latest Betfair odds

[Craigo]

Sanders has the highest implied odds in the general election, 64.5%.

Huh?

Winning individual odds / primary odds

That only works if the market is efficient - which it clearly isn't, as the individual Democrats' chances of winning the WH add up to more than the Democratic Party's chances of winning the WH.

[Craigo]

Sanders has the highest implied odds in the general election, 64.5%.

Huh?

Winning individual odds / primary odds

That only works if the market is efficient - which it clearly isn't, as the individual Democrats' chances of winning the WH add up to more than the Democratic Party's chances of winning the WH.

Do they not allow short selling?

They might. Go make some money if they do (or not, from your state icon - Betfair is illegal for Americans).

[Craigo]

Sanders has the highest implied odds in the general election, 64.5%.

Huh?

Winning individual odds / primary odds

That only works if the market is efficient - which it clearly isn't, as the individual Democrats' chances of winning the WH add up to more than the Democratic Party's chances of winning the WH.

Well there aren't any actual odds here in a scientific sense, even if the market is efficient, as it's a non-repeatable event whose result we can only guess at in a Bayesian sense. Also I imagine there are many people betting that don't even understand that the odds have to add up like that (much less consciously considering that when betting), adding a further wrinkle as to how they should be interpreted. So I think his metric is as good as any basically, given the data there is to go on.

I agree with everything but the bolded sentence. Bad methodology doesn't magically become good just because a better method is not currently practicable. His method is probably the best that you can do with this data without some hardcore math that I left behind in grad school, and I don't blame anyone for wanting to divine that information - but it's still not good. We just don't have the data needed to answer that question with any confidence.

[Craigo]

Like I said earlier, the numbers I'm posting are "Betfair Exchange" numbers, which AFAIK works pretty much like the old Intrade did, in that every transaction is by definition a trade between who's buying an event and someone else who's short selling the same event.  This is in contrast to the vanilla version of Betfair, where the price is set by bookies.

So no, short selling isn't disallowed.  It's *required* that every trade involves someone short selling, to counteract the person buying.

That said, you're right that the volume doesn't tend to be high enough to make the market perfectly efficient.  On the "winning individual" market, a given candidate might go several days without having any shares traded.  Thus winning individual is more likely to lag in certain trends as compared to the nomination markets.

Thanks for that info. I didn't know the volume numbers, good to know.