The Baltimore Orioles Offense: More Boom Than Bust
Prepare yourself for an unpopular opinion – there is a difference between a well-pitched game and a low-scoring game contested by two bad offenses. Not every quality start is a gem, thanks to the current state of offense in Major League Baseball.
On the same night the defending World Series champions eked out seven total bases in a 19 inning game, the Baltimore Orioles slugged six home runs. The following afternoon, they hit another three bombs. After a fallow period of one day, the O’s got back on the bats, hitting three MORE home runs Monday night, fueling yet another win for the first place team in the American League East.
Over the past seven days, the Orioles hit SEVENTEEN home runs. That is more than the rest of the AL East combined in that stretch, more than 11 teams can claim in the second half. It is this power surge that keeps them afloat and in front. It is this suddenly rich mine of home run power that helps them overcome/offset an otherwise underwhelming squad from top to bottom.
While we all stare at their lineup and wonder how they manage to remain first in a deeply-flawed division, the O’s look like a club that zigged while the rest of baseball zagged. Always a team built to score runs, they simply struggled to get all their pistons firing in sequence – until now.
Mike Petriello noted a month ago that the Orioles lineup is a weird mix of underperformers and out of nowhere stories. Adam Jones and Nick Markakis are the only players to look like “themselves” for the balance of the season. Many of their regulars are well off their career offensive marks, with wild swings in production. As a group, they only managed an 82 wRC+ in June. Since then, bombs away. Here in early August, they boast the best offense in the game, crushing 21 home runs, which drives their 139 wRC+.
A half-month arbitrary split is insignificant bordering on dubious, but all those runs count and the wins in the standings do, too. They’re winning these games on the back of surprising performance by players who either struggled early or were not counted on to produce much of anything in the first place. The Orioles stubbornly remain at the top of the division, despite an underwhelming collection of names (and no-names) and a pitching staff that doesn’t exactly strike fear in the hearts of opponents.
Chris Davis is having an awful time in 2014, struggling to make anywhere near enough contact to utilize his prodigious power. That said, he has four home runs this week. He has 14 hits in the last 30 days, six of which left the park. Caleb Joseph turned from a replacement-level backup catcher with great numbers in his second go-around at double-A to….a replacement-level backup catcher who also managed to hit a home run in five consecutive games.
J.J. Hardy knocked out two homers Saturday after managing just four over the season’s first four months. Manny Machado emerged from his post-injury struggles to post great numbers (before falling to injury again.)
Watching Baltimore dismantle the Cardinals this past weekend was a delight. Not only did the barrage of home runs electrify the home crowd, it was a welcome respite from the underwhelming action around the league most nights. Not all 4-2 games are created equally. If the games are going to be long, they could at least be exciting.
For the O’s, It isn’t about the sustainability of this flurry of home runs, it’s about the ability to build bridges all the way to the post-season. Ride the offense for a while as the pitching staff treads water. The Orioles are working hard to ensure a healthy rotation in October, should they get there. If they win more games than they lose in the dog days of August by turning them into slugfests, so be it. Given the stout bullpen and superlative defense (and Buck Showalter’s legendary management of said group), playing low-scoring games might suit them just fine.
The kind of talent the Orioles have on board lends itself to a boom and bust cycle just like this. That talent doesn’t endear itself to our projection models so well, as the O’s forecast to score and allow a nearly identical amount of runs over the season’s remaining 44 games (4.41 for, 4.40 against).
Even going .500 for the rest of the season is more than enough for the O’s to glide into the playoffs in the AL East. Not only is it the style of team they built – crafted in loving memory of patron saint, Earl Weaver – but it takes great advantage of their home ballpark and doesn’t come at the expense of their aforementioned defense, a key plank in their surprising run prevention.
The lineup isn’t pretty, as many of the names regularly penciled in illicit eye-rolling as much as awe. And forecasts reflect this, as ZiPS and Steamer combine to roll their eyes at the Orioles. But here they are, owners of the biggest division lead in the game and the fifth-best playoff odds in the game.
While their AL rivals pursued starting pitching, the Orioles added another deadly bullpen arm. They’re an enigma with a soft schedule from here on out. It’s starting to matter less and less if they can keep it up, as “it” is getting too late to matter if it’s real or not.
Drew used to write about baseball and other things at theScore but now he writes here. Follow him on twitter @DrewGROF
“wins in the standings do to.”
wins in the standings do too, wins in the standings do to you, wins in the standings don’t*
*that edit would be the worst option because it’s factually untrue
Is there a particular reason that you can identify that causes boom/bust teams to have problems with projections? Is it because home runs and defense have a lot of variability which makes it hard to predict?
I would also argue that health and depth are huge reasons why the O’s are still in it.
The Orioles have not been without injuries. Wieters, of course, was the big blow, but they have also lost Davis, Machado, Jimenez, M. Gonzalez and Bud Norris to the DL this year. Their opponents have lost more, but it’s not like the O’s have got off scott-free (how do you write scott-free? Is it Scott-free? Scot-free?)
scot-free, as in “scot and lot,” an old phrase used for certain types of borough land tax owed in England. Scot-free would be exempt from the tax.
To be fair, some of those dudes they lost kinda sucked. (Losing Jimenez, for instance, was addition by subtraction.)
I think it just means their projections have a much higher variance.
honestly every projection on the site should come with some measure of variance, or at least a 25% and 75% projection as well. Then these things would seem less shocking.
So this article is about the Orioles being bad but going to the playoffs anyway… right?
If the Orioles are bad then who in the AL is good besides Oakland? Detroit is great on paper but pretty average when it comes to actually playing games (largely due to their incredibly poorly constructed bullpen). I guess the Angels may be good but they certainly have flaws as well.
I really dont see the Orioles being any worse than any other contender at this point.
Not to slight the Orioles, but the Angels are markedly better than Baltimore. (Not to say the Orioles are bad – they’re not – or don’t deserve to be in the playoffs, etc.)
Yet when they played ’em, they beat ’em.
On paper. On the field the O’s went 4-2 against the Halos and lost the 2 by 1 run each (one in extras).
Everything is terrible.
Everything is awesome.
Random variation.
The Orioles have the 3rd best record in the AL for the last 3 years (A’s and Tigers are better). So their success is not a fluke, and they are not a bad team, as uncomfortable as that might be for projections systems and advanced stats. Remember, stats are guides, not the real thing! To say that the Orioles are bad or lucky begins to lose its merit after a while. It’s not over yet, and the postseason is a crapshoot, but this is a good baseball story.
Welcome to Fangraphs, person who doesn’t know what a standard deviation is!
Was this addressed to me? Or to those claiming the Orioles are bad because they are projected to be? Or do you just like acting superior for no reason?
Well there is a reason.
Oh c’mon, he’s learning and reading man. Plus he’s interested. Just cause he didn’t bring a Z-score cheat sheet does not mean that he can’t comment on the article or read it for crying out loud. More Fangraphs readers is good. More people learning about our passion is great and good for Fangraphs overall. If you have a problem with his post try to help him understand what standard deviations are and what they mean. The old IQ scale is a good place to start.
What do I look like fucking Google?
Honestly you just look like an internet bully.
You look like one of these -> *
Guys I asked you a question.
The larger the sample size, the smaller the expected standard deviation. Granted, the first half of 2012 was a fluke, but we are talking about 361 games (81 + 162 + 118) from the second half of 2012 until now, and I think at this point it is at least fair to question whether or not the Orioles performance falls within normal variation or if the projection models are underselling their true expected performance.
The O’s record from the middle of 2012 until now is 202-159, which is approximately a .560 winning percentage. Considering that the standard deviation of 400 coin flips is 10, then the it seems that the Orioles have outperformed the approximate projection models by 3 standard deviations over this period. Specifically, adding 30 wins and subtracting 30 losses to the actual record would result in a record of 172-189, good for a winning percentage of .476, and this .476 winning percentage closely track the projected record for the O’s during the period.
Therefore, if the projection models are accurate as to the Orioles, then when considered in isolation there is approximately a .3% (100-99.7) of the O’s exceeding their expected win total by 3 standard deviations, which they have done or at least come very close to doing since the middle of 2012. As such, I think it fair at this point to ask the question of whether the projection systems are/have been underrating the Orioles.
Pretty sure Dave Cameron would say this is wrong.
Really nice work here. I’d love to see a response to this.
Pretty sure you should think for yourself.
the standard deviation for a teams’s win-loss record is +/- 10 wins. So if the Orioles are projected to win be a .500 team and end up winning 91 games that’s only one standard deviation, meaning even their most successful season (i.e. 2012) was barely 2 deviation from their projected record.
This is what happens when you open the doors to the masses… *cough*Instagraphs*cough*
Bill.
Chris, you are underselling the O’s performance relative to the mean projection. The probability that they exceed the projection by more than 3 sigmas is 0.15%. The probability that they do better OR worse than 3 sigmas is 0.3%. In this case we only care about the single tailed probability.
Sam, over a 162 game season, the standard deviation wrt a team’s wins (assuming said team was a .500 team) is 6.4. So the 2*sigma bound is about 94 wins. There is a 2.25% chance that a team does better than ~13 wins above their projection (if we assume the projection was good in the first place), so when this happens you should probably look to see what happened.
On the other hand, given accurate projections one team in 44 should end up doing more than 13 wins better, so we shouldn’t exactly freak out when it happens. It should happen once every season or two. But we shouldn’t expect the same team to exceed projections over and over again; the team that overachieves should be random.
In the Orioles case, as Chris pointed out, either they are that one team in 700 that overachieves in this way, or the projections are not accurate (in their case). Thus, we should probably investigate this further.
A point in Fangraphs’ defense though; they may be attempting to research this and may not have found anything conclusive as of yet. We’re not privy to everything that each author is looking into at any moment in time.
Nard, why do you assume that everyone that disagrees with you has not been a long-time Fangraphs reader?
Or did you just take trolling 101 and decide to show us all what you know?
Maybe this is what Dave should be looking into instead.
For a system that can be well modeled by a Gaussian distribution, the standard deviation is independent of sample size. See stats 101.
Welcome to Fangraphs where when we a see a team consistently outperform projections we seek to investigate to see if this is could be due to factors we are missing. It coukd be luck, but it might not be. Ten years ago there were smart baseball people who thought run differential was the be all and end all. The truly smart ones were willing to change once confronted with evidence to the contrary. We are much better at understanding value in baseball than we have ever been, but only an idiot would contend that this knowledge is complete.
Aren’t articles like this the reason Instagraphs exists?
yeah I’ll agree with that, and note that the articles from mr. fairservice i have seen so far have seemed to be a little content-light, or info-light. this was fine and somewhat interesting, but probably not warranting this much length without more depth.
“Many of their regulars are well off their career offensive marks, with wild swings in production. ”
wait a sec – aside from Davis, which regulars are/were “off of their career offensive marks”?
Markakas and Cruz have hit above expectations all year, while Pearce and Young have been way above in part time roles.
Davis has a been a huge underachiever…..and who else? ryan flaherty?
Preseason ZIPS wOBA v. actual season wOBA:
Pearce .325 v. .371 (+0.46)
Cruz .366 v. .340 (+0.26)
Markakis .340 v. .322 (+0.18)
Wieters .364 v. .321 (+0.43) & Joseph .312 v. no proj
Jones .346 v. .342 (+0.04)
Machado .332 v. .331 (+0.01)
Hardy .311 v. .318 (-0.07)
Schoop .266 v. 309 (-0.43) * – was a generic regressed rookie projection
Davis .310 v. .377 (-0.67)
Young .351 v. .309 (+0.42)
Lough .280 v. .296 (-0.16)
Hundley .270 v. .291 (-0.21) and Clevenger .292 v. .290 (+.002)
Flaherty .263 v. .300 (-0.37)
Seems to me that’s a team that’s largely overachieving their projections offensively with one big underachiever mixed in…. not really a “mixed bag”.
One could look at those stats and make a different conclusion. Really Pearce/Young, Cruz, and Markakis are the only significant overachievers in that group. Wieters played a month. Davis and the 2nd basemen (Schoop/Flaherty) cancel out two of that group.
wOBA isnt everything…Hardy should stick out like a sore thumb career SLG .424 this season .388, he’s made up for it to extent with a better BABIP but the drop in power is a big deal.
For the most part though I agree with you, Jones is pretty much doing what he always does and Davis is somewhat a disappointment but Markakis, Cruz, Pearce and Young have all been much better than expected.
Davis himself pretty much makes up for all the players playing over his head. He’s dropped from a 6 WAR player to replacement level.
the O’s are garbage and it hurts to watch
The Orioles are like a fat dumb kid with too many Pogs and you’re like how did you get so many Pogs fatty?
Or like the little fat guy with super models under both arms. Makes you wonder what he’s got going on and jealous as hell.
Except we’re talking about things that actually happen.
O’s fan here.
Great article.
As an O’s fan, front-running doesn’t come naturally. I can’t shake the premonition that this will end in tears. Machado’s injured again?!!! Of course he is.
Why is it so hard to grasp Drew’s point that the O’s homer-reliance over the past few seasons has allowed them to exceed reasonable expectations and sophisticated projections? And, even if they can’t keep it up, it might not matter. Plus, they are damned fun to watch.
The greatest source of unexpected production from last year (Davis) has plexi-glass principled himself back to earth so hard that the irony of his super-hero action figure give away day stung like stepping on a hornet’s nest.
Yet, here they sit with the biggest division lead as Drew mentioned. Even if the SABR-gods of regression rear their ugly head and the home run luck fades, thanks to their emphasis on defense-first at C and 2B and their surprisingly cromulent pitching they’ve still got a great shot at the playoffs.
A thousand years of losing and that 2005 season would make anyone a pessimist.
If they killing their own division the way they’ve been, they could run away with it in September. Their final 20 days are against the east.
What wasn’t mentioned is that before the bats erupted for the most part the starting pitching has been great in the second half, carrying the team while the hitting (two series each against the Angels and Mariners) was silenced. I had a great view from the bullpen of those seven one-run games. But now I get to pitch again, so bombs away I say.
Are you seriously still doing this? Tommy Hunter has a 2.00 ERA since May 13th. With zero HR allowed. You probably know this.
I freaking LOVE. THAT. USER NAME.
It was given to me by the good folks at Camden Chat. I’m thinking of changing it to Tommy “Low Leverage Living Large” Hunter, to keep with the times.
I always called him Tommy “Home Run and 2 K” Hunter.
Great write-up!
Almost every national commentator underrates the Orioles pitching and assumes that it’s an offense driven team. Granted, the team’s excellent defense helps the pitchers, but the team is second best in the AL east in runs allowed (just behind Tampa) playing in a hitters park. The bullpen is good and rotation has quality starts going six deep (although Ubaldo has been inconsistent).
This is somewhat off-topic, but perhaps a reason that projection systems underrate the Orioles is that the team relies so heavily on depth (“the Norfolk shuttle”) which leads to them heavily using hot players. For instance, in the bullpen, the team tried Evan Meek and Preston Guilmet earlier in the year. That didn’t yield desired results, but they called Brad Brach up from Norfolk and he has been excellent. They tried Steve Clevenger at catcher before Caleb Joseph. When Steve Pearce was hot he played regularly. Now that he’s cooled off, they are using the hot Delmon Young. Etc.
Good write up, but once again full of back handed compliments to the Orioles. I don’t pretend to even have surface level understanding of sabremetrics. I never got into it even though I took 4 semesters of stats as a research focused undergrad. I just don’t believe it has the reliability as some proponents believe. Despite what you say and what you convince yourself of statistical analysis is a subjective science. I used to study psychology and I could make that data dance. In fact there is actually good research that studies subjective influence on statistical analysis. I suspect that there is a subjective bias against the Orioles that undervalues them. Baseball, is one field where there are so many factors that are not yet able to be reliably quantified and thus you get teams like the Orioles bucking projections. However, I can tell you why the Orioles are winning. They have allowed the third fewest runs in the AL, have one of the best defenses and can hit the three run home runs. Just like Earl Weaver used to say. They also have incredibly good management and a lot of depth that the stats cant measure as of yet. I also don’t see good measurements on how performance of one player impacts other players on the team, a measure of chemistry if you will. Or any really good situational measurements but that is always almost impossible to quantify. But this is why we play the games, we can’t control or predict the outcomes. I suspect the Orioles will continue to buck the projections but I am one of their avid fans.
Enjoy their success…relax. Breathe. O’s fans are developing quite the persecution complex around here. (It goes in spells, it was Giants fans a couple years back, then Braves fans, I guess it’s Orioles fans’ turn now?)
Are you surprised? Popular opinion has been that the O’s are a mediocre team overperforming for the last 3 years. For the people who don’t tune it out it has to get old.
Or maybe we should wrap our head around the idea that great management is worth more than given credit to? Even a team who looks great on paper can be ran into the ground.
You refer to the AL East as a deeply flawed division.
Per my literal back of the napkin calculations the AL East is 12 games over .500 placing it among the most competitive divisions along with the NL Central (+12) and AL West (+11).
If that constitutes deeply flawed then what are we to say about the AL Central (+1), NL East (-9) and NL West (-27)?
Flawed in the “five mediocre teams shuffling wins back and forth” sense, not the “two great teams” sense in the AL West etc. But this could be my own bias/bitterness.
Except that there is one team – Baltimore – that is not shuffling wins back and forth so much as winning a lot, and building the largest division lead in baseball. Maybe it will all come crashing down in the last 6 weeks, but as of now the 5 mediocre teams narrative doesn’t hold. (And Boston is something south of mediocre this season, I would say).
I think this is pretty much a bunch of BS. And by BS, of course, I mean Buck Showalter. Being born in Bal-dee-moore, I love the O’s, but I’ve been expecting a “regression to the mean” for the past two years now, thinking ’12 was a fluke. Well fluke that! With production coming from different and constantly changing sources (aside from the consistently productive Adam Jones), it seems clear to me that this franchise is the sample case for the potential impact of a manager. Put Showalter at the helm of any half-respectable major league roster, and he’d get them into contention.
You know what makes sports so great? is that You can’t predict it. You may think that teams are good, or that players will breakout/bust, but what makes baseball in particular so great is the x factor of it all.
to focus on the 95 percent(gota)of the time something is right, misses the beauty of the 5 percent something that unexpected happens. You can’t explain it, because it cannot be explained. there is no metric for morale.
winning and confidence are attitudes. players are humans, not robots. Maybe Buck does possess superior player-whisperer magic powers. Maybe he has created magic depth where the people who are called up to play are on the same page with the rest of the team. Maybe it is just a fluke. But the point is to not try to measure what makes a team a team by the sum of individual parts, which is what a lot of this tries to do it appears.
I appreciate the willingness to improve the system, but the reality is until a computer program can predict what happens in the locker room when the lights are off, all of this basically amounts to what Ptolemy did to explain planetary motion.
useful, somewhat predictive, but not elegant.
to me, a system should not be used to predict winning with mathematical models, when the formula used is not elegant. and never will be. but that’s my opinion, and i’m somewhat of a crackpot.
The O’s have dramatically outperformed the computer models for 3 years now. Why? Roster management? Bullpen management and construction? Defense? Depth? Under-appreciation of home runs?
I’m just a little tired of hearing “luck” thrown around. This is a purported community of scientists. If you can’t answer the question, just say so.