Andy Sonnanstine is an unheralded pitcher on a team loaded with pitching talent. Because of the presence of Scott Kazmir and James Shields on the major league roster, and the likes of Wade Davis, David Price, Jake McGee, and Jeremy Hellickson in the minors, many people have overlooked Andy Sonnanstine throughout his career. However, Sonnanstine has the ability to be a solid starter in the majors – right now.

Sonnanstine's ERA last season and so far this season is misleading, and a closer look at his underlying numbers will show you how.

Sonnanstine doesn’t have great “stuff” – his fastball tops out around 90 MPH and he lacks any particularly filthy secondary offerings. However, he pounds the strike zone, has solid pitches across the board, and has a track record of incredible success. In the minors, Sonnanstine has compiled a career 2.58 ERA in 84 games (74 starts), and has a career K/BB ratio of 463/75 in 495 inning.

However, in 22 starts with the (then-Devil) Rays last season, Sonnanstine posted a 5.85 ERA. Critics pointed to Sonnanstine’s supposedly underwhelming stuff, citing the fact that some pitchers are able to post dominant numbers by exploiting weaknesses that exist in minor league hitters – weaknesses which are virtually non-existent amongst major leaguers. Despite his track record, few had high expectations for Sonnanstine, and few people seemed surprised when he pitched poorly.

But if you look beneath the surface, you will see signs that Sonnanstine actually pitched very well last season. First of all, despite having such a high ERA, Sonnanstine posted an excellent K/BB ratio and a solid K/9 ratio, two important predictive stats for pitchers. In 130 innings, Sonnanstine struck out 97 (that’s 6.68 batters per nine), and walked only 26. His main problem was the long ball: Sonnanstine allowed 18 homers. However, even this is not a horribly high number. So why was his ERA so high?

First of all, Tampa Bay’s defense last year was atrocious. Their defense allowed a .338 batting average on balls in play – easily worst in baseball and among the worst of all time. Sonnanstine’s personal BABIP of .326 can therefore be mostly attributed to a poor defense behind him. Additionally, Sonnanstine pitched terribly with runners on base: with no one on base, Sonnanstine allowed a line of .256/.296/.434; however, with runners on base opposing hitters hit .354/.380/.559. With more than one runner on base, this line jumps to .423/.460/.678. Sonnanstine stranded only 60.6% of the runners that reached base, which is not surprising considering how poorly he fared with men on base.

But are Sonnanstine’s struggles with runners on base indicative of a lack of skill, or bad luck? In the minor leagues he had no problems pitching with runners on base: in triple-A, opposing hitters hit .213/.263/.320 with no one on base, and .202/.242/.303 with men on base. In 2006, in a full season at double-A, opposing hitters hit .226/.276/.333 with no one on base and .220/.246/.358 with men on base. Granted, these numbers came against minor league hitters; however, if Sonnanstine experiences a distinctive change in his ability to pitch when runners are on base, such a change would likely show up against minor league hitters as well as major leaguers.

Perhaps most interestingly, Sonnanstine seems to have Fausto Carmona syndrome: Sonnanstine gets drastically different results depending on whether the first pitch of an at-bat is a ball or a strike. Last season, if the first pitch was a strike, batters hit .241/.254./.392 off of him. However, if the first pitch was a ball, opposing hitters mashed to a tune of .335/.404/.606. Although Sonnanstine was very good at throwing strikes (he threw 66% of his pitches for strikes last year), he did not throw as many first-pitch strikes – only 60% of batters received a first-pitch strike. Considering how important Strike One is for Sonnanstine – and how good he is at throwing strikes in general – this an area in which we can expect improvement.

To sum up: Sonnanstine’s peripherals belied his 5.85 ERA from last year. He suffered from a terrible defense behind him, gave and gave us a largely disproportionate amount of hits with runners on base (even though he had not shown any decreased ability with runners on base in the minors). Finally, he was a tremendously better pitcher if he threw his first pitch for a strike – something he did less often than we can expect in the future, considering how much of a strike-thrower he is.

So what about this year? The Rays have drastically improved their defense. So far, however, Sonnanstine sports a 5.55 ERA – and that’s including a complete game shutout against the White Sox. Defense is not the problem – Sonnanstine has allowed only a .256 BABIP this year. His K/BB ratio is an excellent 13/4 through 24 innings. So what’s the problem?

To start with, Sonnanstine has given up five homers already. However, this can be attributed to the fact that he’s giving up an inordinately high number of homers per fly ball (22.4% of all fly balls are becoming homers, as compared to 10.8% last year), and he can expect some regression in this area. Also, two of Sonnanstine’s four starts have come against the Yankees – and he has given up four of his five homers against the Yankees. Pitching in the AL East, Sonnanstine can expect to face the Yankees often, but it’s unlikely that he will have to pitch against them in half of his starts.

Some of the trends from last season have continued in to this season. For instance, Sonnanstine has still given up many more hits with runners on base than with no one on base. With none on base, batters are hitting .267/.283/.644; however, with runners on base batters are hitting .455/.480/.727. Because he has only had four starts this year, the sample sizes for these stats are extremely small; normally, I’d write these splits off to complete luck, but in the light of similar splits last year, we have to wonder if something is going on here.

Sonnanstine is also continuing to show much more ability to get batters out if he delivers a first pitch strike. This year, if the first pitch of an at-bat is a strike, batters are hitting .206/.206/.412; whereas if the first pitch is a ball they are hitting .423/.483/.846. Again, the sample size is tiny, but the splits are once again a continuation of what happened last season.

It does appear that Sonnanstine is throwing more strikes, both overall and on the first pitch. So far this season, 70% of his pitches have been strikes, and 64% of his first pitches have been strikes. In his shutout against the White Sox, 21 of 29 (72%) batters received first pitch strikes.

So what can we expect in the future? To be sure, Sonnanstine does not have dominating stuff, and will always therefore be relatively homer-prone. Plus, pitching in the AL East is extremely difficult for anyone, especially for pitchers without dominating stuff.

Perhaps the most curious thing about Sonnanstine, however, is his splits with nobody on base as compared to runners on base. If Sonnanstine was having mechanical issues with pitching out of the stretch – or if he was simply not able to locate as well or throw as hard from the stretch – it’s likely that we would have seen some evidence of this in the minor leagues. I’m not saying he’d be awful out of the stretch in the minors, but he would probably be worse than he was from the windup. However, he was essentially exactly the same in the minors with guys on base as he was with no one on base. Yet in the majors thus far he has been tremendously worse with runners on base. What do we make of that?

I posit that Sonnanstine’s problems with runners on base can be attributed to a combination of bad luck and a lack of confidence. His sample size in the majors is still fairly small, and some of the differences can stem from pure luck (or lack thereof) concerning in what situations batters are getting hits.

Additionally, as I’ve mentioned several times, Sonnanstine doesn’t throw hard and has rather underwhelming stuff. Thus, it’s possible that he is more afraid to challenge hitters when there are men on base. We know that throwing strikes – especially first-pitch strikes – is essential for Sonnanstine’s success, and if he does this less often with runners on base (for fear of challenging hitters), he’s going to get himself into trouble very quickly.

Sonnanstine has demonstrated that he has the ability to get major league hitters out. He needs to throw first-pitch strikes, but there is no reason to think that he can’t do this very often. If he can begin to retire batters more often when there are men on base – and, given his track record, he should be able to – Sonnanstine could be a very solid #3/4 starter for a good team.