It's fun, but when position players pitch, bad things usually happen

Last Saturday's 20-inning thriller between the Mets and Cardinals featured one of the most unusual sights you're likely to ever witness on a baseball diamond: a position player on the mound with the game on the line. There's nothing quite like seeing an outfielder or a shortstop atop the hill, and as fun as that is to watch in a laugher, it's even more interesting when it occurs in a pressure situation.

Saturday night's classic featured three innings of work from Cardinals position players Felipe Lopez and Joe Mather. While the Cardinals managed to get by in the 18th and 19th, their luck ran out in the 20th inning. While one has to expect the unexpected from non-pitchers, it happens often enough that we can get a pretty good idea of what might occur by looking at the statistics. Of players who played in the 1980s, 1990s or 2000s, there have been a total of 120 non-pitchers who have taken the mound. Combined, they've pitched a total of 182.2 innings. And just how bad was the carnage during those innings? The non-pitchers gave up a total of 162 earned runs, good for an 8.87 ERA. While this is obviously worse than any "real" pitcher, all things considered, less than one run given up per inning isn't all that bad.

An opposing hitter might look forward to facing a non-pitcher with glee -- after all, it's a great opportunity to pad his stats. An at-bat against a non-pitcher seems like an awfully good place to look for a cheap home run, and fans might expect to see a power show when a position player takes the mound. However, while it would seem that facing a steady stream of batting-practice fastballs would turn pretty much anyone into a long-ball threat, non-pitchers don't give up homers at the astronomical rate that you might expect. While it's true that they do give up more homers than average, don't expect a slugfest when a position player trots in from the bullpen. At .039 homers per at-bat, this mark is actually better than the rates that some pretty respected pitchers posted in 2009. Joe Saunders, Rich Harden, Ervin Santana, Jamie Moyer, Jeremy Guthrie and Braden Looper, among others, all gave up more homers per at bat than the average non-pitcher.

The real difficulty with non-pitchers is their stuff and their control. Predictably, non-pitchers aren't fooling a lot of hitters. They do manage to strike out, and thus embarrass, 7% of batters they face, but this is well below the 18% strikeout rate that the average pitcher posts. That means more balls in play and more hits against them. It's possible, of course, to succeed without being a strikeout king. John Lannan for example, struck out just 10% of the hitters he faced in 2009. But to have that success would require exquisite control -- the biggest hurdle that any non-pitcher faces. While real pitchers walk 9% of batters, non-pitchers walk nearly double that amount, 17%. These bouts of wildness could be overcome by great stuff, but non-pitchers can't compensate for walking hitters like a Clayton Kershaw or a Carlos Marmol can.

Any one of these traits might be overcome on their own, but someone who has the strikeout prowess of Lannan, the control issues of Marmol and gives up homers like Looper is going to give up an awful lot of runs. And thus, it's not surprising that things can get pretty ugly with a position player on the mound.

So, knowing this, when is an appropriate time to put that backup catcher or fifth outfielder on the hill? Obviously not with the game on the line. While 120 pitchers have taken the mound, only six have factored in a decision: Jose Oquendo (1988), Jeff Hamilton (1989), Brent Mayne (2000), Jaime Burke (2008), Josh Wilson (2009) and Mather on Saturday. All of these players came into the game in similar situations -- an emergency after the team had burned its usable pitchers late in extra innings. Mayne actually took the mound after the Rockies' last pitcher had been ejected. It's no coincidence that five of the six pitchers, all of whom appeared in the 15th inning or later, picked up the loss, with only Mayne pulling out a victory.

Tony La Russa was the only manager in those six games to have two position players pitch. Cards fans probably figured that they were in for certain defeat with a non-pitcher on the mound, cursing La Russa for pulling Ryan Franklin after just one inning of work and burning through all seven of his relievers, but while the move certainly didn't put the odds in St. Louis' favor, it wasn't a sure game-killer either. An average regular pitcher can prolong the game by pitching a scoreless inning about 72% of the time. However, a pitcher as bad as the one described above -- with an ERA pushing 9.00 -- can still be expected to hold the opposition scoreless 57% of the time. While it's certainly a major downgrade, position players are very capable of throwing up zeroes and keeping their team in the game. These non-pitchers are at a much higher risk of giving up big innings, but of course, in a tie game in extra innings, this isn't really a concern. So, for as much despair as Cards fans felt when Lopez's name was announced over the Busch Stadium PA, these non-pitchers actually gave the Cardinals a fighting chance in extra innings.

No matter your team affiliation, it's hard not to root for a position player on the mound. He represents the ultimate underdog -- someone who has no business being anywhere near a pitching rubber who is thrust into the game and forced to get out major league hitters. Most fans have at one point or another pictured themselves on a big league mound staring down a slugger, despite the fact that they haven't pitched since Little League. In a small way, that fantasy is fulfilled when we watch another guy who also hasn't pitched since Little League attempting to do the same. The scenario becomes that much sweeter when the game is on the line and the position player can pitch his team to victory. On Saturday night, Lopez and Mather almost did just that.