Predictions from betting quotes adjusted for money-flows Recently at least two predictions were made based on (pools) of bookmaker quotes 
one for the UEFA Euro 2016 (see here and Zeileis A, Leitner C, Hornik K (2016). “Predictive Bookmaker Consensus Model for the UEFA Euro 2016” for the paper) and one for the Brexit referendum.
The former predicts the winner of the Euro 2016 as France and can still be true (I am aware that the predicted probability is much less than 100%) and the others were rather wrong predicting Britain not to Brexit.
My question: isn't it true that bookmakers adjust their quotes such that if a lot of people bet on the event A the pay-out in case of event A is reduced - which could look as if A were more probable. 
If this is the case then not only the quotes of bookmakers (professionals) but also the betting behavior of the crowd (not all pros, maybe biased by media etc) influences the quote and therefore the estimated probability of the event.
How is this quote-adjustment to money inflows accounted for in these predictions?
 A: Isn't it precisely the function of a prediction market? Participants wager on the likelihood of a future event. And the distribution of the outcome rewards those with the most accurate prediction[1]. Bookmakers respond to the amount of wager by adjusting the public odds and that's what you see out there.
[1] John M. Gandar, William H. Dare, Craig R. Brown, and Richard A. Zuber. Informed traders and price variations in the betting market for professional basketball games. Journal of Finance, LIII(1):385–401, 1999
A: If you assume:


*

*Market efficiency (i.e. prices reflect all available information)

*Investors are risk neutral (i.e. they maximize expected value)


Then the probabilities you back out from betting markets will be the true probabilities that events will occur!
If you drop the 2nd assumption:
Then the probabilities you back out will be what's referred to in finance as probabilities under the risk neutral measure. That is, the probabilities will be somewhat tilted versions of the true probabilities, and that this tilt reflects covariance with variables of hedging concerns to investors.
If you drop the 1st assumption (your scenario):
Then you can't say much of anything at all. If prices are totally arbitrary, reflecting merely the whims of non-sensical punters, then you can't use betting markets to estimate the probabilities of events.
A: I have to answer my own question...
In the paper on the Euro 2016 they write on page 2:

In order to forecast the winner of the Euro 2016, we obtained
  long-term winning odds from 19 online bookmakers (see Table 3 at the
  end). However, before these odds can be transformed to winning
  probabilities, the stake has to be accounted for and the profit margin
  of the bookmaker (better known as the “overround”) has to be removed
  ...

One can read the assumptions and formulas for this overround in the paper.
Thus: if it is done correctly then the probabilities can be found under certain assumptions.
