My prior post about the television show "Deal or No Deal" here led to a Google search that showed my blog and the person searching happened to be a teacher. He emailed me and asked if I had made a spreadsheet for the game so that he could use it with his kindergarten class to teach them about statistics. I made it last night, spending much more time than I should have, but it turned out well. I did it for the kids, (and because I'm a spreadsheet nerd).
Download deal_or_no_deal_final_atul_patel_v2.xls
Click on it to download it and as it opens, you'll see a box that asks about macros. Enable the macros. Then follow the directions on the spreadsheet. It's quite spartan in appearance, but if any of you want to spruce it up, feel free.
+ Atul
Nice spreadsheet. Why did you calculate the banker's offer at 85%? Is that a consistent discount from the mean? I thought the gap was larger in the early rounds and smaller in the later ones.
Posted by: Peter Kim | March 31, 2006 at 08:35 PM
I only did the calculation with 85% because I hadn't really watched the show enough to notice trends. It was a crude spreadsheet in that respect. And I don't think that their rules are consistent. Sometimes they offer over the expected value, sometimes under. I think it depends on what proportion of suitcase values are over the E.V. and what proportion are under. As I'm typing, I'm realizing that if the offer is over the expected value, then maybe they know that you have a suitcase with a high value. Just a theory.
Posted by: Atul | April 01, 2006 at 10:25 PM
I think I see the formula of how the come up with the "Bank's Offers" But now there rasing the winning offer.
My question is how did you make the spreed sheet formulast to change the no. in the calculation for the bank offer.
Posted by: A. Lipscomb | September 20, 2006 at 08:33 AM
I just made it give a percentage of the expected value of the remaining suticases. This is not what the banker on the show does. He adjusts based on a number of other factors like number of suitcases above or below the expected value, as well as how recently the contestant started playing.
Posted by: Atul | September 21, 2006 at 12:52 PM
This spreadsheet, although quite elegant, has some serious problems. Choosing a random case at any point in the game results in the bankers offer being a null figure. I can't figure out why, but it can't be to tough to figure out if I have more time....at other points after eleminating the top values, the banker's offer dropped to 1K then after another selection turned to null also. Hmm...
Posted by: Chuck @ Podunk | September 23, 2006 at 09:57 PM
If you ever want to make the banker's algorithm more realistic, learn about valuing real options. It's also what explains why people purchase lottery tickets -- rather than a "math tax" on those who can't do the math. Like investing for retirement, the right answer all comes down to your personal and current tolerance for risk. If the thugs will break your legs tomorrow if you don't hand over a certain amount, you're more likely to take a deal as soon as it reaches that amount, as an extreme example.
Posted by: Jeff Buchmiller | November 26, 2006 at 10:11 PM
Stumbled across this site and was intrigued. In the UK we have a version of the gameshow and after watching it I too was compelled to model it in Excel. Ours is a little different and here is my effort if you want to check it out: http://www.nordberg.co.uk/ExcitingExcel/leisuresummary.htm#DOND
Posted by: Chris Nordberg | July 23, 2007 at 06:34 AM
I envy your style, the idea that your post is a tiny bit unusual makes it so interesting, I am fed up of seeing the same stuff all of the time.
Posted by: Puma Golf Shoes | June 28, 2010 at 03:18 AM