By the Numbers: One slippery Bannister

This past Sunday, Brian Bannister threw a complete game against the Twins, giving up only one unearned run and three hits. It was an impressive feat, but nowhere even close to the most amazing thing about the Royals' righty.

Last season, Bannister did something improbable. He compiled a 3.87 ERA and 1.21 WHIP while striking out a measly 77 batters in 165 innings. Bannister made Jamie Moyer look like Johan Santana, yet it was Moyer and not Bannister with the 5.00-plus ERA. Bannister's above-average stats were largely built on the foundation of a 26% H/BIP (hits per balls in play) rate, which is exceptionally low for a starter. Still, this is not what I find incredible about Bannister.

Brian Bannister's success is as impressive as it is improbable. (US Presswire)

He not only knows what his H/BIP was last year, but contrary to what the research suggests, he thinks his low rate was neither random nor accidental. Bannister thinks he knows how to keep hit balls from becoming base hits. It's not that Bannister is ignorant of the research findings; he is actually a student of them. Through his own statistical research, he concludes that his H/BIP, ERA and WHIP could easily remain at their 2007 levels. This is what blows me away about Brian Bannister.

Everyone knows that major leaguers would never resort to the level of statheaded geekiness that you sometimes find in Fantasy leagues and baseball blogs. Read what Bannister had to say in an interview on one of those blogs -- MLB Trade Rumors -- and think again.

"I think a lot of fans underestimate how much time I spend working with statistics to improve my performance on the field ... I don't claim to be able to beat the .300 (H/BIP) average year in and year out at the Major League level. However, I also don't feel that every pitcher is hopelessly bound to that .300 number for his career if he takes some steps to improve his odds -- which is what pitching is all about."

Bannister revealed his secret for success in the interview; he actually tries to manipulate his H/BIP rate with his pitch selection, inducing weak swings, particularly in two-strike counts. That's why he knows his H/BIP. It's an indicator of how well his pitching strategy works.

Now that's an innovative use of statistics.

If there is truth to Bannister's claim that a crafty pitcher can prevent balls in play from becoming base hits, then Fantasy owners can also take advantage. Low-strikeout guys like Bannister may suddenly provide us with a hidden source of value. We could punt on Ks entirely, or at least look to only three or four starters to help us in that category, while stocking up on low-strikeout pitchers who can help with wins, ERA and WHIP. That would free us up to use high draft picks or auction dollars to address other needs.

Below is a list of every starting pitcher last year with at least 50 IP who struck out no more than 4.5 batters per nine innings. By looking at their skill indicators, H/BIP rates, ERC (a better indication of performance, and therefore, future ERA, than ERA itself) and WHIPs, we can see if the Bannister Theory can help us to find effective bargain pitchers. We see that Bannister and Jon Garland stood alone with sub-4.00 ERCs, and in fact, ERC says that Bannister's ERA should have been a half run lower than it was. In the next tier, Aaron Cook, Carlos Silva, Braden Looper and Kyle Kendrick all delivered ERCs in the low 4.00s and WHIPs below 1.35.

If Bannister's theory is right, then these six pitchers can succeed without strikeouts simply by keeping their H/BIP low. However, Bannister was the only one in this group to have an H/BIP below 27%. Looper was the next closest at 28%. It seems that if a low H/BIP is the key to success for low-strikeout pitchers, then Bannister was the only major league pitcher last year to have figured this out.

What, then, could be the secret of success for Garland, Cook, Silva, Looper and Kendrick? The one common denominator is outstanding control. All five of these pitchers walked 2.6 batters per nine innings or less. Nearly all of the pitchers on the list with average or above-average ERCs lacked this level of pinpoint control. Those who did have excellent control but high ERCs, namely Paul Byrd, Chris Sampson, Bobby Livingston and Zach Duke, all had serious problems with the long ball.

This analysis tells us a few things about low-strikeout pitchers and Bannister's theory. First, the conventional wisdom that low-strikeout pitchers are poor candidates for success appears to be true. The Bannisters and Garlands of the baseball world are rare exceptions to this rule. Second, excellent control is a necessary, but insufficient, condition for a low-strikeout pitcher to succeed. Third, a low H/BIP rate is not necessary for a low-strikeout pitcher to do well, but a high one is the kiss of death (see Livingston and Duke).

Finally, if there is a way for a pitcher to consistently keep his H/BIP rate low, the evidence for it is not here. None of these pitchers managed to post even a sub-29% H/BIP in consecutive years. Actually, if you include data from seasons where a pitcher threw fewer than 50 innings, there was one exception. Naturally, it's Bannister. He recorded a 25% rate in 2006, but he did it over only 38 innings.

Maybe Bannister has figured out how to stymie big league hitters without striking them out. Until we know for sure and other pitchers start to figure it out as well, Fantasy owners should still give priority to pitchers with a well-rounded skill set that includes strikeouts. If quality strikeout pitchers are in short supply, though, getting a low-strikeout pitcher with a track record of great control, like Garland or Silva, is a good alternative.

Al Melchior was recently a Fantasy columnist and data analyst for Baseball HQ and will be providing advice columns for CBSSports.com. Click here to send him a question. Please put "Melchior" in the subject field.