Crowdsourcing Psi: Super Bowl 2015

Mark Boccuzzi

The idea is straight forward: If a group of people is given the chance to vote on the outcome of a future event under blinded conditions, can the simple majority of the participants accurately predict that outcome?
Mark Boccuzzi

I have been interested in the idea of using large groups of people to try to predict future events for some time and every now and then I get a chance to informally play around with it.

The idea is straight forward: If a group of people is given the chance to vote on the outcome of a future event under blinded conditions, can the simple majority of the participants accurately predict that outcome?

Recently, I ran a little, very informal trial to see if a group of people on Facebook could accurately predict the outcome of Super Bowl XLIX.

The method for this test, which was based loosely on an Associative Remote Viewing (ARV) protocol, was simple. I first selected two images. Image selection is really the trickiest part. I needed to find two images that were as different from each other as possible. In this case, I chose a picture of a colorful, round balloon and a monochromatic picture of a pair of rectangular framed glasses sitting on a page of black and white text.

Once the images were selected, I gave them to Julie so she could randomly assign each one to one of the teams playing in the game (the New England Patriots and the Seattle Seahawks). Only Julie knew which image went with which team. Then I created a simple on-line web form to display the images and collect votes. I also randomly selected the order that the images would be displayed on the form. It should be noted that I remained blinded to the image assignments until after the game was over.

After the form was tested, I posted about the test on my Facebook page (the first post was on January 30, 12:52pm Central Time) and the Windbridge Institute Facebook page. Julie also cross-posted the form link on her Facebook page. Over the next couple of days I also posted a few reminders. Some Facebook users also shared the post on their pages and the link to the voting form was picked up and posted on an on-line community forum outside of Facebook.

The post was intentionally vague, as I didn’t want to tip our hand as to the targets.

“Can we accurately predict the future? Let's try! On Monday I'm going to post a picture on my wall. Can you guess what the picture will be?” along with a link to the online form.

The form remained open until just a few moments before the game kicked off on Sunday, February 1.

I ‘m not a big football fan, but Julie is, so we watched the game together.  Overall, the game wasn’t too exciting. Then, in the last few minutes, when it looked like the Seahawks were going to win, there was dramatic turn as the Seahawks lost the ball to the Patriots who were able to maintain their lead and win the game. The last play that cost the Seahawks the game has been called by some the “the worst call in Super Bowl history” 

Here are the results from the on-line guessing form:


72 people started to fill out the form, but only 61 completed it.

We did not collect any demographic information about the participants.

Image Assignments:

Glasses = New England Patriots (winners!)
Balloon = Seattle Seahawks


32 people picked the Glasses (52%) and 29 people (48%) picked the Balloon.

Since the goal was to see if a simple majority could accurately predict the outcome, I consider this test to be successful.


The form service I use tracks the number of seconds it takes a participant to complete a form. For this test, the average completion time for those who chose the image of the Glasses was two minutes and 45 seconds. The average time for those who chose the image of the Balloon was four minutes and 18 seconds. However, this difference is not significant (p=.52 two-tailed t-test).

So, what can we conclude from this little test? Honestly, not too much, but it’s a nice first step. I’d like to conduct more of these on a larger scale and see if this is a robust finding. In the future I’ll most likely expand the voting form to include some basic demographic information along with set of short questions to gauge the participants’ levels of interest in the outcome and belief in the phenomenon being tested.

There is some exciting work being done in area of Associative Remote Viewing. For those who are interested, I’d suggest checking out this video and the recent article “Stock Market Prediction Using Associative Remote Viewing by Inexperienced Remote Viewers” from Smith, Laham, and Moddel which was published in the Journal of Scientific Exploration.