As real-money prediction markets become legal in the US, there’s surely going to be a lot of money in writing about them and driving rich readers to them.
Carlos Graterol [*] and Ben Shannon both tried to popularize their prediction market blog (featuring InTrade 95% of the time), and they never managed to take off. The fact that InTrade needs websites to drive people to its betting operation does not mean that readers will appreciate prediction market journalism.
Prediction market journalism (which sums up news and probabilities harvested from newspaper sites and from InTrade) does not produce any scoops. You need scoops to draw readers into your blog. No scoops, no readers.
[*] He is a smart and sociable young man with a bright future.
P.S: I am not picking on Joe, who is a good web curator, and a nice chap. Just disagree with him (and Andrew Golberg) that prediction market journalism is easy to do. Good prediction market journalism is very hard to do. Requires scoops and pertinent analysis. Scrapping exchange probabilities and aggregating it with re-written news is not enough.
Joe should mention whether there is volume on each market.
Joe should cite BetFair, not InTrade, for any UK-related event.
Joe should be aware of InTrade’-s long history of fucking up contracts and settlements on non-sporting events. (Type “-North Korea missile InTrade”- in Google, and review the various InTrade forums for traces of past fights.)
Posted by Jesse Livermore on Sunday, January 3, 2010
I’-ve been gone for a while. There just hasn’-t been that much happening on Intrade, and I’-ve been focusing on neuroscience.
Intrade has definitely gotten tougher over the past year. I think the 2008 election drew in a lot of people who weren’-t very good at politics or gambling. By now those people have either lost their money or gotten better. Hopefully the 2010 elections will draw in a new crop.
In the mean time, Intrade’-s management has not done a great job in developing the brand. My impression is that volume is off by more than 50% compared to last year. Chief difficulties: - Absolutely no advertising whatsoever. - Diminished interest in politics in an off-year. - Getting money onto the site requires a lot of determination and a visit to a gas station to buy a Netspend card.
Future updates on this blog will be less-than-daily, basically when I have an opinion about politics that I feel like sharing.
Here’-s a new application of prediction markets now under review for a grant: Prediction Market Journalism.
Here’-s how it works: a journalist in London wants to investigate government plans to fund long-acting reversible contraceptives (LARC) in schools.
The journalist opens an either/or real-money, public prediction market question: Will the government fund LARC for girls in school clinics?
The journalist is interested in questions like: How safe is LARC? What companies make LARC? Which company would supply school clinics with LARC, and why that particular company? Who would profit from LARC distribution in school clinics? What other contraceptives are currently government funded in schools? Both the journalist AND prediction market traders want these questions answered. The former, to publish. The later, to make better market predictions from.
So, in Prediction Market Journalism, traders can help answer these questions. Traders contribute documents, video, photos, tips, links- whatever helps the journalist do the burdensome investigative research. With sufficient vetting, User Generated Content (UGC) is posted for public viewing, thus helping everyone to make better predictions.
In short, PM traders can indirectly effect the market on which they make predictions. Journalists get both crowdsourced investigative research and a percentage of trading commissions.
That’-s a new application of prediction markets.
As a prediction market aficionado, what lights me up are stories about…- (of course)…- how the prediction markets are assessing important news. The HubDub blog publishes, every Monday morning, a post that rounds up the 5 most prominent (that’-s subjective) news stories of the week, with the prediction market charts, so we can spot which outcome is the more likely, for each issue.
I find this weekly newsletter addictive. I read it with attention each Monday. It is simple, short, but well done and effective.
I wish InTrade, BetFair, and NewsFutures would publish such a prediction market blog.
BetFair Predicts (a blog run by BetFair) titled “-The Power Of Objectivity”- a post giving the latest odds produced by BetFair on the race for the White House.
The real “-objectivity”- would have been to quote the odds produced by the other prediction exchanges, too —-InTrade, Iowa Electronic Markets, Betdaq, NewsFutures, HubDub, etc.
Midas Oracle is the only blog that lists prices and probabilities from all the prediction exchanges. No wonder, our daily readership is much, much bigger than the audience of all the other prediction market blogs combined. A blog that gives the odds of one exchange only is like a dead end —-no one trusts a dead end.
UPDATE: The guy in charge says in the comment area that this is just an early beta website, which is going to be much improved soon.
WHAT I LIKE IN BETFAIR PREDICTS
BetFair Predicts is a (clumsy) response to some of the (harsh) criticism I directed at them. So, it shows that they listen up.
They chose WordPress as the content management system (CMS) for this project —-it’s the right choice. (DruPal would have been OK for a more sophisticated project.)
They are web-hosted on an independent domain name (as opposed to a sub domain on the betfair.com website.) It’s good because it gives them freedom to use whatever information technology they want. (For security reason, they get limitations on their main website. That’s why the BetFair blog had to be run on MovableType, and not on WordPress.) – [UPDATE: The website now redirects to http://predicts.betfair.com/.]
The architecture of this new website is well done. It’s standard. (Unlike the BetFair blog, which is a price of crap, technically.)
They created a new compound chart on the 2008 US presidential elections. (I have hot-linked to it, just below.)
WHAT I DISLIKE IN BETFAIR PREDICTS
It was an error to call this project BetFair Predicts“, singular. “BetFair Predict”, plural, would have been better, because the wisdom of crowds requires both a collective judgment mechanism (here, a predition exchange named BetFair) and the event derivative traders associated with that prediction exchange. By using the singular, BetFair appropriates a predictive power that should be credited to a community (lead by BetFair).
Overall, it’s a very thin website. It does not bring much to the kitchen table. (That new compound chart above is really the novelty at this time. But it’s not Earth shattering, since the BetFair Politics Zone already displays compound charts, which are dynamic, and which can be hot-linked to.)
That compound chart is probably not a dynamic one (that is, a chart that will update itself in the future). This compound chart is just an image that has been uploaded from the BetFair exchange to this new BetFair Predicts blog, I doubt that that image will be updated in the future.
And they seem light-years behind when it comes to embeddable chart widgets. (They claim on the frontpage that you can “embed” their charts in your blog, but what they offer is just static charts. Those idiots don’t know what they are talking about.)
Their explainer on prediction markets is…- so-so.
Their blog is written by a “Joe Seither”, whose background is not disclosed. The writings (well, only 2 posts, as of today) are typical of any copy-writer who would have been instructed about the mechanism of the prediction markets the day prior to undertaking this project. You won’t find any insights- just banalities. What I see is worthless.
They can’t spell “RSS”. (See below. )
BetFair is still struggling with the web publishing technologies.
BetFair is still struggling with the prediction market approach, embodied by InTrade.
BetFair has a hiring problem: they pick up incompetents on the job market and put them in charge too quickly. The result is the disaster detailed above, and chronicled in the Midas Oracle archives since 2006.