Buy Now Or Forever Hold Your Peace

The Intrade prediction market is giving Hillary a 53% chance and Obama a 47% chance of winning the Democratic presidential nomination.  Hillary is down 7.5 percentage points in just the last day.  (Note:  Between when I wrote the above, and when I posted this, Hillary went up to 54.)

From what I've read on Intrade, you can fund your account with up to $250 using a credit card, and it should land in your account immediately.  (More than this takes time.)  Also, remember that you can sell contracts at any time afterward - you don't have to wait months to collect your payout.

If you think that Hillary is going to do better than the polls on Super Tuesday, and you're going to sneer afterward and say that Intrade was "just tracking the polls", buy Hillary now.

If you think that Obama is going to do better than the polls on Super Tuesday, and you're going to gloat about how prediction markets didn't call this surprise in advance, buy Obama now.

If you don't do either, then clearly you do not really believe that you know anything the prediction markets don't.  (Or you don't understand expected utility, or your utilities over final outcomes drop off improbably fast in the vicinity of your current wealth minus fifty bucks - you don't have to bet the full $250.)  It is free money, going now for anyone who genuinely thinks they know better than the prediction markets what will happen next.

Prediction markets do not have supernatural insight.  If they give the candidates fifty-fifty odds, it means that the market collectively doesn't know what will happen next.  Even if you're well-calibrated, you get surprised on 90% probabilities one time out of ten.

The point is not that prediction markets are a good predictor but that they are the best predictor.  If you think you can do better, why ain'cha rich?  Any person, group, or method that does better can pump money out of the prediction markets.

If prediction markets react to polls, they're getting new information, that they didn't predict in advance, which happens.  Being the best predictor doesn't make you omniscient.

Everyone's going to find it real easy to make a better prediction afterward, but if you think you can call it in advance, there's FREE MONEY GOING NOW.

Buy now, or forever hold your peace.

58 comments, sorted by
magical algorithm
Highlighting new comments since Today at 11:57 AM
Select new highlight date
Moderation Guidelines: Reign of Terror - I delete anything I judge to be annoying or counterproductiveexpand_more

You're not accounting for the costs of investigating the prediction market itself. Perhaps if it was well known to be an extremely liquid market where I could be sure my bets would pay off. But I'm busy right now working on my start up, and even looking into the question will have a high opportunity cost.

My friend did a great job of arbitrage on the 2004 election, and made like $250 dollars. He would have made a lot more, but the market simply wasn't liquid enough to support it (he bought up all the contracts anyone would sell). It took him probably 5 hours to examine the market, find the arbitrage, and place the bets.

My wage consulting would be at least $100/hour. So it simply doesn't make sense to play in a market where, even if you realize the market is way out of whack, there simply isn't enough money in play to pay better than $50/hour.

Has anybody thought of prediction markets as a form of insurance? Suppose you don't like Hillary, then you can bet she wins the nomination. If she doesn't, you're happy because you don't like her. If she does, you win some money, either way you win.

Of course, if people did this it would make prediction markets less accurate.

I know Steven Hawking did. He bet against the existence of black holes. If it turned out all that work he did was worthless, at least he'd get a free magazine subscription.

Why would it make prediction markets less accurate? Does this problem adversely affect the prices of actual insurance?

Does this problem adversely affect the prices of actual insurance?

This might be an interesting question. Also, the following version:

Would prediction markets adversely affect the prices of ordinary insurance, or vice-versa?

If you are getting the insurance by placing bets, then people will be buying the candidates they don't like. If lots of people don't like a candidate, that is not a sign that the candidate will do well.

Insurance adds liquidity to the market. Someone has to pay the speculators a premium for being right, and sometimes that's the insurance-buyers. Higher premiums for speculators -> better quality speculation.

If you are getting the insurance by placing bets, then people will be buying the candidates they don't like. If lots of people don't like a candidate, that is not a sign that the candidate will do well.

But speculators should be able to get free money from those buying insurance, which will even it out some.

To the extent that they can distinguish, yes, and this may very well leave the markets performing well enough (and perhaps still better than anything else). It does necessarily add noise, however, which I understood to be the original point - it makes things less accurate.

But wouldn't rational bettors' willingness-to-pay for a stake in a candidate be the same in both cases (buying for insurance vs. buying as a speculator)? Their WTP would be determined entirely by odds, right?

Example: Llewelyn has (in your view) a 5-1 chance of winning. Maxine, whom you despise, has 1-5. Let's say I offer to sell you a voucher that is redeemable for $5 in the event the hated Maxine wins (and is just worthless paper otherwise). How much would you be willing to pay for this voucher?

I can't speak for you personally, but wouldn't the money-maximizer pay up to $1 for it? (In five possible worlds, L wins and you lost a dollar; in one, M wins and you get five bucks; pay a cent more and expect to lose money.)

And my point is this: The fact that you hate Maxine was irrelevant all along. You (or, again, our hypothetical rational agent) should be willing to pay up to $1 no matter how you feel about Llewelyn and Maxine, assuming a given estimate of their likeliness to win.

If I have errred, please do point it out where my map is wrong.

Their WTP would be determined entirely by odds, right?

No. In short: with insurance you're paying money to reduce risk. Thus, WTP goes up.

How much would you be willing to pay for this voucher?

It depends how much I stand to lose regardless of betting, if Maxine wins.

Why should it depend on how much you stand to lose if Maxine wins? No matter the value (as long as it doesn't approach your total wealth--or is this what you meant?) to you of Maxine's election, you'll gain more if you bet following the odds more closely.

Yes, if the people selling you insurance are rational, then you would gain more on average by not buying insurance, putting edge cases aside. That is true both of ordinary insurance and bets taken for insurance.

But the point of insurance is to reduce risk, not maximize gain.

For example, suppose Maxine winning would cost my business $200, and I cannot lose more than $50 and stay in business. Then I see a bet that pays $200 if Maxine wins, which costs me $50 to buy. It would be worth taking that bet regardless of the probability of Maxine winning, if I'm very risk-averse regarding losing my business. It turns a possible loss of $200 into a guaranteed loss of $50.

If the actual probability of Maxine winning is 10%, then the expected value of not betting is $-20, while the expected value of betting is $-50, so if I want to maximize gain I should not take the bet. However, taking the bet has a maximum loss of $50, while not taking the bet has a maximum loss of $200 (costing me my business), so in taking the bet I've gone from a 10% chance of losing my business to a 0% chance of losing my business. So if I want to minimize the probability of losing my business (all else equal) I should take the bet.

Dan - so did you get that easy 10% ROI on Ron Paul? If no, why leave all that free money on the table?

Andrew: I favor Bryan Caplan's explanation: when a belief you hold starts to have material consequences for you, when before it did not, your brain switches to more "realistic" thinking. From introspection, I find this to be true. It's a great experiment you can do on yourself anytime. Think about something you believe strongly, and then consider taking an offer to put money on it. In my experience -- and I'm the only one I can observe for this -- I get a weird feeling, like a complete perception shift. (If I studied psychology, I might toss in a "Gestalt" somewhere.)

Eliezer_Yudkowsy is right: If you really believe something will happen, and you like money, and betting that it will happen will get you more money, that implies that you should make the bet. (Issues about the bookie can be resolved by having a friend secure the bet.) The refusal to bet therefore suggests the person is being dishonest about his stated level of certainty. I believe Robin_Hanson has made a good point before about how claims like "Oh, Hillary can't win, I know it" shouldn't be read as actual claims of knowledge, but a "signal" about whose "team" you are on, and therefore asking to bet misses the real point of that person in making that statement.

Dave D... yes, I did. I didn't short at the peak, I didn't cover at zero, and Silas is absolutely right about the opportunity cost. I didn't commit a lot of capital because of the logistical problems and my unfamiliarity with the platform, so this was definitely more "edutainment" than investment, but yes, I did, I sold Ron Paul short and made money doing it. If I see another slam dunk, I'll do it all again, with more capital.

Robin: Actually, yes. One time I thought of the idea (okay, fantasy) of silently obsoleting government through a massive system of insurance whereby any time legislation is proposed, you make a bet (or equivalently, purchase an insurance policy) such that you become indifferent to the outcome. The result: any legislation will merely make everyone take a monetary hit, but otherwise leave their behavior unchanged. And if that doesn't make everyone cynical about the democracy that results, the fact that politicians will vote in ways that maximize the value of their bets, should do the trick.

I never bothered to look up if anyone actually developed the model more deeply though.

Well, any legislation you like would make you take a hit, and legislation you don't like would pay out.

But it's not a great idea overall without a very canny system in place to regulate what measures are put forward. Suppose EvilCorp puts up $1M in favor of making puppy smashing legal. Opponents then put up $1.1M and the measure is defeated. This pays out to EvilCorp. Then EvilCorp opens a motion to legalize kitten smashing and puts up $2.1M, including what they gained from the puppy smashing result. Now the opponents need to put up $2.2M...

At this point, EvilCorp can recover their $1M investment without slowing themselves down appreciably.

Every time you get your way, you're paying your opponents to win or extort you harder next time.

Right, he stayed at 60% or so for about an hour, then around 8:20 PM EST he fell back to 54%, held that for another hour, then suddenly plummeted to 35% with a quick rebound to the 40%-45% range, where he is as of 10:30 PM EST. Obviously these numbers track the swings of fortune as the various polling numbers come in.

It may seem wrong for a market to fluctuate so wildly, but I am beginning to suspect that many markets would show large fluctuations if participants were fully rational. Uncertainties dominate the future, and so valuations of securities might legitimately show tremendous variation with even slight changes in current information. Only the stabilizing influence of market makers, along with traders' irrational reluctance to venture outside the current trading range, would be responsible for the degree of stability we see in our financial markets.

Hot off the presses from the Wall Street Journal:

"The Wall Street Journal political market, run in partnership with, shows the Democratic nomination race in a dead heat: Hillary Clinton is rated a 50% chance of eventually securing the nomination, to Obama's 49%. Notably, the chances of an Obama victory have risen from 40% only yesterday morning, likely buoyed by strong national polling which shows Mr. Obama drawing within striking distance of Mrs. Clinton."

Obama received a similar bounce immediately after winning the Iowa caucuses, briefly vaulting up to 70% in the Intrade market before falling back to the 40% range. One thing to watch out for in these markets is the possibility that campaign strategists will see them as a tool to manipulate coverage, pouring funds into the markets to give their candidates a brief but well-timed boost. If and when this happens, it is a profit opportunity for traders whose strategy is to moderate sudden swings.

I suppose people could write in their predictions here, and gain some local cred for being right. It's not clear what exactly we should be predicting though. The election situation is actually quite complex as many or most states are no longer winner-take-all, so it comes down to hundreds of individual districts. Districts assigned an odd number of delegates are particularly prized!

Fortunately our disappointment about being shut-out of the markets due to gambling-oriented financial regulations may be partially assuaged by considering the various no-trade theorems, which state (under various idealized conditions) that you wouldn't trade in the markets even if you could, because the fact that someone else is willing to take the opposite side of the bet is prima facie evidence that his information is as good as yours.

From what I've read on Intrade, you can fund your account with up to $250 using a credit card, and it should land in your account immediately.

Not in the United States; US credit card processors will deny the transaction. A (slow, expensive) wire transfer or mailed check is required.

Of course, this opportunity presents itself every day in the form of sports betting. Sometimes the markets are "wrong" (e.g. yesterday's Super Bowl), but in the long term, it is pretty hard to beat the sports markets. FWIW, the sportsbook I use currently shows the same odds as intrade ( which makes sense as otherwise someone would arbitrage the difference away.


I see what you're saying, but I don't think I have a moral obligation to take every available opportunity to make money. I'm reminded of an event when I was about 10 years old: I took some small change and threw it in the trash. I don't remember why I did it, but I do remember that my dad was really offended. But, hey, it was my money. Betting is fine but I don't see why it should be privileged over other means of expression.

Silas and Jason, the insurance/hedging of redistribution policies would be valuable if the transaction costs could be made low enough. But are low transaction costs feasible? It appears that there are lots of issues where only special interest groups can afford to acquire information needed for hedging. E.g. the benefits to me of insuring against the harm I suffer from ethanol subsidies are small enough that I don't want to estimate the value of those benefits, much less evaluate potential ways to buy the insurance. If redistribution disputes could be simplified to be disputes over one or two numbers (e.g. the slope and y-intercept of income tax rates, with negative tax rates for the poor), then insurance might be feasible. But special interest politics seem to be firmly pushing us away from that goal.

follow up on the poker player's results (he put up 10K because he was convinced Intrade was easy to beat):

Can't sleep, so, postmortem:

I did two things very right, and one thing wrong. Almost all of my predictions prior to 2/5 were dead on, and they largely continued to be once the count actually began, except for one thing: because they fit my preconceptions and what I hoped the last minute vote was doing, I trusted Drudge's leaked polls. As such, I immediately jumped into the Dem market with both feet when I was previously committed to staying away from it until I saw EST results. The upside was that, when the aforementioned results came in and proved those polls horribly flawed, in most cases I got to take the money and run everywhere else before the market knew what hit it. I immediately got out of the national Dem market only down $500 or so and actually made a big profit off the states.

In short, I give myself a B+ for the night.

Oh yeah, results...up 2800 on the night, 35% total, and nearly 100% since NH with a lot of money still sitting in those guaranteed $$$ Pres.McCain shares.

In comment 2, Emmett said, "You're not accounting for the costs of investigating the prediction market itself."

Which is true, but I think the point is not that you should bet. The point is that if you have strong beliefs on the subject, you should bet. Obviously, if you think Obama is going to do better than the polls suggest, you have strong opinions and have already investigated the issue.

The combination of "I have a strong rational belief in X and believe the betting market is wrong" and "I am nonetheless choosing not to bet" indicates that the belief is not rational at all.

To those people who say that they don't bet on principle ... I say, set up a sting operation where you offer them a no-brainer, like even money that Obama will get more than 10% of the vote. When they jump at that one, they will never again have a decent excuse for not betting.

Dave_D: The problem with betting against long odds is that InTrade pockets most of the interest (the "float") between when you cover your position and when they pay you. That's how they make a lot of their money. So, let's see, put up $90, then 18 months later, get $100, minus fees. That's about the return you'd get in a money market mutual fund, but the latter would lack the structural risks (InTrade being shut down, unreliable, etc.)

It would be more tempting if you could cover your position by buying a government bond dated at the latest time the contract can expire, but that has lower interest than the equally-reliable MMMFs (and yes, this matters for low-odds bets). Simply crediting you with interest would improve things, but then, who would run the site? ;-)

I say this because I had actually considered doing exactly that.

As Eliezer said, prediction markets may react to the information conveyed by polls, but they certainly don't just follow the polls; for example, there was a period of time when among the Republican candidates, Huckabee was the front runner in national polls. For solid reasons he was never the front runner on Intrade.

Only some US cc processors will deny the transaction. The transaction fall under their category for betting & gambling, same thing that prevents you from pursuing cc transactions with online poker sites. But I've seen cases where these transactions are unblocked with certain banks.*

  • Not me, of course.

Nick: I didn't know that. Any way for US folks to get into a prediction market before tomorrow - an alternative to Intrade?

I actually wanted to bet about $250.00 that Obama was going to win in 2012 on Intatrade, because I thought Nate Silver's models were more accurate then the market, but my wife really didn't want me to, heh. Still, there was a pretty big gap between the market (60% Obama or so) and the mathematical models (Nate Silver had Obama at over 90%; some other models were even higher).

On a more general note, I don't think the prediction markets in general were that accurate in 2012, because A: a lot of people were relying on provably false information, and B: I think a lot of people invested in the markets, not because they thought they knew the result, but because they wanted to manipulate the result (if the Intatrade market showed a close race, then the media might report on that, which might influence people's minds, ect).

Indeed, Intrade didn't necessarily perform very well for the 2012 races:

I suspect Intrade will perform even worse next year now that they have banned American users due to persecution from CFTC.

Ah; I had missed that news story, that's a shame.

This seems a little bossy to me. Beyond the issue of transaction costs ("the vig") and the effort of gathering the information to try to beat the market (this would an intellectual hobby, like blogging, doing crosswords, or following the horses, that would make sense to do if enjoyable in itself), maybe some people don't want to bet. I have no problem with betting--I enjoy it--but I'm a little puzzled by the statement that people should be betting, or that they have some sort of moral obligation to put their money where their mouth is. Maybe you personally don't "really believe" things unless you put money on them, but not everybody feels that way.

Peter, you are right and the first step is to just be able to insure against taxes. This is what I'm trying to do over at Intrade:

I am using the highest federal marginal income rate as a proxy for effective rates. There are other factors that determine effective rates like AMT and the social security cap, and I think we will launch those markets at some point if liquidity builds.

Hal, I expect that fully rational traders would produce larger bid/ask spreads when the results are being announced, and trading volume that is closer to what is suggested by the No Trade Theorem. Traders have a widespread bias to overestimate their ability to profit from following news. For example, see an experiment done by Andreassen where subjects trading stocks did worse if they saw a constant stream of news than if they saw no news once they started trading.

I'm a little puzzled by all the above. A prediction market is supposed to squeeze out the last drop of (solvent) expert knowledge concerning a given outcome, but voting is a complex and chaotic phenomenon where 'expert knowledge' is thin on the ground or non-existent. If anyone could reliably forecast election results, I think we'd know about it by now. Think weather forecasting, as a comparison. So we're left with, at best, a few people who really believe in their favorite groundhog, and are prepared to place serious money, and others who are prepared to have a flutter on their hunch or their hopes. The result, even ignoring any possible manipulation, looks awfully like a poll of forecasts, distorted by individuals' like of betting and amounts of spare cash as selection biases. Polls of forecasts are less reliable than polls of intentions. And as for Phil, who seems to think not betting is suspect, perhaps Un-American, or perhaps downright criminal, consider that many people have a conservative mindset, don't bet as a general rule, and so just couldn't be bothered. Apart from which, my Dad told me you should work for a living and not to bet, and my Dad is an Authority.

I am beginning to suspect that many markets would show large fluctuations if participants were fully rational

Maybe. But I'm willing to bet that that is not the explanation for Intrade fluctuations...

Regarding the Intrade fluctuations, maybe we're seeing an example of the Law of Conservation of Expected Evidence.

Obama has just spiked in the last hour or so to 62% on Intrade. The early poll numbers seemed encouraging and it seems everyone is taking Eliezer's advice to heart.

I think I am correct in saying that has seen more action than Intrade on the Democrat Candidate Contract but I presume they are off limits to you poor yanks...

Yes, you can sell contracts as well. In fact, whenever you are buying, you are doing so from someone who is selling.

However, seeing that Obama and Hillary are really the only candidates left, shorting one is (more or less) equivalent to going long on the other.

As far as getting money into an account quickly goes, can you sell short at Intrade? If so, I'd short Hillary, and long Obama, but not confidnetly enought to actually do it.

(And by "expected utility" in the above comment, I meant "expected value" not taking into account risk attitude. One must be precise about such things.)

What if one thinks (as do I) that not only do prediction markets do badly, but so do I? If both me and the market aren't doing better than random, do I have positive expected utility for betting?

Also, I'm not sure how intrade's payoff calculation works -- how much does one stand to gain per dollar on a bet at those odds? I think I'm pretty risk-averse if I'm gambling $250.00 for a $10.00 gain.

Anyway. My cash-free prediction is Obama by 2 points in general.

The intrade markets are tiny and illiquid for real players. If I can get similar or better odds elsewhere on larger size, which I can, why would I use them?

The intrade market is purely reactive, not predictive. It has been for year and years. You can't have it both ways by saying they are going to be predictive, but when they are wildly wrong and switch overnight, then they are STILL predictive but taking into account new news.

Secondly, there are real transaction and contract costs involved. Plus the risk of not getting your money back and having to investigate the site, alone, with no help from the gov't if they stiff you.

If you think the tiny volume in that market is representative of ALL the collective wisdom as the NYSE is, or the Eurobond market, you're clearly mistaken.

Real stock markets work the same way. If markets really were able to predict the future with perfect accuracy, no bubbles would ever form, and no one would have ever invested in Enron or Lehman Bros or Bernie Madoff. I don't see how you can demand that a market make predictions based on information that doesn't exist yet.

There's a great article on arbitrage opportunities in the FA Cup at the website Although it doesn't conclude there actually is free money, it just posits there are wagers with positive expected values.

The Iowa Electronic markets is legal for US citizens to use to bet on the elections.

But they only take checks - not what Eliezer was looking for. And they have low liquidity, very few contracts, and a $500 account limit. :-(

(AFAIK [IANAL], it's not actually illegal for US citizens to bet on Intrade, it's just illegal for US financial institutions to transfer funds to them.)

I remember a cold call from a stockbroker years ago, wherein he argued that if I didn't believe that the market was going to go down, then I must believe that the market is going to go up.

Leaving aside the stay-the-same option, that isn't A or not A.

``Believe'' has its own grammar.

Wittgenstein : 575. When I sat down on this chair, of course I believed it would bear me. I had no thought of its possibly collapsing.

Adanthar, a poker pro who helped break the Absolute scandal and developed a "robotic" small-stakes algorithm on a lark that supposedly returned approximately 20% before it became well known put down 10k and has been updating his Intrade progress here: I'm interested to see how he'll do today.

The Iowa Electronic markets is legal for US citizens to use to bet on the elections.


no-trade theorems, which say (under idealized assumptions) that even if you could trade in a betting market, you wouldn't, because the fact that someone is willing to take your bet is prima facie evidence that his contrary information is as good as yours.

Hey, shouldn't you guys be out, like, voting or something?

A couple of months ago, Intrade had Ron Paul at close to 10%, the polls had him at 4 or 5%, and anyone with even the slightest hint of a suggestion of a clue knew that Ron Paul had 0% chance of getting the nomination. Now THAT was easy money. The Hillary/Obama trade is nowhere near a slam dunk.