Pari-Mutuel Betting = No Good

Robin Hanson (on Foretal):

This is parimutuel betting, and a big problem is that people have an incentive to wait until the last possible moment to make their “votes” (= bets). Waiting lets you take advantage of new info that comes out, and look for mis-pricings.

About Chris F. Masse

Founder and President of Midas Oracle
This entry was posted in Analysis (Meta), Forecasting (Science & Practice), Mechanism Designs and tagged , , . Bookmark the permalink.

3 Responses to Pari-Mutuel Betting = No Good

  1. Parimutuel betting = No Good?

    I guess that’s why betting on horses is so unpopular…, except, of course, billions of $$$ every year are bet on horses.

    Of course what works for very short term markets — completed in a matter of minutes after betting begins, or at most a few hours — may not work for a longer term market. So, parimutuel betting = no good for longer markets, but apparently just fine for short term markets.

  2. Robin Hanson says:

    Michael, yes, parimutuels can do well if there is strong interest and if they are controlled to only be open during a period when relatively little information is revealed. Thus turns out to hold for the hour before a horserace, but is unlikely to hold for the actual example claims now shown at Foretal.com

  3. The basic premise here is correct. We are required by Malta’s LGA to limit cash voting to predictions with a proposed outcome date less than 24 months away, but aside from that, there’s very little incentive for people to tie up any cash for weeks and certainly months in advance. So, cash voting is likely to be for short term predictions certainly.

    We are also able, according to our own rules, to determine the outcome time and date based on the substantiating sources that must be included in the prediction proposal. There is a mandatory 24 hour period after we close voting on a prediction during which we can verify this. If people cash vote after an outcome had happened, but before we had closed the prediction and determined the outcome time, their cash votes are simply returned without penalty.

    However, our starting point is slightly different from pure gambling. The free voting mechanism is designed to engage and stimulate interest in the site, while incentivising people to make their own predictions. As such, the collective data we gather overtime becomes potentially significant and revenue generating in other ways.

    And while we are starting out with the simplest of cash voting formats, there is nothing to stop us offering different cash gambling formats over time for Foretal members with different levels of experience and expertise.

    Why not make your own prediction about this on Foretal ;-) ?

Leave a Reply