So far, the Kentucky Derby draw format of trainers/owners selecting their own post positions hasn't generated much in the way of meaningful strategy.

Two years ago, Rick Pitino chose No. 4 for his Halory Hunter because that was the uniform number of Lou Gehrig, who happens to be the native New Yorker's "all-time favorite ballplayer." The process by which trainer Bob Baffert studies post-position possibilities is no intellectual pursuit, either. "We go out and have a few beers," he said.

At least Baffert enjoys his method. Judging from facial expressions, some other participants behave as if they have just discovered they're a target of an IRS audit.

Perhaps the whole procedure is much ado about nothing. After all, last year Charismatic drew 18th in the selection order, and Menifee was 19th. They finished one-two. But just in case there is a correlation between post position and performance, let me present a scientific analysis.

The inspiration for my attempt at advanced mathematics was a Daily Racing Form article last year that challenged the traditional way of measuring post-position success. It isn't good enough simply to know that post No. 1 has a 20 percent win ratio at a mile-and-a-sixteenth compared to only 5 percent for post No. 10, since those stats inherently favor inside positions. That is because in a five-horse field, post No. 1 - or any post, for that matter - would according to raw probability have a 1-in-5 chance of winning, or 20 percent, while a horse breaking from No. 10 has at best a 1-in-10 chance, or 10 percent.

In 1948, Citation started from the rail and defeated a six-horse field. In 1956, Needles started from the rail and beat a 16-horse field. Yet, even though Citation outran only five horses and Needles had to beat 15, most methods of post-position analysis give each victory the same credit.

(If you're already lost, please hit the "back" button on your browser, because you'll have no chance to follow the rest of this.)

With a variation of field sizes in modern Kentucky Derby history ranging from six horses to 23, this mathematical quirk might seem at first glance to be the Rubik's Cube of racing. But here was my logic: assign points for Derby finishes based on field-size probabilities of post positions.

Sound confusing? Try this. In a 20-horse field, each horse has a 1-in-20 chance of winning, or 5 percent. Conversely, each horse has a 95 percent chance of losing. Therefore, in each 20-horse field, credit the winning post position with 95 points. Using the same math, credit the second place finisher with 90 points (a 2-in-20 chance of a top-two finish, or 10 percent, and a 90 percent probability of running worse than second).

Using that formula, I gave points for each top-six finish in every Kentucky Derby since 1930, the first year a starting gate was used.

The most productive post position, believe it or not, is No. 18, with an average of 48.3 points per starter. Of course, if a trainer had the first choice of post positions and took No. 18, he might be immediately fired by his owner. But Gato Del Sol won the Derby in 1982 from the 18 hole, and Menifee (1999) and Hudson County (1974) ran second from that post. In fact, six of the last nine runners from No. 18 have managed top-five finishes. However, only 15 horses have started from No. 18 in the 70 years of the test, so perhaps the sample is too small for any horseman to risk public ridicule.

But the second-best post by the numbers is conclusive: post No. 10, with 45.6 points per starter, a full 10 points better than the rest. If I had a horse running in Saturday's Derby, and I had first crack at post selection, I'd take No. 10 in a heartbeat.

I asked Chris McCarron, the Hall of Fame jockey and two-time Derby winner, for any reason why the middle would be the post of choice.

McCarron, who rides The Deputy on Saturday, shrugged his shoulders. "I know I don't want No. 1, and I don't want No. 20," he said. "Beyond that, I'm not sure how much it matters."

If the rail is terrible, and the extreme outside is disastrous, maybe it makes perfect sense that 10 would be tops.

Each of the last two years, the trainer with the first post-position selection has taken No. 7, and indeed, that post is next on the scale with 35.6 points, followed by No. 5 with 35.1 points. But after that, a surprise, at least to some: posts No. 1 (34.2 points), No. 2 (34.1) and No. 3 (34.0).

The worst post positions based on the numbers are 19 (ranking last with 12.6 points), 15, 20, 12 and 6, the latter's poor ranking something of a puzzler but clear-cut nonetheless.

Don't expect many owners or trainers to use this logic when selecting post positions on Wednesday. But it couldn't hurt, because their current methods aren't working. A year ago, none of the first nine selectors managed a top-four Kentucky Derby finish.

Randy Moss joined ESPN in 1999 as its main horse racing analyst. Prior to that, he reported for a number of newspapers including the Fort Worth Star Telegram and the Dallas Morning News.