What if Mariano Rivera did start?

During spring training the baseball community was given the news we all knew was coming: 2013 would be the last season for Yankees' closer Mariano Rivera. At 43 years old and coming off of an ACL surgery that caused him to miss the majority of last season, it would have shocked no one if Mo struggled during his final campaign. Yet once again he is among the best relievers in baseball, posting a 1.47 ERA and racking up 17 saves. Rivera will certainly earn a trip to the All-Star game this year, and as many of you may have seen Bill Chuck has started a movement to let the famed closer start the game for the American League team. While his role in an exhibition game doesn't matter much to me, the movement has made me wonder about his usage in New York. How would Rivera fare as a starter at age 43?

It's no secret that pitchers can provide a much greater impact and more value to a team as a starter than as a reliever. So it's understandable that we are seeing more and more attempts to transition good relievers into starting pitchers. The majority of these moves do not work, at least not in developing a really good starter, but there are success stories like C.J. Wilson and Chris Sale that make the idea a worthwhile risk. Unfortunately figuring out just how a pitcher will perform after the switch is easier said than done.

That's not to say there hasn't been research on the subject. Just this past October, our own Alex Kienholz wrote about some of the best pitchers that have made the switch, and not long before him, Grantland's Jonah Keri weighed in with his thoughts. Jonah's article contained this table from ESPN Stats & Information looking at pitchers over the past five years that spent two or more seasons exclusively (or nearly exclusively) as relievers before making 20+ starts the next season:

*Samardzija was added

Now I apologize if it seems I have gotten off track, but to make any sort of projections for Rivera's hypothetical performance as a starter, we need some sort of baseline. Unluckily, the table above gives us a tiny sample that is made even more useless by selection bias. Teams won't try the move with bad pitchers, and they also won't try the switch unless they feel reasonable confident the pitcher can handle it. So where does that leave us? Well Keri's article noted a Dan Szymborski study from 2003 that looked at pitchers switching from the bullpen to the rotation and vice-versa dating back to 1920. He found that the relievers struck out more hitters, allowed fewer home runs, and posted an ERA that was almost a full run lower (3.46 to 4.31) than the starting pitchers. Both Alex and Jonah also referenced this 2009 blog post by Tom Tango, which urges us to use the "rule of 17" for such cases:

I went ahead and calculated the wOBA for all the starters/relief, and for pitchers since 1993, it was .352 for starters and .323 for relievers, a gap of 29 wOBA points, almost identical to what I found in The Book. The relievers had a wOBA that was 8% less than starters. When I looked at the whole dataset (1953-2008), it was also the same 8%. Basically, this reaffirms the 1 run per 9IP gap I've been using.

Basically, use the "rule of 17": difference in BABIP is 17 points higher as starter. K/PA is 17% higher as reliever. And HR per contacted PA is 17% higher as starter. Walk rate is FLAT.

If the Yankees made such a move, Rivera would certainly be limited in his innings this season, probably between 150 and 160 innings. Since the beginning of 2010, Rivera has pitched 148.0 frames, with a 1.82 ERA, 2.49 FIP, 3.15 xFIP. His strikeout rate and walk rates over the same time are 22.5 and 4.1 respectively, and his BABIP was .249. Applying the trends from Szymborski and Tango's findings, we would expect Rivera to be one of the better starters in the American League, with an ERA between 2.85 and 3.50 over his 150 innings.

Sadly this is mostly guesswork though, as Rivera is a pitcher much different than most. He relies on one pitch, his cutter, throwing it over 95% of the time, making the challenge of turning over a lineup that much more daunting. He also has pitched out of the bullpen for his last 1,060 games, a much more extended time than any of the pitchers in the sample.

But this is no ordinary pitcher. In fact, as Jason Stark stated recently, we may be watching the greatest season by any 43-year old professional athlete ever, and that's without even mentioning that Rivera's knee operation. Perhaps using any rational method of projection for Rivera is foolish. The man continues to amaze, and I would bet he could be worth 4-5 wins as a starter based solely on the principle that one should never bet against Rivera.

Of course the Yankees won't move Rivera to the rotation, so we will never know the answer. All we have are the 10 games he started way back in 1995 and 20 years of dominant performance in the bullpen. So while it certainly is fun to think about the what-ifs, I'm just happy to get one more year of watching Rivera pitch regardless of his role.

Andrew Ball is a writer for Beyond the Box Score, Fake Teams, and Fantasy Ninjas.

You can follow him on twitter @Andrew_Ball.