Win-Show Mutual Pool Overlays - The Ziemba System

[PairTheBoard]

Around 15-20 years ago a book came out authored by a pen-name "Ziemba" I believe, in which he claimed an edge could be achieved in Racing where there are mutual pools for Win-Place-Show. The dynamic involves the observation that people sometimes are reluctant to place bets where there are very short oddss payoffs. If this is happenning it can be theoretically measured in the case of Racing mutual pools. For example, the win pool for a favorite horse is 1/3 the total win pool for the race, giving the favorite 2-1 odds to win. Now, assuming market efficiency for the win pool at these odds, an efficient show pool should also see 1/3 of the total pool being bet on that same horse. Yet often times, especially in the later races of the day, there will be a much lower ratio bet on the favorite to show than to win. For whatever reason, maybe people are looking to get even for the day, sometimes as little as 1/6 of the total show pool is bet on this horse. When there is that much of an overlay you should theoretically be able to gain an edge by betting this horse to show. The method is to wait until just before post time and when you see the divergence in pool ratios, place your bet.

Ziemba claimed to have enough data from one session's results from one track to prove the system works. Theoretically, you could employ this method sitting in a Casino's Race Book if the Televisions were flashing the mutual pools close enough to post times.

Has anyone actually worked this method and made long term profits from it? I tried it a little years ago with dismal results. Offtrack there is too often a large change in the final mutual pools compared to the last ones flashed on Television. I've also thought there might be a problem due to the possibility that when the favorite doesn't have it that day and it's clear he's having little chance to win, the jocky may let up and not ask him for all he's got just to beat out horses for 3rd.

Any experience with this here?