One interesting observation on the bets. While more money was piled on "Remain", more individual bets were placed on "Leave", at least at Ladbrokes. A rare instance of those placing the bets also being able to affect the outcome (by voting) so there was information content but it didn't fit the conventional view of smart/dumb money.
I raised that point in an earlier post when I questioned the wisdom of using amount wagered rather than number of bets to determine bias in a 1 person 1 vote situation. There may be smart money and dumb money, but votes are all equal, and the chap who clears Cameron's garbage has a vote equal to his. I can't remember the details of the paper I read on this subject years ago, but surely there is enough data to analyse which has better predictive value? Or has that already been done and this was a genuine anomaly?
This is a pretty interesting question. I have a ton of experience with sports betting, where money bet and by whom has no affect on the outcome of an independent event, at least in theory. I also have a little, by comparison, with financial markets where money bet actually drives the outcome. However, I have zero experience where the actions of all participants has an equal affect on the outcome, and they are also simultaneously betting on the outcome. I haven't seen any volume numbers on the sides from the big bookmakers, but I assume they made money. The suckers (dumb money) were they guys, no matter how rich and smart they are, who were willing to lay 4 for 2 on a 50/50 prop where polling was inside the margin of error. The end result, known only in hindsight, is that the bookmakers got them to take a bad price, and the long odds they offered on Exit weren't enough to offset their profits on the Remain. They did a great job, people need to stop relying on betting markets, which has become fashionable among the "half sharp", for making predictions. They aren't trying to predict anything. They are trying to set a price that strips Expected Value out of their offerings, and puts the player in a long term negative position. They succeed and fail in this endeavor a lot, but their is no predictive magic in the pricing process.
My memory of that paper is really fuzzy, but I seem to recall something about the buzz over the wisdom of crowds. You make a very valid point about the bookmakers self-interest in offering odds, and yet after all it is human nature to react to the odds offered as much as the expected outcome. I regularly pay less than $3 for a lottery ticket here that offers a first prize of $170K. If I lose, I don't even realise it, if I win the asymmetric return outweighs a lifetime of losses. I know there are greater odds I will be struck by lightning on my weekly trip to the supermarket, but the asymmetry outweighs such silly reasoning. It would be pretty nonsensical imputing anything from the fact I'm indifferent to losing less than $3 twice a month. There is surely such influence on betting, if not by number of wagers, then by amount wagered. I suppose if you look hard enough you can find something in anything.
It's my experience in terms of political bets, analysts are really just watching the "trend" in the bet, not the actual bet. There is some predictive value in this at least here in the states. In 2004 the election was called for Kerry as all data was pointing strongly in his favor in all the swing states. What they were ignoring was the "trend". Bush started trending the final days even though he was far behind. This "trend" gave us some insight into how the late deciders were going to break, in other words, there is some value there measuring sentiment during meaning periods in the cycle. In the Brexit vote I wasn't watching carefully the trend as it appeared to be all over the map. If anything, the late votes probably trended in the direction of remain if I recall correctly.
Wisdom of the crowds is an indeed important predictor of future events. What it has no bearing on is the value of the actual price being offered to express that opinion. Bookmakers have a tough job, compared to say casino operators, in that the true value of any side is pretty nebulous. That is why the "juice" is so much higher in sports/election betting than in say blackjack or baccarat. They need the padding of those big spreads, just like a market maker who increased his in times of great volatility/low liquidity. At a certain level, Leave money would have overwhelmed Remain, given a certain price. Who knows? Maybe 20-1, 30-1, surely 50-1. My point is any bookmaker worth his salt balances his risk while maximizing profit. The fact that they were heavy on the losing side at a short price, while under played on the winning side on a long price, simply is evidence they did a great job trading this one. But then again, LOL, sample size. It's not like there are a ton of trials when dealing with election bets. So maybe every single participant got it wrong from a long term EV stand point and we will know in 1000 years. One Presidential Election every four years, 30 base ball games everyday, pretty sure which market they are better at pricing.
Well, a lot of guys I know here in Vegas thought it was ridiculous they could get 9-1 on Exit late Friday considering how tight the actual polling was. A few had outs overseas and played it small as a throw away. I couldn't because election betting is illegal in the US, even in Nevada, lol. Got a couple hundred off at 8 with buddies.
There were some great trades. I think what made the setup really attractive was the rally going in, that changed everything from a risk standpoint. My thesis from the beginning was when we were sitting on the lows at 2040 there was nothing to do, the market had sold off and if the result was "remain" we would get a rip roaring rally. If "leave" then massive selloff. However, we got an 80 handle rally into the event which "greatly" reduced the upside risk and made the downside asymmetric. That changed everything for me. I could lean into it accepting maybe a quick pop on "remain" with a sell the news after. And on the off change we got "exit", the downside was wide open. I really do believe the only reason we saw that much action on the downside was because of the squeeze going in. However there were really good option set ups, some of them I got, some I tried desperately to get but could not get fills because of the lack of going in.
Pretty much only hang out with gamblers, but I felt this way very strongly as well. At 1.50, the Pound was pretty much exactly where it "should be" on a Remain vote. I had no options account and was begging friends to play 1.45 puts Friday afternoon. I thought that was a much better position than doing a Binary "Exit" bet. The idea crystalized for me when I started to consider the upside risk of an event that was being viewed already as an 87% certainty. Pretty sure you could have shorted the Pound up there, had the vote go "Remain", and still made some "sell the fact" money.