Tuesday, March 13, 2012

Until Math Do Us Part: John Tierney's Refined Prediction Model for Celebrity Marriages

Click to enlarge.
In 2006, New York Times columnist John Tierney published the equation above that was developed to calculate the probability that a celebrity couple would stay married.

The variables in the mathematical model included: the relative fame of the husband and wife, their ages, the length of their courtship, their marital history, and the sex-symbol factor (determined by looking at the woman’s first five Google hits and counting how many show her in skimpy attire, or no attire). 

Now, with more than five years of follow-up data and empirical testing, John Tierney reported in yesterday's New York Times that there is firm statistical support for the "Sundem/Tierney Unified Celebrity Theory." For example:

"The 2006 equation correctly predicted doom for Demi Moore and Ashton Kutcher; Pamela Anderson and Kid Rock; and Britney Spears and Kevin Federline. It also forecast that Will Smith and Jada Pinkett would probably not make it to their 15th anniversary, in December 2012; so far, they’re still married, but gossip columns are rife with reports of a pending split.

On a happier note, the 2006 equation identified two couples with a good chance to make it to their fifth anniversary, in 2010: Ben Affleck and Jennifer Garner, and Matt Damon and Luciana Barroso. Sure enough, they made it (and are still married)."

But now Tierney and his co-author Garth Sundem have updated and refined the celebrity marriage prediction formula, and believe it now has even greater predictive power.  John explains:

"Whereas the old equation measured fame by counting the millions of Google hits, the new equation uses a ratio of two other measures: the number of mentions in The Times divided by mentions in The National Enquirer.

“This is a major improvement in the equation,” Garth says. “It turns out that overall fame doesn’t matter as much as the flavor of the fame. It’s tabloid fame that dooms you. Sure, Katie Holmes had about 160 Enquirer hits, but she had more than twice as many NYT hits. A high NYT/ENQ ratio also explains why Chelsea Clinton and Kate Middleton have better chances than the Kardashian sisters.”

Garth’s new analysis shows that it’s the wife’s fame that really matters. While the husband’s NYT/ENQ ratio is mildly predictive, the effect is so much weaker than the wife’s that it’s not included in the new equation. Nor are some variables from the old equation, like the number of previous marriages and the age gap between husband and wife.

In the fine tradition of Occam’s razor, the new equation has fewer variables than the old one. Besides the wife’s tabloid fame, the crucial ones are the spouses’ combined age (younger couples divorce sooner), the length of the courtship (quicker to wed, quicker to split), and the sex-symbol factor (defined formally as the number of Google hits showing the wife “in clothing designed to elicit libidinous intent”)."

Here is a list of some of the predictions using the new model, e.g. Prince William and Kate Middleton have a 71% chance of being married in 15 years, while Jennifer Lopez and Ojani Noa have only a 1% chance.   

HT: Morgan Frank


At 3/13/2012 4:42 PM, Blogger Ironman said...

That's the old equation! The new one is here, and also coming soon as a new tool over at Political Calculations!... In the meantime, here's our review of Garth Sundem's "Brain Trust".

At 3/13/2012 9:54 PM, Blogger PeakTrader said...

I wonder if comedians stay married longer?

At 3/13/2012 11:29 PM, Blogger juandos said...

"“Research has documented that women who wear skimpy or sexually provocative clothing tend to be higher on the trait of narcissism,” says Dr. Buss, a psychologist at the University of Texas"...

Hmmm, probably an excellent vector for some new and barely curable STD also...

At 3/14/2012 10:02 AM, Blogger morganovich said...

garth went to HS on bainbridge with my co-portfolio manger's wife.

he's an interesting guy and i love the way he looks at data (like swapping out google hits for ny times mentions divided by enquirer mentions to get a measure of fame toxicity, which is brilliant) and that fact that he publishes predictions and then evaluates his success.

the number of "bold predictions" i see about markets and the economy that then disappear and never get mentioned again when they go wrong is outlandish.

there's a whole newsletter industry based on tossing predictions against the wall then crowing about the one of ten you got right.

garth and co seem to have actual integrity.


Post a Comment

<< Home