What's Up With... Joe Saunders?

Ok, so who had Joe Saunders finishing the month of April leading all qualifying pitchers in ERA? If you tell me you saw that coming, I trust you are enjoying your half-billion lottery jackpot as well. But that's where we stand, with Saunders' four starts for Arizona have resulted in three earned runs over 30 innings of work, with opposing hitters being held to a .186 average. On the basis of this first month, we could be mouthing the words "Cy Young candidate" and "Joe Saunders" in the same sentence. But is this just one of those early-season illusions? Or is it a new, better, Saunders for 2012?

How good has Joe been?

There's an undeniably magnifying effect that happens as the result of a player's or team's performance, good or bad, happening at the start of the season, because it comes on a blank slate. That said, Saunders has thrown at least seven innings and allowed two earned runs or less in all four of his starts this season. In 2011, Only six National League pitchers had a longer streak of such games last year [Cliff Lee did it twice, and curiously, three of the others were called "Tim".] Joe does have some way to match the franchise record, however. Randy Johnson had 14 such starts consecutively in 1998. Of course, this being RJ and the D-backs, his record in them was 5-6...

Certainly, it's unprecedented in Saunder's career, culminating in last night's gem in Miami. That game was the third complete-game shutout he has thrown, and the first since becoming a Diamondback, coming in his 50th start for us. It was also something of a stamina test for Joe, the 116 pitches thrown tied for the third-most among his 165 career appearances as a starter. But using Game Score, a stat developed by Bill James to measure the strength of a starting pitcher in any particular game, we can see that the Mauling in Miami was arguably the best game Saunders has ever thrown.

Saunders' best 10 starts, by Game Score

So, what's changed?

Before getting into that, we should note that Saunders was also pretty damn good at the end of last season as well. Over the final seven starts of 2011, he had a 2.68 ERA and only 9 walks in the 47 innings pitched there. Combining the two years, it means that in all save one of his last eight starts, Saunders has done the seven or more innings, two earned runs or less thing. Going back further, since the start of June last year - so we're now looking at 26 starts and 176 innings - his ERA has been 2.81. This might help alleviate legitimate concerns about the small sample size worked by Saunders, that we'll be focusing on.

Let's see if we can detect some trends, however. The numbers below showcase some stats for Saunders - there are three tows, one for his career numbers to date, the next for the 2011 season and the last for the 2012 campaign thus far.

These are cherry-picked for significance. The first three columns more or less measure how many base-runners Saunders has allowed, and the answer is, not very many. Any WHIP less than one is very, very good, and Saunders is well below that. Some of that is probably not sustainable, because the BABIP - batting average on balls in play - is likely unsustainably low, showing that the defense behind him has been making plays. The lack of home-runs is probably a bit fortunate too, though that may not be entirely accidental, as we'll see in a bit. But the decrease in walks and increase in strikeouts are a positive sign.

The GB/FB column is his ratio of ground-balls to fly balls and is particularly interesting. Previous, Saunders was almost dead-average here - for about every four ground-balls delivered, five were in the air. This year, however, the ratio is now more than six to five - he's getting a lot more ground-balls. This may help explain both his low home-run rate [obviously, balls on the deck don't leave the park very often] That may also play in to the insanely-low level of extra-base hits. In his four starts, Saunders has allowed a double and a home-run. And that's it. Again, ground-balls don't tend to reach the wall. If he can keep generating those, it will certainly help his numbers.

The question is, has Joe's changed his armory of pitches? Or is he just using them more effectively?

An Armory of One...

Fortunately, fangraphs.com has a wealth of information that can help us sort through the data, based off the Pitch f/X data which you see on MLB Gameday. This allows them to collate precise details on the type, speed and location of every pitch thrown by Joe since the 2007 season, when the system first came into play. Here's what the system shows for Joe. The abbreviations for pitch-type are as follows: FA% = Fastball, FT% = two-seam fastball, CH% = change-up, CU% = curve-ball, SL% = slider, IN% = intentional walk, PO% = pitch-out. Pace = pitches per inning, and Pitches = the total number thrown.

What should immediately stand out is the sharp increase in Saunders' use of the two-seamer, to the point where this now represents about half of all his pitches, and is more than twice the number of "standard" fastballs this season. That appears to have been an ongoing process - as recently as 2009, four-seam fastballs were the majority of all Joe's pitches.But if you look at the breakdown of his complete game on Friday, 50 of the 116 pitches thrown were two-seamers, with four-seamers almost matched in frequency by change-ups (30:25), and curves forming the balance, with ten of those.

I also note the almost complete absence of the slider from his repertoire this season - it appears he has thrown it only once to date, with most of those now being replaced by two-seamers. The change-up and curve are still also in there, being thrown about two out of ten and one out of ten times respectively. This does kinda make sense, as the velocity of Saunder's average fastball has been slowly declining. It's now clocking 89.3 mph, down from 91.5 mph in 2008. If you can't pitch as hard, you have to pitch smarter, and it seems Joe has been doing that.

Can he keep it up?

No. At least, not in the sense of a sub-one ERA, at least. That's clearly going to go up, because when his pitches get put in play, they will end up dropping in for hits a bit more frequently, and fly-balls will end up leaving the park more often. However, even the defense-independent numbers like FIP and xFIP, which have been brutally down on Saunders in the past, are much more impressed with the 2012 version. FIP has him at 2.98, while xFIP (which uses an expected number of home-runs allowed) is at 3.68. Both of those are well under Saunders' career figures of 4.60 and 4.49 respectively, and I'd be happy to settle for either as his season's ERA.

The re-signing of Saunders, at the time, seemed a somewhat questionable decision, with the team appearing to have younger, cheaper alternatives a plenty, waiting in the wings. But at this point, with the loss of Daniel Hudson and the dismissal of Collmenter from the front five, it has been to date one of the best moves of the winter. He will allow more runs, but if Saunders can pitch the same way he has, then he could prove to be extremely valuable, giving the team stability and a genuine chance to win every fifth day. For someone considered our fourth or possibly even fifth starter on Opening Day, that's more than you should expect.