Prediction Market Tutorials
What Happens If I Don't Sell?
Let's go back a few steps - let's say I still own 10 units/stocks, and prices keep going up.Now let's say that I still believe this product could sell a lot - and I stick with it until the end.
Choosing the "end date" (or expiration date) is tricky, but let's assume for argument's sake that we expire the market when the company finally decides which products to develop next year.
Note that another option is to wait out until end of year 2009, and see what really happens - but that's very far away...
If in the final management meeting, company decides to go with this product - the question value will automatically become 100%.
This means I'll get back 10 X 100 = 1,000 points, and my earning will be 1,000 - 230 = 770 points.
If, however, company decides against developing this product - the value will automatically become 0% - and I will get nothing.
Which means I lost the 230 points that I initially invested.
How Can I Sell Something I Do Not Own?
In the example above, I pointed out that if I come to participate in the market, and see that the current probability is too high - I can "Sell" - even though I never bought anything in the past...How does this work? Let's see an example:
Say you join the market when the product's success probability was 40%. Your do not believe in this product. Furthermore, you think other products have higher chances to succeed. So, you think there's no way that product X has a 40% chance to succeed.
You choose to "Sell" 10 stocks, "betting" against this product.
Selling (or "selling short" as it is called in most places) has many implementation alternatives - we will discuss the one used by Qmarkets, which is by far the simplest one for you - the end user.
So - by selling 10 stocks, you will pay 100-40 = 60 for every stock. Selling 10 stocks will cost you 10 X 60 = 600 points.
After a month, company management meeting decided NOT to go ahead with this product, just like you thought.
Market value is now automatically 0% . But for you, since you "sold short", market value is 100-0 = 100. Which means that you will get back 10 X 100 = 1,000 points. So in this process, you have earned 1,000 - 600 = 400 points!
To summarize - if you sell short, you want the probability to go down as much as possible. The more it goes down, the more points you earn. If the probability goes up, you will lose all the points you spent.
If I Am Buying - Who Is Selling?
When you buy something in life, which could be a real product or a stock in the stock market - normally someone else is selling it to you.What is the equivalent in prediction markets?
There are two well known ways for managing a prediction market:
- Using Book Orders (like in the stock market)
- Using Automatic Market Marker
In the first option, if you want to buy 10 units of the market, someone else has to own at least 10 units, and want to sell it to you. On the other hand, if you want to sell, you'll need someone that wants to buy from you - and at the right price.
This is how the stock market works, and there are some prediction markets platforms (exchanges) that use this mechanism.
However - it really complicates the user experience for participants, and requires quite a lot of participants ("volume") in order to work.
At Qmarkets we went towards the simpler method, in order to keep it simple for our users, and not require them to be skilled stock market traders.
We use an automatic marker maker method called LMSR. What it does, basically, is uses a fixed algorithm in order to decide how much will the market value increase when you buy x units, or how much the value will decrease if you decide to sell.
It means, you can decide to buy or sell regardless of what other people in the market want to do - it is the market owner (us) that sells the stocks to you (or buys them from you).
Read more about this method, invented by Prof. Robin Henson in his article on LMSR.
If you are looking for a simplified version of this, you can also look at this article by David Penock.
This method has its drawbacks, but we believe that the advantages are much bigger than the flaws.
Previous <== What Is A Prediction Market?
Next ==> Does It Really Work? Wisdom Of The Crowds..