Horse Racing Does the Wagering Economics Thing, Kinda Sorta

The Powerball was won! For those of you who took an 8-27-A-A-A wheel because the 8 had the top pace figure and the 27 was sneaky good the last couple, you played the wrong game, but congrats.

The Powerball lottery we saw over the last couple of weeks should be studied with a grain of salt when it comes to gambling behavior, of course; but there are concepts that explain why we had a case where an impossible to hit bet had otherwise fairly rational people standing in line for hours, wanting to give the state their money.

In general, gamblers trade off the expected return of a game for skewness. If it's 1 in one trillion to hit something, but the payoff is some obscure amount, like $1 billion or more, hey, you take a shot. This convex function is shaped differently because of the the size of the payout.

For horse racing, this also helps explain the lottery bet attitude. Why do people bet 52% takeouts for a twenty cent bet they won't hit, with that terrible expected return? It's simple skewness; a mini-powerball.

Horse racing is much different than a lottery however - it needs its customers betting more frequently, can't bust them, and has to offer sound, rational bets. The higher the probability of a bet, the less skewness and the more rational the behavior is. For example, a person on twitter who just took a Rainbow Six ticket with no hope in hell might tweet about a win bet in leg one, "3-2 is too low, if he goes up to to 2-1 I am betting." That's the difference between a 40% probability of cashing to 33%, and it completely changes behavior.

Carryovers provide an excellent example of the best of both worlds.

There are irrational bettors betting into a pick 4 carryover, with say an expected $500,000 pool, because the pool is $500,000. The chances of one ticket paying $500,000 is like getting hit by lightning three times in one day, but that doesn't matter to this user. He or she bets because of this hope (and then wonders why they got so excited when their spread ticket paid $54).

Meanwhile, me and you are betting  five or ten times what we'd normally bet into this pool, because with a $50,000 carryover we are getting negative takeout. We are acting like any rational bettor would. The money carryovers take is not a mystery.

Powerball maximizes their skewness well, as we all saw. But horse racing doesn't maximize it skewness very well at all.

Win takeouts, where the probability of hitting a bet is in the 35% range should have 5% or 6% juice to take advantage of the same skewness. Two horse exotics should not be a heck of a lot higher. Generally, as the hit rate goes down, the rake can go up, where with super high fives, or pick 6's, 25% juice can be just fine. If a lottery component is added, even over 30% takeout is workable.

Horse racing, unlike Powerball, is big tent, with open arms to many people who wager. It's really a tremendous asset. That the sport tends to shoehorn everyone into one system, with pricing and promotion, is one of its most frustrating failings.