Non-Linear Growth

For those that are still skeptical of the answer to the Monte Hall dilemma I posted on a few days back, my colleague Tom Simonson has written an excel simulation. Tom’s simulation proves that across 5,000 observations, you are two times more likely to win by “switching” doors than by keeping the door you originally chose. The version posted is static (Google Docs doesn’t handle data tables very well); for those of you who want the dynamic version, drop me an email and I’ll send it to you.

Update:

For those of you still in disbelief with this, these links may help.

• Wikipedia entry on the Monte Hall dilemma; here.
• Peer reviewed paper on the Monte Hall dilemma and the associated cognitive disconnects it creates; here.
• A nice simulation of the Monty Hall game that you can play online; here.
• Another simulation that enables you to run multiple trials; here.

Remember, the key to this dilemma is understanding that the game show host knows he is opening a door with a dud prize and not a door with the valuable prize.

For me the beauty of this is how a simple game can be so deceiving to our internal probability calculators. Imagine how we perform making complex probabilitic decisions with incomplete information.

Filed under: Random, Monte Hall problem, three prisoners dilemma