At its heart, Inspectd is a simple game: it shows you a chart of historical data on a random stock, asks you to bet on whether the stock’s price will rise or fall, and then immediately tells you if you won or lost — with another performance chart showing you why.
And maybe that instant gratification is what makes this game so addictive.
What surprised us most was how difficult it can be to go broke — at least on Inspectd’s virtual Wall Street. If that ever does happen, it sure will be fun to see how the business press explains it.
and 
. You can also watch it on 
Of course its difficult to go broke when you can’t use any leverage. A stock would have to go down by 100% for you to go completely belly up, no matter how much you have left in your account. Although this happens occassionally (Enron), it is very rare.
However, it IS quite easy to loose MOST of your money. I was down about 95% pretty quickly after adopting a “buy only” strategy, which suggests to me that the game is weighted to the downside and not completely random.
It is not necessarily weighted downward. To see this, set up a simple spreadsheet where you start with some amount of money. If you sequentially buy stocks with all of your money and the stocks alternate between gaining x% and losing x%, you will lose money over time.
Y*(1+x%)*(1-x%)*(1+x%)*(1-x%)…..
Once you take into account the broker fees, it’s easy to see how easily you can go broke.
10 trades, total profit 187569 Account value 287569.
No strategy…. buy/sell by first intuition.
Random luck?
Brad,
It’s not percentages that should be in normal distribution. There should be a lognormal distribution [approximately] centered around prevailing market returns from the given periods. If Ben were wrong, and there’s no skew to the data (and his strategy consistently loses money), then Wall Street should have been a losing game for a long time, which is clearly not the case.
After 50 consecutive (blind) BUY trades: $1.49 Million… random fortune, but isn’t the “buy everything” approach somewhat similar to an index fund, just without the appropriate weighting?
Hi, I’m the guy behind Inspectd.com. Maybe I can shed a little more light on the site: it selects a random US stock and a random date between 1987 and 2007, and displays that chart. The default trade length is 20 market days. Ben and miki have shown that a lot of it is luck; however, after a large number of trades, the prevailing higher prices over the last 20 years makes it hard to lose money.
Brad: that math is technically correct, but even in a flat market, the chance of a 20% rise is higher than a 20% loss. Think of it this way – which is more likely, a 100% rise or a 100% drop? Obviously the former.
Hi Sam, why did you close inspectd.com?
Do you have a new site like this one?
Any interresting statistics you got out of it?
thanks
wil
Yes, random trades in a generally upward market are going to get you, on average, random fortune (but you’ll call it investment genius). The question is how often do you pick up the Enron/Worldcom/Delphis that will clean out a long stream of “wins” in one shot, and how often does that happen?
If you want to have even more fun, do the “advanced” strategy where you can extend a trade over multiple periods … e.g., you can close out a win (or loss), or let it ride another day. Makes the martingale aspect of the problem even clearer – if I just hang on another day I’ll see if I can lose less, or convert back to a win.