Predicting stable marriage

So I stumbled across this story referenced on slashdot where Randal Olson did a fascinating predictor analysis of what makes a stable marriage. After digging a little further, I noticed that Randal took information from the paper "‘A Diamond is Forever’ and Other Fairy Tales: The Relationship between Wedding Expenses and Marriage Duration" by Andrew M. Francis and Hugo M. Mialon and plotted the data.

I do not mean to trivialize Randal's effort at all. In fact, one could say that his effort is very noble in the following sense. He takes a seemingly complex paper and summarize it with a few plots. Story telling with graphs and plots these days appear to the most efficient way to get people's attention. Remember Ross Perot's ill-fated presidential campaign? He was successful in telling his side (and plan) with lots of plots and graphs.

Maybe there is something behind data science afterall...