An A/B Test for Halloween Candy

My wife and daughter are out and about, being the scary witches they are. That leaves me at home, manning the candy bowl. To make the time pass, I decided to do a little experiment on the trick-or-treaters.

The Experimental Setup

First, I needed a proper setup. I arranged the candy in a single layer, making it easy for the kids to see all their options.

The Method

For each trick-or-treater, I followed a simple two-step process:

1. “Wow what a great costume! You look so scary!”
2. “Would you like to pick a candy?” (I noted what they took first)
3. “Here, have one more.” (I then noted their second choice)

The Hypothesis

My hypothesis was simple: the peanut butter cups would be the clear winner. I mean, they have to be, right?

The Results

Here’s how the first round of choices stacked up:

• Cookies ‘N’ Creme: 10
• Reese’s Cups: 6
• Kit Kat: 3
• Oh Henry: 3
• Aero: 2
• Smarties: 2
• Coffee Crisp: 0
• Reese’s Pieces: 0
• Whoppers: 0

And for the second-choice runoff:

• Reese’s Pieces: 9
• Kit Kat: 5
• Aero: 3
• Coffee Crisp: 3
• Oh Henry: 3
• Reese’s Cups: 3
• Smarties: 3
• Whoppers: 2

The Winner

The overwhelming winner of the night was Cookies ‘N’ Creme!

And in second place, my original hypothesis, Reese’s Peanut Butter Cups.

Kids these days, amirite?