Back in May, I wrote this piece at the ESPN SweetSpot mothership -- it was a reaction piece to that particular Saturday's action, with the specific action I decided to focus on being Jeff Samardzija throwing six scoreless innings against the Braves but failing to get a win because his team, yet again, did not support him. At the time, that kept his record at 0-3 with a 1.62 ERA. He's now all the way up to 2-7, though the ERA has come back to the pack at 2.83, thanks to a 4.33 mark over the 52 innings since I wrote that article.

Well, he'll get a lot more support now that he's joining the only team in the league that's scoring five runs per game, not to mention the team that turns batted balls into outs at the second-best rate in the league (behind only Seattle), not to mention not to mention the team that plays its home games in one of the toughest run environments in baseball, especially for left-handed hitters.

So the first question: should you be worried about that 4.33 ERA over his recent starts? I wouldn't be, not unless you have a better reason to be worried than "he has a 4.33 ERA over 52 innings." Every sample is a small sample, but that's a particularly small one when it comes to drawing conclusions about a starting pitcher's present talent and likely performance going forward. His preseason projections took into account far more information than those 52 innings: Steamer had him at a 3.83 ERA, Oliver at 3.85, ZiPS at 3.56, and PECOTA, the outlier in the other direction, at 4.37. The rest-of-season projections have moved only negligibly, which makes sense because a half season of 2.83 ERA shouldn't move the needle on any robust projection system.

What about his pitches over that period? Here are a variety of PITCHf/x stats, courtesy of Brooks Baseball (and therefore relying on PitchInfo's classifications):

Do you see anything there? I'm not sure I see anything there. Of course, this is the trouble with PITCHf/x "analysis" -- you start breaking things down into ever more component parts and quickly overwhelm yourself with mountains of data. The key movement on Samardzija's pitches seems to be down, but his velocity is holding steady and he's increased his grounder percentage substantially. Ground-ball rate actually stabilizes quite quickly, and while this is only nine starts, it's close to 150 balls in play, so it's harder to dismiss the extra grounders as a mere aberration. Which is different from saying it's not an aberration, and that this is what Samardzija is now, but on the spectrum from "pff he hit a homer in one at-bat, whatever" to "he's been terrible at baseball for 15 seasons now, why would we expect him to be good now?" the grounder rate may fall more on the real side of that line.

The thing about everything I've said above, and in particular the projection systems' beliefs in Samardzija's true ERA talent, is that one thing he definitively is not is an ace. Frankly, this is the part that makes me queasy -- do you trade Addison Russell, who is by universal acclamation one of the best prospects in all of baseball, along with a variety of others, for a second or third starter? Plus Jason Hammel? Yes, it's a full 2015 plus three months of Samardzija, and yes it's at below-market rates because he hasn't hit free agency, but these things are true of David Price, too, for whom I'd have been more comfortable giving up Russell (and whom the A's attempted to acquire, according to Jeff Passan's sources). Dropping down a tier (or two?) to the level of Samardzija, regardless of how good he was over his first N starts this season -- is that worth it?

All projections are the "rest-of-season" variety, i.e. the preseason projection adjusted a bit by this season's performance. Caveats abound, of course, in particular given short track records (Gray, Straily), new pitches added (Chavez), and weird career arcs (Pomeranz, Kazmir).

Jason Hammel may be a more important part of this deal than we're all giving him credit for. If you take a sort of consensus of the projection systems and ignore everything else (which is a bad idea on a variety of levels, but we're just doing exercises here -- if you want to give me access to an entire scouting bureau and a database program I can plug all that information into to make heads or tails of it, then super, but until then, shortcuts!), a rotation chosen from all available options, post-trade and pre-trade, probably lines up this way:

1. Sonny Gray
2. Jason Hammel
3. Jeff Samardzija
4. Scott Kazmir
5. Tom Milone

You can flip-flop any of no. 2 through no. 4 if it'll make you less mad. You can also swap in Jesse Chavez for Milone, and it would seem that the A's may be figuring out that very alignment soon enough. Happily, the "choose your best five" is also what the rotation will actually look like post-trade (again, depending on the Chavez/Milone situation), which suggests that the trade represents a real, and possibly marked, upgrade for Oakland:

1. Sonny Gray
2. Scott Kazmir
3. Tom Milone
4. Jesse Chavez
5. Brad Mills

Brad Mills? No longer necessary. Jesse Chavez? Can now regress or run out of gas to his heart's content. Drew Pomeranz? Can take as long as he needs for his hand to heal. Tom Milone? ... well, I have no idea what to say about Tom Milone.

Instead of those types of pitchers making up the middle of the A's rotation, they've got four very good options, each of whom is probably a second or third starter, and one of whom may be a no. 1 starter (if perhaps not an "ace," depending on where your cutoffs are for that particular loaded term), and two solid choices for the last spot, one of whom (Milone) is durable and has options and has performed well this season and the other of whom (Chavez) has had success in the bullpen and may be a solid starter with the addition of his cutter.

Still, has any of this answered the question? Is the upgrade of Samardzija and Hammel over Mills and the lesser of Chavez or Milone worth a top-20 (and maybe top-5) prospect? A player who will fill one of the hardest positions to fill at the league minimum salary for three years, and at a below-market rate for years after that? The best prospect the A's have had since Miguel Tejada in 1998, Eric Chavez in 1999, Mark Mulder in 2000? Tejada put up 22 bWAR at the cost of $11.6 million before hitting free agency. Chavez 25.7 bWAR for $12.4 million. We're a decade on, so the dollars have gone up, but a bargain is a bargain.

Of course, not every top-notch prospect hits: Jesus Montero, Mike Moustakas, Travis Snider, Cameron Maybin, and Delmon Young have all been top-10 prospects. With the Angels breathing down the A's neck, do they hold on to Russell in the hopes that he can anchor the team from 2016 through 2021, or do they survey their competition (a down year for many powers, real shots at World Series runs both this year and next) and solidify their best-in-baseball position to take what may turn out to be the best chances they'll have to put a flag in the stadium? It's hard to blame Billy Beane for taking the latter course, even at the expense of the best prospect he's ever traded, better than Michael Choice, A.J. Cole, Grant Green, Andre Ethier, John-Ford Griffin, Mark Teahen, Jeremy Bonderman, Jason Hart, Mario Encarnacion, and hey, look at the track record of those guys. A few too early to tell, a few hits, a few misses, and probably only one major regret (Ethier for Milton Bradley).

In fact, the name I left out, the guy who was closest to being rated as highly as Russell, is Angel Berroa, who was BA's no. 15 prospect in baseball before 2002, the year after he was included in the pre-2001 deal that netted the A's Johnny Damon and Mark Ellis. Berroa'd had a good year at Visalia in 2000 but was not in the top 100 the next year, but then raked his way through the Carolina and Texas Leagues in 2001, earning his top-prospect status ... before proceeding to have exactly one good year in the major leagues. These are the games you play with prospects, and while you play the same kinds of games with pitchers (who's to say Hammel and Samardzija don't simultaneously blow out their elbows with a pretend wrestling match in the A's locker room next week?), they're at least major leaguers whose performance status is more or less a known quantity.⁵

One more note on prospects: the A's were mediocre (though never quite bad) for a few years in between their 1999-2006 and 2012-present runs, which netted them the ability to draft higher than many of us got used to while joyfully celebrating the Moneyball years. What did the A's get from those drafts? Here, the list of players drafted in the top three rounds from 2008 (the draft that resulted from the bad 2007 finish) to 2012:

It's easy to look at that list and sigh about how poorly the A's have drafted or developed, but don't forget the lessons we've learned from draft research -- the expected WAR return after the first handful of picks is extraordinarily low. Like the player never makes it to the major leagues at all low. Vast swaths of the minor leagues and the draft are filled with players who will wash out, even at the relative top of the draft. Out of 16 picks, getting Choice, Gray, Russell, and maybe Robertson may not be all that bad.

So where do we go from here? Tom Milone has already been optioned out to get Jeff Samardzija onto the roster in time to start tomorrow. This, by all accounts, is no real indication of how the roster will stack up after the All-Star break and is instead more of a quirk of scheduling -- Milone started yesterday, so he wouldn't be available for a while anyway, even in an emergency situation. To get Hammel onto the roster, the trade will presumably "cost" them Brad Mills, because between Jim Johnson and Mills in the bullpen, even with Johnson's struggles, you've got to imagine the team is still sticking with the ex-closer, not the guy whose fastball can't dent bread. Mills wound up costing the A's one dollar plus, what, $60,000 in salary? Lew Wolff dropped $60,000 out of his pocket when he reached in for his car keys and it wasn't worth the effort to bend over and pick it up.

The question after the break is whether you put Milone into the rotation in Chavez's spot and put Chavez into the bullpen, keeping him available to move back into the rotation as needed. The problem is "at whose expense"? Jim Johnson is the only player who has earned a boot, but he'd have to consent to being sent to the minors, and if you're Johnson, why do that? If you force the A's to either keep or release you, you'll either be in Oakland's bullpen or someone else will take a chance on you for basically free, knowing that the A's are picking up the rest of the check. All of which means that any booting of Johnson will result in the A's probably paying for the privilege of watching someone else try to fix him. Now, if the A's think they can't fix him, or that he can't fix himself on the A's, then it's a sunk cost one way or another. But if there's still hope, and if you have two pitchers in Chavez and Milone who you probably can't realistically choose between without getting deep into the minutiae, then you may as well keep all the bodies you can, even if that means sending Milone and his 107 ERA+ to Sacramento to await his next opportunity, which may not come until next year, after Hammel departs in free agency, and even then may be short-lived, what with Samardzija, Gray, Kazmir, Chavez, and Pomeranz all being under contract and with A.J. Griffin and Jarrod Parker hopefully coming back at some point midseason.

The A's had something like an $80 million Opening Day payroll this year. Their commitments or quasi-commitments for next season look like:

Depending on some of the question marks, that's basically $80 million right there. I may have been overly generous with the raises for Samardzija, Moss, Jaso, and Donaldson, so maybe you think the commitment looks more like $75 million. On the one hand, payroll has a way of ever increasing, in particular after a playoff year. On the other hand, there aren't any shortstops on that list. On the other other hand, sure there are, there's a $2.75 million option on Nick Punto. On the other other other hand, Nick Punto, starting shortstop for the defending champion Oakland A's?

There are a lot of moving pieces, including the possibility of long- or medium-term deals for Donaldson and Moss in which the A's would presumably attempt to trade some long-term value that kicks in after the Cespedes, Kazmir, and Crisp contracts expire for short-term savings. But as a sketch, this probably gives some insight into the odds of the A's doing something like re-signing Hammel, especially since he may never get a better chance to cash in than putting up a career-best ERA in his age-31 season, with half of it coming for a team running so far out front that even the national media is paying attention.

Last question: what are the odds the A's make a qualifying offer to Jeff Samardzija? The QO is probably going to be in the range of a bit over 2-WAR money, but that's at the market rate, which doesn't really apply to the A's, a team that needs to find bargains in order to put together a roster. Samardzija is probably worth more than that, and is almost certainly someone who would, without any strings attached, sign a multiyear deal for moderate money, but the QO free agent market is still shaking out, and $15 million or however much the offer winds up costing after 2015 may be enough money to get Samardzija to accept and play out the year to see what happens. The strategy right now has been that every player rejects the offer, but as we see players come to regret that choice, the balance should start tipping back the other way. Maybe pitchers aren't where the scale will tip because if anybody is unwilling to go year-to-year, it should be pitchers, who can lose it all with one elbow-sproinging slider. The A's might not regret $15 million in theory, but, again, that theory doesn't much apply -- the question is whether they'd regret typing up 1/7 (?) of their payroll in one 31-year-old starting pitcher. That risk may be too much to bear, and damn the draft-pick consequences.