Slashdot Mirror
Google IPO Swami
The Google IPO Swami writes: "I'm running an experiment and Slashdot readers would be good contributors. As you may know, Google recently announced that they will be using a unique dutch auction structure to price shares of their IPO. Instead of having the underwriters determine the opening price, the price will be set by the demand of investors that register to participate. I'm interested in how well the public can estimate this demand and the price of the shares to be offered. I'm giving away free shares in Google to find out. The person that comes closest to estimating the opening and closing price of the stock on the IPO date will win shares in the company."
It looks to me like the only piece of info needed to register is an e-mail address. With people here capable of supplying thousands of e-mail addresses each, I think you're looking for abuse.
No, no, no, you got it all wrong. Winning is certain:
/wonders if a few minutes to write the script is worth it for a few shares...
Two 0-999 ranges
i.e. 1000*1000 possible combinations Need 1,000,000 submissions to 'win'
Spamgourmet.com allows dynamic forwarding
g000000.2.name@spamgourmet.com
To
g999999.2.name@spamgourmet.com
Here I come!
Given the large amount of publicity Google's IPO is getting, and assuming the dutch auction will be pretty open to the public (I trust Google for that), the whole point of the dutch auction is to dampen the first day effect and make sure the company, and not day traders, gets most of the upside the market is willing to give it. Notice the assumptions up top, though... Theory doesn't always translate into real life.
And another thing: read the prospectus: the wall st guys are still getting a pretty good cut!
Yes, googles stock has to compete with other stocks/bonds/etc but that doesn't mean that the stock will exhibit rational behavior on the day it IPO's.
You are looking at the stock market as a pure numbers game when a large factor in the rise and/or fall of googles stock on IPO day will be how people 'feel'.
That is what caused the tulip trouble and the dot com trouble.
Is there any reason to assume that the average consesus also is the expected value of the stock?
First of all, most people who enter the contest won't be experts like the pilots of your example.
And next to that, even if they were there still could be situations where this approach would not work. Let's consider the plane example: If there were to probable routes A and B, some experts will choose a crash site along route A and others will choose a crash site along route B. The average of these guesses will be somewhere between routes A and B, on a highly improbable route.
I'm not saying this guy won't get useful data out of this game. Au contraire, I think it's a smart idea. But I'm not sure whether using the data in the way you suggested to calculate the stock value will yield a reliable estimate.