in the playoffs when the season ends. But to get there, they’re probably going to have to have been right about at least one or two of these youngsters.
New York Yankees
I think I’ll take the under here. This .500ish forecast includes a +3 WAR performance from Masahiro Tanaka, who is already publicly stating that he’s expecting to throw in the high-80s this year. Maybe Tanaka will be one of the first guys to figure out how to pitch through an elbow problem successfully, but I don’t think he’ll throw 150 innings for New York this year, and I don’t think he’ll be as good as the projections think when he is on the mound. And if you take him out of the picture for an extended period of time, this rotation does not look good enough to make up for a pretty mediocre offense.
I think the Yankees top four position players — Jacoby Ellsbury, Chase Headley, Brett Gardner, and Brian McCann — are probably underrated, and they’re all solid pieces who can easily play on a contender. But you don’t really want any of these guys to be your best position player, and there’s a huge gap between these guys and whoever you think is the fifth best everyday player on the Yankees. Maybe Carlos Beltran has a big comeback year. Maybe Didi Gregorius figures out how to hit lefties and becomes a star at shortstop. Maybe A-Rod hits 40 home runs and becomes a media darling. Anything is possible — okay, not that last one — but these are long-shots at best, and assuming several of these bets fail, there isn’t enough at the top of the roster to carry the Yankees, especially without Tanaka.
If Nathan Eovaldi takes a huge step forward while Michael Pineda and Ivan Nova prove healthy, a Wild Card run isn’t out of the question. But this team is a far stretch from the days of the Yankees having star players around the diamond; you have to squint to even find one, really, and there aren’t enough above-average role players to make up for the lack of a true franchise player anywhere on the field.
Tampa Bay Rays
In a world where Alex Cobb and Drew Smyly can both give the Rays 200 innings, I’d find the Rays case as a Wild Card contender pretty interesting. At that point, they’d have one of the strongest pitching-and-defense combinations in baseball, and maybe just enough hitting to be the 2015 version of the Kansas City Royals. But Cobb and Smyly are both starting the year on the disabled list, which will force the team to try to win low-scoring games with Nate Karns and Erasmo Ramirez on the mound. That’s a tougher sell, and unless the team’s magical pitcher-fixing abilities are about to turn Jake Odorizzi into an #1 starter, the April version of the Rays could actually be pretty terrible.
And if the hole becomes sizable enough early in the season, then a team in the Rays position will have to start thinking about selling, which likely starts shipping all the good relievers out of town and leads to a second half where the team is simply evaluating talent rather than trying to win. That’s the danger of starting slowly for a team that can’t afford to wait too long to make their in-season buy/sell decisions, and losing Cobb and Smyly early in the year make it more likely that the team ends up as sellers, pushing them down below their pre-season win total from the forecasts.
That isn’t the only way this season goes, of course. Smyly and Cobb could come back at 100%, the outfield defense could be amazing enough to make even inferior pitchers look great, and perhaps the numerous job-shares end up being more effective than one would expect based on the individual parts. There is a chance for this team to put it together and make a run this year, even while the Rays mostly look towards the future.
But that would require the team to turn a lot of lemons into lemonade. And as a wise man once said, if you don’t also have sugar, your lemonade is going to suck.
Baltimore Orioles
The Orioles have spent the last few years making the pre-season projections look silly. They’ve even made the in-season models look dumb, as those of us who have believe in the predictive power of things like BaseRuns keep pronouncing that regression is coming, only it never really has. Despite all kinds of statistical markers downplaying their strengths, the Orioles have won 274 games over the last three years, fourth most in baseball.
So I understand completely if Orioles fans don’t particularly care that the algorithms still don’t buy in, and are projecting yet another sub-.500 season. I imagine these tools have little credibility in Baltimore at the moment, and that won’t change until the Orioles actually play like the projections expect one of these years. And it is possible that the models are broken and the Orioles are the franchise who have figured out how to exploit those holes. But it’s also possible that it’s just been how baseball works, and the Orioles outstanding 2012-2014 run might not be the sign of a team that has a sustainable ability to beat the forecasts.
As with the last few years, the Orioles are again betting that their defense and Buck Showalter’s magnificent bullpen usage can make up for a weak starting rotation, and that they’ll be able to get enough offense out of undervalued pieces like Travis Snider and Alejandro De Aza to compensate for the bats that left via free agency. If Chris Davis has a big rebound season — he doesn’t even have to be the 2013 version to be quite valuable — and the O’s can get J.J. Hardy and Matt Wieters on the field for most of the season, the Orioles certainly have a chance to make another run.
But they’re going to need Buck Showalter’s magic to actually be a thing. If they win 90 games again with this roster, it will likely be time to admit that the Orioles have something figured out.
Dave is the Managing Editor of FanGraphs.
why are pedroia, napoli, hanley, and ortiz all projected to stay exact same as they were last year, as opposed to the dramatic declines that they’re all clearly experiencing?? and the projections are just supposed to make us believe that bogaerts will magically become a 2.5 war player after basically being replacement level last year?
i’m sorry but these projections just don’t pass the sanity test. i’d much rather have a young guy with upside like smoak at 1b then bet on an again napoli to somehow reverse his clear downward spiral. ditto for saunders in lf as opposed to hanley. and don’t even get me started at c. projecting blake swihart (a poor man’s jp arencibia) for 0.4 war before he’s taken a major league at bat is ludicrous
and just a friendly reminder:
2014 war of current red sox starters: 6.6
2015 projected war of those starters: 6.9
how the projections systems somehow got that all of the red sox starters would take a huge leap from their 2014 performances is beyond me
You can certainly be skeptical, and you may even be right. But the flipside of the coin says: Mike Napoli has a healthy finger and is sleep apnea free, Dustin Pedroia has a healthy wrist and thumb, Ortiz had a bad luck on balls in play last year, and Bogaerts should improve somewhat (maybe not all the way to 2.5 but still somewhat). That’s how the projections must see it, and I am actually inclined to agree.
All the things you’re inclined to agree with appear to be somewhat sunny projections/interpretations. Given the injury problems those players have had, their increasing age (w/the exception of Xander), and what is arguably some age related decline that was already going on a safer bet would probably be somewhere in the middle – which would suggest at least the projections for those four players are a little optimistic right now.
Pedroia: regression (upwards) to the mean vs. age-related decline.
Napoli: health vs. age-related decline.
Hanley: health vs. age-related decline.
Ortiz: steroids.
Bogaerts: everdiso is nuts.
Smoak over Napoli? Everdiso is nuts.
Rotation: I would expect slight improvement from Porcello and Miley based on improved defense alone. Buchholz may be in line for a huge improvement if he can stay healthy. Think that the projections are overrating Kelly and probably being overly optimistic on the health of Masterson, but the Sox have enough behind him (Wright, Owens, Rodriguez, Workman, Barnes, etc.) and enough in trade ammunition that I think the Sox on the net are better situated w/ SPs than last year. 6.9 vs. 6.6 sounds about right, except that I think there’s considerably more upside this year.
Let it noted that on this day, April 6, 2015, noted fangraphs commenter everdiso found irrefutable evidence of pro-red sox bias in mathematically-derived projections through use of several poignant anecdotes. Thanks for your contributions to the world’s understanding of statistics, projection, and prediction of future baseball performance. We are forever in your debt.
Say there, lad! You’ve the wrong appellation on, you have. Sergeant is a perfectly noble profession. Nick Fury did it, as did York. On th’ other hand…there is Bilko and Gomer Pyle.
Carry on, general.
Not Gomer Pyle. Sergeant Carter. ?
Let it also be noted that “the projections” are the gospel and they should be taken as the absolute word of GOD. I jest but this is what the response seems like to everyone that has some question about anything found in any computerized model. And to anyone that insists on “showing your work” or any other trite remark, these are opinions. And we only have to show how horribly wrong the models have consistently been with a team like the O’s to shut down that argument.
And of course the O’s are once again laughably underrated. I just wonder if the models themselves will be re-examined if the O’s blow the projection away yet again.
Maybe you don’t understand WAR? 6.9 WAR is essentially statistically the same as 6.6. It is less than half a win difference and far from a “huge leap.”
I’m not looking to argue or take sides, but if the numbers are correct (6.6 & 6.9), then perhaps you should revisit, as a friendly reminder, what the definition of WAR is.
He isn’t interested in statistics or data except to the degree than can be used to validate his deep, Bellevue-caliber fixation – not even with his favorite team, one he roots as part of a healthy, happy existence – but with a team he roots against so much that he spends all his idle time yelling at people as they walk by.
In short, he is not a person you can talk to. You can only squint, nod assuringly, or laugh and, realizing the tirade isn’t going to stop any time soon, drop your spare change in his RED SOX CONSPIRACY-scribbled cup and continue on your way.
Justin Smoak is a young guy with upside? Did I fall into my time machine to 2011?
*yankee fan cries over Seattle choosing Smoak package for Cliff Lee* That was a shift of 2 pennants right there.
I read this article after it was posted and then said to myself, “I bet if I refresh this page, there’ll be an everdiso rant about the Red Sox all being overrated.” Nailed it!
“bogaerts will magically become a 2.5 war player after basically being replacement level last year” vs. “young guy with upside like smoak at 1b”. Never change everdiso. Never change.
The Smoak over Napoli nonsense isn’t worth a damn but one doesn’t have to be a hater to think that projection systems might be overrating the red sox.
Certainly not. But that doesn’t mean he isn’t a “hater” and hasn’t proven it over and over again.
A substance-free critique is not worth a damn just because a critique with substance that reaches the same conclusion also exists.
First part: “Why aren’t these projections acknowledging the age of these older players! Age matters!”
Second part: “Why are these projections acknowledging the age of this very-young top prospect? Age doesn’t matter!”
You really just compared Blake Swihart to J.P Arencibia, favoured Smoak over Napoli, said that Hanley is worse than Saunders in one comment? Wow.
another effective preemptive strike by the red sox fan everdiso troll replere with intentionally bad stats. man you work hard at this. i can’t even keep up with you.
now for a real everdiso comment:
Do we realize that every single red sock starting pitcher and position player is projected for either identical or better war rate pace than they had least year….except for mookie?
on top of that, the collective bench and bullpen are projected to near identical paces as they put up last year, too.
that’s right – the only startung sox player to be projected to decline in performance is the guy with the negligible sample size who is still projected to put up the most war of any full first year player.
The projections don’t see one significant decline on the entire roster, despite the majority of it being on the wrong side of 30, with significant health concerns too. That is, at the very least, extremely unlikely.
now, no matter how hard the pseudo everdiso works to divert attention, this is still truly amazing stuff from the projections that shouldn’t be ignored by any objective observer. including the pseudo everdiso.
You realize projections aren’t just based on one year right? Projections have been proven to be superior to just using last years stats. Many Red Sox players last year underperformed compared to career norms and many suffered bad BABIP luck. This means a huge majority of them are going to be projected to be better than they were last year.
Everdiso is annoying enough without having to have clearly fake verions on every single article. Let him make his own ludicrous rants.
Not sure why replying, but comparing Swihart to Arencibia is the most absurd thing ever slipped into a trolling comment. If Swihart busts, it won’t be because of a 30% K rate or because he’s a terrible defensive catcher who shouldn’t be catching.
Not sure why replying, but comparing Swihart to Arencibia is the most absurd thing ever slipped into a trolling comment. If Swihart busts, it won’t be because of a 30% K rate or 3% BB rate or because he’s a terrible defensive catcher who shouldn’t be catching.
“But they’re going to need Buck Showalter’s magic to actually be a thing. If they win 90 games again with this roster, it will likely be time to admit that the Orioles have something figured out.”
Bookmarking this for posterity. I’ll believe Fangraphs admitting the Orioles are good when I see it. According to Fangraphs, even the 2014 version was supposedly worse than the 2014 Rays, despite winning 19 more games and scoring >100 more net runs.
Oh my god, give it up. Do you REALLY think that the FanGraphs authors have a personal vendetta against the Orioles? Do you ACTUALLY think they give a shit?
The team has succeeded every year by doing things that all of baseball history says are unsustainable. That’s why they’re projected low. Who cares? Obviously and fortunately, baseball is a sport where you can defy the odds. Just enjoy it and stfu about this anti-Orioles BS. This whole site is ABOUT objectivity. No one cares that your fan-feelings are hurt.
The everybody hates us mentality is a little tired but it’s silly to think that fangraphs is some paragon of objectivity. This site is inherently about trying to interpret statistics to make sense of them and then trying to project team quality. The writers here admit WAR doesn’t capture everything and that their projections are based upon their interpretations of empirical evidence. That interpretation of the evidence is absolutely biased by whatever priors are held by forecasters – that’s true of any projection system. Hell Dave even admits they might be missing something in these forecasts (and also that natural variability may explain the O’s).
As much as I’m annoyed by friends who think everybody just hates on their teams, I have more patience for fans of those teams that have been consistently overperforming projections.
It’s a common problem that’s hardly limited to baseball. Avery Pennarun, a Google Engineer, wrote an excellent article entitled “The Curse of Smart People” that discusses this. Bing it. WAR is an excellent tool, but one’s faith in this measurement can cause one to reject evidence to its deficiencies. There probably is something WAR is missing with regards to the Orioles. However, if we don’t what it is and the Orioles front office doesn’t know what it is, it will be difficult to copy it or, indeed, sustain it. I think the Orioles will again exceed there expectations because they haven’t changed the team make up much. I don’t know what it is about the make up that causes them to exceed expectations, but as it hasn’t changed, I have no reason to believe this as yet intangible “skill” has changed.
So 3 years of being wrong isn’t enough to admit it?
Look, I don’t believe FanGraphs authors have a personal vendetta against the Orioles. I just don’t understand how people who analyze data for a living can keep ignoring it.
Hey, I’m an O’s fan from way back, say from 1964 or so. I was at the game in 1967 when Steve Barber/Stu Miller threw a no-hitter and lost, which led to my irrational, life-long, dislike for Mark Belanger, who, as a rookie playing second base, dropped the feed from Aparicio and lost the game. I was at the third game of the 1970 World Series where Dave McNally hit a grand slam and Brooks made some amazing plays. I’ve seen a lot.
It’s great when the O’s win, but over the years they have had their share of ups and downs. I don’t believe FanGraphs is ignoring any data. They’ve already demonstrated that with normal statistical variance, you would expect outliers to occur that would result in some teams overachieving in multiple years. If this happens to the same team enough times, to the extent that the odds of this occurring are extremely low, then it’s time to look and see if other factors are at play. This hasn’t happened yet, but as Dave says above, after this year, if it continues, it may be time to take another look.
BTW, planning on going to see the Wednesday’s game with the Rays in St. Pete. (I’m in Orlando)
Ignoring it? That’s rich. They have addressed it head on, repeatedly.
http://www.fangraphs.com/blogs/the-orioles-and-accepting-random-variation/
You haven’t ignored articles like this, have you?
Don’t link to really one of the worst statistical analysis’s fangraphs has ever produced. It’s like it was a postulated by a freshman community college stats 101 student. I could get into it, but some of the comments on the article were spot on. I know I am kind of insulting Mr. Cameron here, but his reputation took a hit in my eyes after this. However, he is a writer first and statistician second.
I think the models are fine, but I expect advancements in forecasting to become even better. This is a new science, and one day we will probably all laugh at what we think is important or unimportant today.
There have been 90 team seasons over the last three years. Baltimore’s three beating the odds isn’t a significant sample in the grand scheme of things.
Every team/season is an outlier, viewed through that lens.
Unless it’s the Red Sox and they’re good. Not that FanGraphs plays favorites.
And no one mentions the team that stopped us- KC. The numbers don’t love those guys either. As Earl was moved to say as his career evoked ” Pitching , Defense, and three run homers, in that order!” Defense is still not quantified in projection models. It will get there, and when it does the world will be a better place.
I think the Royals will be lucky to finish above .500 this year.
The O’s will fall apart, maybe not this year but they will. So will the Red Sox, Dodgers, and Nationals. Players will experience age related declines and the bullpen luck dragon could change sides. When this happens to Baltimore many here will declare that the Orioles are regressing to the mean and their past success was all luck. Personally, I think that any team that sends out 5 above average starters with a dominant bullpen and an above average offense will win most of their games, but what do I know? I guess it could be all luck that all these starters are above average. It’s pretty clear that the current Orioles core that really came into being the second half of 2012 is a true talent 85-95 win team. If all goes right like it did last year, they’ll be one of the best teams in the game again. If it all goes wrong like say in 2013 minus Davis’s surge, they’ll win fewer than 85.
I suspect they will fall apart next year when virtually their entire roster hits free agency.
I freely admit that I am a die hard Orioles fan who’s love for the Orioles readily eclipses most things in my life between march and September (and very recently a few Octobers). I read the articles because I am a baseball fan and think they are very interesting. I absolutely love when my team outperforms expectations, models, and columnists opinions. I really don’t understand why some O’s fans are so worried about what a model they don’t believe in says about our team. In any argument we have the trump card, because we win, and baseball the only stat that ultimately ever matters is Wins.
I’d say it’s more like the O’s need to get more offence out of the return of Machado and (sigh) Wieters to offset what they lost in Cruz and Kake. It’s not complicated to see how the O’s win, it’s just tough to quantify. They send out 5 pitchers who will go five or six innings and allow 3 or fewer runs. The bullpen then shuts down the other team and the offense manages to get more than 3 runs. Three straight years of basically the same pitchers doing basically the same thing is a pretty good sample. If they Orioles crash and burn it will because their bullpen is unable to pick up the slack. (or injuries)
I’m an Orioles fan. Man, other fans, reasonable writers, of which every Fangraphs person is, do not hold weird vendettas or something.
The O’s are a good team. No one thinks they are a great team, this exercise bears that out.
Certainly if everything breaks right for the O’s, they’re a playoff team. However, it’s baseball, when does everything break right.
I personally think the O’s have better chances than this, but it’s not like the O’s are the Nats in the NL East — every team in teh AL east is solid…
If the Sox Red were starting five #3’s they’d win going away, but they look like #4’s at best. The loss of Vasquez becomes enormous.
Once Victorino hits his way out of the lineup they will have only one dead spot. But they are still going to need to acquire starting pitching.
87 won’t win the division. No idea who will step up, but someone always does.
Yeah, I’m not buying the idea that Boston is running out a rotation of “solid mid-rotation starters”. It’s Porcello, maybe Miley, and then a bunch of guys who could be really bad. Like 5+ ERA, dig-team-into-hole-the-offense-can’t-get-out-of bad. You absolutely do not need an “ace” to survive the regular season, but let’s not get ahead of ourselves calling this rotation solid from front to back.
I like him going forward, but Bogarts could be a dead spot as well offensively. Actually, the current Red Sox team design is very similar to Baltimore’s. Five (hopefully) above average starters, good defense, a deep and solid bullpen, a lot of depth and very few dead spots on offense.
I think the interest here, from the perspective of a Baltimore fan, is that we do think that our team has something figured out that allows us to consistently beat the projections found here and elsewhere. Just like Angels fans probably felt like they had something figured out last decade.
Analysts here have no particular prejudice against the Orioles, but it would be reasonable to assume that they think their predictive models are better than others (and so would defend them with non-analytic gusto), because they’re the ones they’ve designed or chosen to use. And, when it comes to the O’s, they’re off. Most of the human predictors, here and elsewhere, think the Orioles will win more than 80 games, despite the statistical models. And the people are probably right.
The interest and difficulty at the end of the year will be determining how analysts here and elsewhere adjust their models (as presumably happens every year) to account for another better than expected season from the O’s. A couple articles that appeared toward the end of last season seemed to find a stronger than expected correlation between minimizing the awful and teams that outperformed expectations. The O’s and Royals squaring off last year in the ALCS seemed to suggest that team defense and bullpens remained undervalued factors (maybe especially in the American League, where pitchers can’t save pitches against other pitchers–it seems like bullpens could be more crucial in the American League).
Anyway, once is a fluke, twice is a coincidence, and three times is a trend, so it’s not unfair to be a bit perturbed that the cutting edge has been so slow to adjust. Maybe this site just needs more O’s fans as regular contributors.
The thing is that three times may very well be a coincidence as well…
You’re going into this already having made up your mind that the O’s know something everyone else doesn’t and that it will lead them to a successful 2015 season. Unless you at least have a theory on what specifically they know, it’s hard to see that as unbiased.
A huge part of the O’s success in recent years has simply been guys outplaying expectations. Pearce and Cruz combined for 8.6 fWAR in 2014. Chris Davis jumped from 2.1 fWAR in 2012 to 7.1 in 2013. Maybe the Orioles are better able to identify talent than other teams, but they may also have had some exceptional luck in back-to-back years. And even if it was purely skill that led to the Davis, Pearce, and Cruz acquisitions, are they going to find that next hidden gen again 2015? It’s hard to imagine they’ll hit on one or more every single season.
There may be something the Orioles know the fangraphs models don’t, but it may also just be a case of a general model not being able to predict extreme events.
I’ve got three theories in there (weighing negative contributors more heavily, defense remaining undervalued, and weighing AL bullpens more heavily than their NL counterparts–a precedent for health among a teams’ starting pitching is another idea previously unmentioned).
My points, which seemed clear enough, were that professional human prognosticators (not the biased math-haters like me) nearly uniformly disagree with a 79-win projection for the Orioles, and that three years is a large enough sample to start to draw some new conclusions. The second is obviously an opinion, and the opening of Rosencrantz and Gildenstern suggests there are people that disagree with the old saw I quoted. Those people would logically continue to disagree after the fourth iteration matches the first three. The first point suggests that the models have been slow to evolve.
As to gems (or gens), a rotographs writers think Travis Snider will hit 30 homeruns (but can that replace Markakis’ leadership in the clubhouse?). I think the linebacker we’ve got at second base will hit 25 if he gets another 550 at-bats, but again, I’m crazy. The other three-year trend has been undervaluing the O’s pitch-to-contact, runner-stranding starters (could xFIP weigh strikeouts too heavily?).
Anyway, after some light reflection, I do think that bullpens are more important in the AL than the NL, and that therefore having a deep bullpen would be more predictive of team success in the AL than the NL. Can somebody respond to this?
Pointing out a few areas where the Orioles may have beaten expectations or might be particularly strong isn’t really a theory on where they are systematically undervalued by predictive models.
As I mentioned before, you are going into this with the assumption that there’s something wrong with the models (that specifically undervalues the O’s) instead of analyzing whether there’s actually something wrong with the models. You have decided the “people are right” before the season even starts.
That a model doesn’t predict every outcome doesn’t necessarily mean that there’s something wrong with it. No model predicting anything worth predicting is going to be right all the time. There could very well be a flaw in these models that undervalues the Orioles… but just pointing out that they’ve out-performed expectations for a few years doesn’t really prove that.
My question to the Balitmore mob is why do you care what the projections say? If Baltimore really has figured something out, you should be pleased as punch that state of the art projection systems don’t know what it is. This means your advantage would persist into the future!
But instead you chalk it up to bad faith and willful blindness, and demand that the analysts here figure out what the O’s are doing, and broadcast it to the wide world. That would mean your advantage is arbitraged away!
I really just think that these projections are mirrors, which since fans (myself included!) have way too much of their self esteem wrapped up in a baseball team, cause people to react with rage if they project something ugly back
Baltimore fans will always feel like underdogs. Football, baseball, badminton, roller hockey, it doesn’t matter. Other people like to pick on Baltimore.
Yankee preview seems pretty arbitrary / subjective… For example:
-“this rotation does not look good enough to make up for a pretty mediocre offense”
-“But you don’t really want any of these guys to be your best position player”
It’s a little disappointing to read a Fangraphs article that largely ignores any actual evidence in favor of what things “look” like and who has the most “star-power.”
The article also seems to ignore defense, chalking winning up to only offense and pitching.
On another topic, do the Red Sox really have 5 mid-rotation SPs? I count more like 2 or 3, based on projections. While the MSM meme might be “they don’t have an ace,” reality is more just that their SP is questionable overall. Some guys might outplay expectations and they might just out-hit their defense, but it seems inaccurate to say they have mid-rotation guys front-to-back.
Chiming in with other Orioles fans here —
I don’t think the Orioles are a good team, but I think that the Orioles Magic is stemming from getting AAAA scrubs to play at their talent ceilings by magically getting them on the field when variance swings their way and off the field when variance doesn’t swing their way. This might be Showalter, this might be Duquette, it might just be dumb luck that can’t possibly continue. I don’t see how a model that accounts for floor and ceiling performances can reasonably predict that, and I don’t expect anyone at Fangraphs (or Szym) to tweak the models to predict it… because, why? What proof do we have that this is a skill and not just dumb luck?
You’ve captured my feelings very well!
If you have enough similar-quality options readily available and can interchange them with accurate foresight to maximize the hot hand and minimize the decliners, then the overall average production is enhanced – sometimes by a considerable amount. Buck makes Dan look good and Buck could not do this unless Dan first understood how to play the game.
Can it continue? Maybe – Earl stayed ahead of the curve and typically produced a team that was more than the sum of its parts. But even Earl’s teams missed the playoffs if the talent was simply too weak.
Another challenge is 2015 and a large number of potential free agents. A slow start might result in a sell off, and this would be interpreted, unfairly, as proof of regression to the mean and supporting statistical models betting against the Orioles’ flaunting the odds.
How Dionner Navarro the Blue Jays opening day DH?
Orioles fan here. I’m more optimistic than the models, but the idea that Dave has some sort of vendetta against the team is ridiculous. His take here is completely reasonable.
Keep up the good work.
It’s very simple. Fangraphs is in the clickfarming business, just like every other online publishing property, therefore, they have a (perhaps unconscious) bias to produce content that makes the clicks clickety-click.
Since they know that everyone from Baltimore is a thin-skinned rage monkey with an inferiority complex, they may have a bias towards underrating the Orioles, thus encouraging hundred comment screeds. Since they know everyone from Boston is a solipsistic thumb sucker, they may have a bias toward overrating the Red Sox, thus encouraging multiple re-reads as they bask in the onanistic glow of their own self-regard.
Or it could just be math, what do I look like a scientist?
Awesome
FYI I am from Baltimore, and we are thin skinned, but you have to realize the context. We are a small city that is directly jammed in between Washington DC, Philadelphia, and Pittsburgh. Those are all huge sports cities with huge fan bases. We have to earn our niche.
We in Baltimore get this crap all the time with our dominant, but unlikable franchises the Ravens and Orioles, and we always have the best damn fans in the game. Our fans are local, and passionate, and we sell out everything. The Orioles are back, and the AL east knows it. DEAL WITH IT.
Lol at calling the Orioles “dominant” one playoff series win in 17 years gets you there? Pretty low standards
Oh also you sell out everything and have such great fans? Before 2012 when the Orioles played the Sox or the Yanks there were more Sox or Yanks fans at Camden than Os fans
Is there an article that provides transparency on how Fangraphs develops these models? It would be good to know how much manipulation goes into the process.
As for Baltimore’s projection, it’s hard for me to believe that every O’s starter will have an ERA north of 4.00.
Oh, how the mighty AL East has fallen! Behold, the division of mediocity! Unlike the NL Central (the other division where every team does indeed have a legitimate chance to win the division, even if it’s not anywhere near as balanced), where there is expected to be fierce competition and at least one strong playoff team, these guys are going to be fighting each other to just not lose more games than the other teams. We could see our worst playoff team since the 2005 Padres won the NL West at 82-80 or even further back!
Let’s see: The Yankees are a strange mix of mostly overpaid former stars past their prime and unproven youth. The Red Sox have spent their way into a “stars and scrubs” approach with a merely decent starting rotation and embarassing LF defense. The Blue Jays have made some nice upgrades but still have no real 2B, an inexperienced rotation, and a lousy bullpen. The Rays’ usual low-budget MO has finally caught up to them with few young stars left to replace the latest wave of talent that they were forced to give up through trade or free agency. The Orioles have lost a good chunk of their offense to free agency, still have their All-Star catcher recovering from TJS and unlikely to be at full strength until late season at best, and a rotation that was nothing special even when they won the division last year. I would’ve picked the Blue Jays before the Stroman injury, but now I have no idea who will win.
The weakness of the AL East was often commented on last year, and yet it had the highest winning percentage of any division in baseball. Tampa might drag that down a little this year, but the other four teams are all good again (or at least the yankees and red sox will be as they keep adding to their rosters, neither has any reason to sell off players). The NL East is actually a weak division, and appears very likely to finish with the lowest winning percentage amongst the six.
No, they’re not weak, but they are uniformly mediocre. Every team in the division has good points but also major question marks to the point that the division is very unlikely to provide more than one playoff team this year.
The rotation I get but seriously,LF defense? We have no idea if he’s “embarrassing” yet. You have to cover less ground than the average LF and they’ve won world series’ with Manny Ramirez playing LF. How about their great defense almost everywhere else? Why don’t you mention that
Fangraphs doesn’t hate the Orioles, they’re just bad at statistics.
If you doubt this, show the article that Dave linked in his analysis of the O’s to any professor of statistics at a decent university (or even any competent high-school AP stats teacher).
I laughed out loud.
“Boston’s front office has made a bet that you don’t need an ace to win in the regular season, especially if you have solid mid-rotation starters from front-to-back. ”
this has been repeated too often this offseason imo.
eyeballing last year’s leaderboards, a rough estimate would define “solid #1” as a reasonably healthy 4+ war/200ip pitcher, a “solid #2” as a 3+ war/200ip pitcher, and a “solid mid rotation guy” as at least a 2war/200ip pitcher.
this year’s per200ip projections:
porcello: 2.5 (good #3)
miley: 1.7 (#4)
buchholz: 1.7 (#4)
masterson: 1.3 (#5)
kelly: 0.8 (fringe sp)
that just cannot be described at all as “solid mid-rotation starters from front-to-back”. that’s just a false description of what they have, even if they stay healthy.
Glad someone pointed this out.
Boston could be a serious contender if they manage to offload some of their excess position-player talent for some pitching, but I’m not seeing them make it with the rotation they have.
Perhaps they could work something out with the Nationals.