One of the famous scenes from the film Moneyball involves Billy Beane revealing his plan for replacing Jason Giambi. He explains that the A's are too poor to find another first baseman with a .477 OBP and 38 home runs, but that the team can replace his production by finding a decent successor at first base and then also upgrading at a couple of other weak positions. By offsetting their losses at first with gains elsewhere, he claims they can recreate Giambi's numbers in the aggregate.
In the movie, Beane replaces Giambi (.477 OBP in 2001), Johnny Damon (.324) and Olmedo Saenz (.291) with Scott Hatteberg (.374 OBP in 2002), David Justice (.376) and Jeremy Giambi (.390). In reality, Jeremy had already been on the team in '01 and therefore wasn't really an addition, but in reality Saenz also stuck around in '02 and raised his OBP to .354, so the overall theory still makes sense. Multiple upgrades made up for one large downgrade.
You could argue that having several solid players isn't a full replacement for having a truly great one in the middle of your lineup, and you might be right. But I can't help but notice that the top stars in baseball -- Mike Trout, Clayton Kershaw, Andrew McCutchen, Giancarlo Stanton, even Josh Donaldson -- haven't come particularly close to winning rings, while the last two World Series have been played by teams with relatively even distributions of talent. And of course, we'll always remember the 2001 Mariners, who set the AL record for wins right after losing all-time greats Randy Johnson, Ken Griffey Jr, and Alex Rodriguez. There might be something to this idea of building a complete team rather than focusing too much on collecting big stars, especially when your budget allows you to choose only one or the other.
The task at hand: Simulate rain
In this case, the player who needs to be replaced is Josh Donaldson. The A's don't have another infield slugger who is a good bet for 25-30 homers and an .800 OPS. In order to make up the downgrade from Donaldson to Brett Lawrie, they'll need to upgrade somewhere else -- for example, the rest of the infield. We already looked at the potential improvements at first base, so we'll leave that out and just focus on the real weaknesses, which are second base and shortstop.
My methodology is simple. For last year's players, I used last year's stats. The 2014 season is our baseline, and recreating it would be a success considering that Oakland finished fourth in MLB in scoring. For 2015, I used Steamer projections, because that's the only projection system out yet and it seems like a respectable one. This is strictly a case of "what did we get last year, and what can we expect to get next year?" For the purposes of this discussion, it doesn't matter how Donaldson does in Toronto, it only matters what hole his 2014 stats leave in next year's lineup. It doesn't matter how many games Lawrie missed last year, only what he can be expected to contribute in 2015.
Donaldson hit .255/.342/.456 last season with 29 homers. His numbers were bogged down by his June slump, but that's irrelevant; this batting line is the one that contributed to the A's strong offense, to their 729 runs. Overall, we're looking at the entirety of second base, third base, and shortstop, and those three positions were covered by six different players:
That group combined to post a .792 OPS at third base (4th in MLB), a .666 at short (20th), and a .579 at second (30th). (Note that those marks only count their plate appearances while playing those positions. Callaspo also played some first base and DH, and Donaldson DH'ed a handful of times, but they did some of their worst hitting at those positions. If you're worried about cherry-picking stats, know that the group's numbers actually look slightly better than they would have looked if we'd used their full 2014 stats.) Add them all together, and you get the following batting line:
2B/3B/SS: 2,005 PAs, .244/.319/.365 (.684 OPS), 39 HR
That .684 OPS is uncanny, because it's exactly the same mark that Oakland's first basemen put up last year. The A's collectively got a .684 OPS from their entire infield. To put that into perspective, Jed Lowrie, who had a really bad year at the plate, posted a .676 OPS on his own; he earned an OPS+ of 93. MLB shortstops combined for a .678 OPS overall. The A's essentially had an average-hitting shortstop batting at every infield position. Given that they had the real Lowrie at short, another way of looking at it is that, for as good as Donaldson was, second base was so much worse that combining the two created a below-average batting line. Again, second base was worse than Donaldson was good, and improving on it should be all but guaranteed.
And what can we look forward to in 2015? We know that Brett Lawrie will start at third, and we can guess for now that Marcus Semien will get every chance to stick at short. For now, we'll factor in a platoon of Sogard and Parrino at second, since that's the most likely scenario until/unless Beane makes another move for a replacement. Here's what Steamer has to say about them:
Lawrie: 513 PAs, .263/.324/.427, 16 HR
Semien: 571 PAs, .239/.317/.396, 16 HR
Sogard: 442 PAs .250/.314/.337, 4 HR
Parrino: 419 PAs, .224/.292/.322, 5 HR
Total*: 1,945 PAs, .245/.313/.375 (.688 OPS), 41 HR
* percentages were simply averaged, weighted for number of plate appearances
* For the sake of curiosity, Donaldson is projected by Steamer to hit .264/.342/.470 with 27 HR
The immediate takeaway from that list is that this group is projected to be nearly identical to last year's infield. In other words, the upgrade from Lowrie to Semien and the improvements from Sogard and Parrino could be enough to make up for the downgrade from Donaldson to Lawrie.
Let's take a moment to see if we buy into these projections. Lawrie had a .722 OPS last year despite having his season repeatedly interrupted by injuries, so an increase to .750 as he inches toward his physical prime isn't unreasonable. Steamer is giving him credit for playing roughly 117 games, so a couple DL stints are factored in already (and since second base won't demand 850 plate appearances, Steamer is assuming Parrino or Sogard will be filling in for him when he's hurt). He hit 12 homers last year, so 16 is reasonable with an increase in playing time.
As for Semien, this projection feels like a worst-case scenario more than anything. He's given credit for 130 games and a batting line similar to what he did last year. However, the whole point of his acquisition is that Billy Beane thinks he has some upside at the plate and could improve in 2015. Furthermore, if Semien does outperform his Steamer stats then he'll likely play more than 130 games, and every extra game he plays pushes Sogard or Parrino to the bench and improves the group's numbers. We can use this conservative projection for now, safe in the knowledge that we aren't being overly optimistic, but also hope for more in the backs of our minds. The one generous part of Semien's Steamer is the 16 homers, but given that he hit six in 64 games last year and has shown power in the minors it isn't an unreasonable estimation.
Sogard probably has the most favorable projection. Steamer is assuming his BABIP luck will stabilize and that he'll get some of his modest power back; neither are unfair assumptions, but neither are guaranteed to happen. Sogard rebounded in the second half last year and hit more like his 2013 self, but he'll have to prove that his weak first half was a fluke rather than a new trend. Given that he has only seven career homers in nearly 1,000 plate appearances, I don't believe for a moment that he'll hit four next year, but that slugging percentage isn't out of line if his batting average does indeed recover. This projection is basically one notch down from his 2013 breakout year, putting it on the happier side of realistic.
The numbers for Parrino look about right. With extended playing time he could probably hit .220, and with enough at-bats he'll run into a few mistakes and pick up those five dingers -- he seems to have slightly more power than Sogard. Steamer still has him with a sub-.300 OBP, so we're keeping things realistic. He could easily be worse than this or even fail to hold down the job at all, but he could also luck into a BABIP-fueled .270 average. I'm comfortable with this modest projection.
Of course, offense is only one side of the ball. What about the defensive changes? Donaldson was a stud last year, but all reports are that Lawrie is an ace at third as well and that there might not be much drop-off at all. Semien won't be amazing at short, but he probably can't be worse than Lowrie was; we'll call it a wash for now. We know that Sogard is comfortably above-average at second, and he'll be playing the same position as last year. Parrino is a plus at all three positions, which means he should be an upgrade over Punto (who was average last year) and Callaspo (who was below-average). So, any downgrade from Donaldson to Lawrie is likely made up for by Parrino taking innings from his inferior predecessors. And if Semien turns out to be better than Lowrie (i.e., not the worst in MLB), then the defense might actually be improved overall.
As you can see, a lot of this hinges on what Marcus Semien does. Will he be able to handle shortstop, or will he have to fake it like Lowrie did? Will his bat develop, or was last year the best he can do? The good news is that this current projection, the one in which Semien has a mediocre bat and plays porous defense, still gives the A's identical production to last year. Heck, if you cut Sogard's homers from four down to two, the three positions even combine for the same 39 dingers. And if Semien beats his projection and pans out into an above-average player, then the infield has actually gotten better despite the loss of a superstar.
But wait! What if there is another option behind Door No. 2? What if Beane finally finds that shiny new second baseman, the one who can rid us of our keystone concerns? That player probably doesn't exist, but at this point nothing Billy does will surprise us. What would the player have to look like in order to further upgrade the infield?
Realistically, a new second baseman will have to replace Sogard. I think Parrino is a lock for this roster. With Punto out of the picture, Parrino is the only guy with MLB experience who can back up Semien at short -- Sogard can't cut it there on a consistent basis. Parrino is also out of options, whereas Sogard can still be safely sent down to Triple-A if needed. If a new infielder is brought in, Sogard can be stashed and Parrino can't, and the A's hate to lose players from the organization -- especially when they play such a thin position.
Where we stand
So, we're right back in a familiar place: go for a second baseman, but only if he's better than Sogard. That either means clearing a .650 OPS with good defense, or clearing .700-.750 with a mediocre glove. You'd think it would be easy to find that player, but apparently it isn't considering how many teams are struggling to do so. Even the Yankees are going with an untested youngster, Rob Refsnyder, and they are practically allergic to rookies. There isn't even a decent veteran retread out there who the Yankees will take a chance on. That's when you know things have gotten serious. The Blue Jays and White Sox would also probably love to have Sogard, and maybe the Orioles and Angels as well, so he's not even the worst bet in the league. Not by far.
The good news is that even if Semien underwhelms and Lawrie misses 45 games and Sogard spends another year platooning at second, the infield should withstand the loss of Donaldson with no decline on either side of the ball. That's pretty incredible, and it speaks to both how weak second base was in 2014 as well as Beane's ability to piece together useful parts into a greater whole. And if Lawrie or Semien breaks out, or Beane finally finds that new second baseman, the lineup could even be better than it was last year.