UsableMarkets

Who Is …
Inkling is another recent entrant into the public prediction market field. Already generating a lot of interest and commentary in the Prediction Market community, Inkling does something that appears (at least to me) to be very novel. As they say in their Q&A: “Inkling has been designed to be easy enough for our moms to use, but satisfying for stock market gurus as well. We feel if you are going to try and gauge the “wisdom of the crowd,” the crowd has to be able to use your application and not get bogged down in market-ese.”

The Good
Sets itself an admirable goal and has some nice features, such as the Trade Assistant.

The Bad
Sloppiness in market structure and design, and contract language generates more confusion than there needs to be.

The Tags
Market Purpose: entertainment, predictions, corporate use
Market Control: Inkling, individual users
Targeted End User: general trader, prediction market aficionado, corporations
Currency: play money, inkles
Underlying Event Types: events (business, entertainment, sports, etc.)
Contract Types: binary
Trading Mechanism: market maker
Size of Market: not sure how many traders are registered
Number of Contracts: 24 markets (by my count, although it says there are 58 markets on the home page), each with between 1 and 10 or so contracts

The Metrics
Communication of ideas
To a degree Inkling stakes its reputation on ease of use. I applaud that, of course, and I applaud that Inkling tries to teach you not through lots of text and help pages (although certainly more of that would help), but through the interface and process of trading.

Things get off to a relatively good start on the home page. The text, while a tad vague, is generally straightforward and sets the stage. “Buy and sell shares in what you believe will happen.” Once the user registers it’s then fairly easy to find a market and get started.

Each market revolves around a question. For example: “Will General Hayden be confirmed to be the next head of the CIA?” Or: “What team will win the NBA Championship?” Or: “What will the price of Oil be on 5/31/06?”

Each of these markets is intended to be presented in a straightforward way, but language and market structure get in the way.

The structure that Inkling has set up for their markets and contracts works best in the multiple outcomes for one event scenario – the World Cup, for example. First I’m asked what I think the winning outcome will be – France, for instance – and then the probability that this will happen. The first question susses out your opinion, and the second question susses out the strength of your opinion, or how much you are willing to pay for that opinion.

This structure adds redundancy to a one contract market, since the only important question is what I think the probability of this event happening will be. And it is irrelevant for a linear contract, since linear contracts should reflect estimated outcome – say $72.00 per barrel – as opposed to a probability – a 72% chance that oil will go over a certain price. What would happen to this market and it’s predictive ability if oil went to $150/barrel?

Example 1

One could also argue that the first question – what I think the winning outcome will be – is always irrelevant. For example, if I believe that France will go the farthest of the teams listed in the market, but I’m not willing to buy at $9 (or 9%), then do I really believe that France will go the farthest?

This is where language comes in. Structuring markets based on your beliefs, but then buying (or selling) based on probabilities, mixes a voting model and a betting model. Remove the voting aspect of it (“What team among the current FIFA top 10 will advance the farthest in the 2006 World Cup?”), and suddenly the markets make a lot more sense (I don’t believe that France has even a 9% chance of going the farthest of the teams listed). The second question incorporates price and belief, while the first one only asks about your beliefs.

Separating these two questions probably seemed like a good simplification, but I’m not so sure.

This is not to say that one can’t make their way through a trade. The problem is that the site actually becomes more confusing with experience, and at some point you just operate on faith (e.g. I know price is supposed to be a probability, but in the oil market I’ll just assume it’s the expected price at the end of the month). While simplification is an admirable goal, over-simplification is worse.

It is worthwhile noting a couple other aspects of the site. The concept of shorting – in that you work off margin, borrow shares, earn profit from the sale, and then need to cover at a later point – is more realistically reflected here than on other public prediction sites where “shorting” occurs. While I did like this, I can easily imagine that if I didn’t know anything about shorting I would be terribly confused, though contextual help does a good job of explaining the impact of shorting on your Total Assets (example 2).

Example 2

HSX, Smarkets, and others have gotten out of this problem by allowing “shorting” which is really the purchasing of opposite contracts. NewsFutures makes explicit that you are purchasing “opposite outcomes.” These are perhaps simpler solutions and easier for users to understand, but Inkling is more like the real thing.

What continues to be a problem is how one covers a short sale. You can go ahead and purchase shares in the contract you’re shorting, and this does indeed cover the short sale, but there are no clear calls to action that this is what you’re doing, and so one does it on faith. In other words, if the user doesn’t know how to cover a short sale, they will be totally lost.

Even more problematic is if I want to sell a purchase that has gone awry. If I own 100 shares in something, and I want to sell those shares because I’ve changed my mind (or lost too much money), and if I don’t have enough margin in my account, then I can not sell my shares! This is just plain wrong.

Also, given the low liquidity of the site, when a user purchases or sells a contract they move the market, in that the price changes, and has an effect on your average cost-basis. I can easily imagine that if you didn’t understand this, you might be confused why a purchase you thought you were going to make for one price turned out to be a higher price.

Functionality
The functionality of the site is robust and the owners appear to be working diligently to add even more features.

The most recently added feature is the ability for users to create their own markets, as on CrowdIQ (see my review). If they haven’t already (and I believe they have … more about that later), Inkling will run into the same problems with contract clarity.

This piece of functionality also shows where Inkling believes the future is: corporate prediction markets.

In terms of the basics Inkling does a pretty good job. There is a dashboard where traders can review their positions, see their performance, their assets, and how they rank among all users.

There is a Trading Activity tab that allows you to view your trades over pre-determined periods of time. And there is a Top Traders tab, which allows users to see Top 10 in the past week, and all time.

In terms of trading, on a market page (see right) the trader can select the contract they want to trade, view a chart of all contracts within that market, and get market and stock specific information. Your performance, total assets, and trading rank are listed, as they are on all pages.

On a contract page is where the neatest piece of functionality appears. This is the Trade Assistant (example 3). It tells you the price of the contract, your investment and shares in it (if any), the total shares traded in this asset, and the last trade.

Example 3

Selecting an answer prompts you with a list of choices that gives you guidance on how you should bet (example 4).

Example 4

This is great functionality, and, I think, would be really useful for someone who doesn’t have any experience trading. It allows them to translate feelings about an event or outcome into an action. If the trader doesn’t want to follow any of the recommendations, they can enter in their own amount.

Once the user makes their selection they are asked to confirm that trade (example 5), telling you specifically what you are buying, for how much, and how much money you will have left after the trade. If this sits well with you, you can then execute the trade.

Example 5

Interaction
While market structure and design and contract language are problematic, there are some nice features on the site other than the Trade Assistant.

For example, the contextual help (the question mark icon) is useful, particularly for the Total Assets area on every page. While there could be confusion about the difference between “inkles in bank” and “available today” (and in truth, I would think about how to make this language clearer), the help feature nicely explains the difference.

In addition the chart feature on the contract and market pages is a nice touch. The ability to view differing periods is very useful. Too bad the chart can’t be enlarged.

Market and contract information on the market and contract pages are also a great idea, but don’t always have the most useful information. For example, on the Fed Rates (6/28) market page we are given the following market information: “Fed rates have gone up and up and up. What will the committee decide to do in June?” Is this helpful to someone interested in betting in the market? I doubt it.

Clarity of contracts
While some contracts appear quite specific …

“Will GM file for bankruptcy by May 31, 2006?”

… others leave something to be desired:

“Will Blackberry Cream be the best seller for Jones Soda after 3 months of sales?”

The immediate question upon reading this is: which three months of sales? Looking at the contract expiration date does not help matters, since the contract expires at the end of the year. From this do I conclude that we’re talking about the last three months of the year?

Other markets make almost no attempt to be clear, or just assume the user will understand. For example, in the NBA Playoffs market it never explicitly states on a contract page how this contract will be judged.

This varience, I assume, is the result of people creating contracts and markets who are not sure how to be clear, or are not interested in being clear.

Account basics
In general setting up an account and managing it is fairly easy. Once signed into Inkling, user settings and account information are not all lumped under one Account area, which is nice. Instead Inkling has Dashboard, Trading Activity, and Settings tabs where a user views and interacts with information specific to them.

In the Dashboards tab the display of your positions is nicely presented. I’m not sure why newest markets and most active markets are available alongside your current portfolio (shouldn’t this belong under the Markets tab?), but it doesn’t detract from the simple and straightforward presentation of your positions. Too bad, though, that one can’t initiate a trade from this page.

Example 6

In addition, the Trading Activity tab is a nice touch, and easily allows a trader to see what they’ve been up to over various periods of time. Perhaps in future iterations one can set their own time period.

Market fairness
It is not clear how market fairness is ensured on Inkling.

Trader incentivization
Traders are incentivized by bragging rights. There is a Top Traders of All Time list, as well as Top Traders of the Week.

As always, thanks for listening.
~alex

If you enjoyed this analysis please Digg it