Data, Actually

I feel it in my fingers, I feel it in my toes. 

Christmas is in the air and someone, somewhere is watching Love, Actually. Last year, David Robinson at Variance Explained took the entire script and analyzed the social networks within the film. 

With his interactive shiny app you can select a scene and see which characters have interacted with each other up to that point in the movie.

Early on, you see multiple distinct story lines developing.

Scene 7

Scene 7 - Love Actually

As the movie goes on, more and more characters interact.

Scene 38

Scene 38 - Love Actually

By the end of the movie, the social network has become completely entangled, when most of the characters come together at the school Christmas concert.

Scene 57 

Scene 57

Robinson shares his analysis along with all of his R code and even wraps it all up with a nice message. Find it here.

Dater’s Index

Emily McDowell at chronicled her experience with online dating with her Dater’s Index. Here’s a selection:

  • Number of first dates I went on from the Internet between January 2009 and November 2010: 40
  • Percentage change over the previous decade: +3,900
  • Estimated percentage of profiles containing the description “fun-loving”: 80
  • Estimated percentage of people who do not actually love fun: 0
  • Estimated percentage of single people in Los Angeles who both work and play hard: 85
  • Number of first dates resulting in trips to the E.R.: 1
  • Estimated number of times I swore off online dating: 5
  • Number of first dates it took before meeting the right one: 39
  • Number of times we flaked out on each other before we actually met: 3
  • Time, in months, that we have now been happily living together: 9
  • Likelihood, in percent, that I will tell you online dating is the best thing ever: 100

Find the full version here.

Awkward Celebrity Couples

In this post, we’ll take a look at how some famous couples stack up against the age rule of thumb, mentioned in an earlier post. For reference, here are the equations:Dating range calculationsAccording to the rule of thumb, dating someone older than your max or younger than your min would be considered objectionable.

In the graph below, each line represents a relationship. If the line falls within the blue zone, the age difference of the couple was socially acceptable for that portion of their relationship.


If the graph is confusing to read, hopefully the following diagram helps:How to Read the GraphFor example, looking at the pink solid line for Demi Moore and Ashton Kutcher, the coordinates for the circle are [42,27], so they got married when Demi was 42 and Ashton was 27. The pink triangle at the other end of the line means the relationship ended in divorce. The coordinates for the triangle are [51,35], so their relationship ended when Demi was 51 and Ashton was 351.

Hugh Hefner

Hugh Hefner, the king of icky relationships, almost made it into the zone of social acceptance with his 20 year relationship to second wife, Kimberly Conrad, before their divorce. In his current marriage, 90-year-old Hefner would need to stay alive and married to 30-year-old Crystal Harris until he’s 134 and she’s 74 for the couple to cross into the blue zone.

Woody Allen and Soon-Yi Previn

The 35 year age difference between Woody Allen and his wife, Soon-Yi Previn, isn’t the only thing creepy about this relationship. Woody first became involved with then 21-year-old Soon-Yi when he was still in a relationship with Soon-Yi’s adopted mother, Mia Farrow. He had even adopted some of Soon-Yi’s younger siblings. 


At age 40,  Sun Myung Moon, the controversial founder of the Unification Church (the church I was raised in) married his wife when she was just 17. During their 52 year marriage, they had 13 children with varying degrees of craziness, the latest iteration being their youngest son’s arms manufacturing business with endorsement from Donald Trump’s son (read more here).

Donald Trump

Surprisingly, of all the couples, Donald and his wife Melania have the least objectionable partnership, at least based on age alone. They’ve been safely within the blue zone for most of their relationship. Perhaps what makes them an awkward couple is their mismatched levels of attractiveness. (Or because everything about Trump is objectionable.)

Demi Moore

Demi (Guynes) Moore married Freddy Moore when she was 17 and he was 29. Later she switched sides, marrying 27 year old Ashton Kutcher when she was 42. For most of their relationship, the age difference between Demi and Ashton was very close to the blue zone. This could suggest that their awkwardness as a couple was because she was an older female dating a younger male, rather than their relative age difference alone, further evidence that the rule of thumb could use some adjusting.

1. Note: Ages are estimated from Wikipedia, which often only lists the year of marriage or divorce, rather than the exact date. Exact ages at marriage or divorce may be slightly off because of this.

The Code


age_plot() #see "Calculate Your Dating Age Range" post for code

dark_blue <- rgb(68,84,106, max = 255)
blue <- rgb(96,147,125, max = 255)
yellow <- rgb(217,192,7, max = 255)
purple <- rgb(122,98,145, max = 255)
pink <- rgb(247,190,202, max = 255)
hot_pink <- rgb(201,6,45, max = 255)

#plot couples

####Hugh Hefner and Kimberly Conrad
add_seg(26,63,47,84, end = "divorce", col = dark_blue, 
     lty = 'dashed')

####Hugh Hefner and Crystal Harris
add_seg(26,86,30,90, end = "", col = dark_blue)

####Woody Allen and Soon-Yi Previn
add_seg(26,61,45,80,end="", col = purple)

####Rev. and Mrs. Moon
add_seg(17,40,69,92, end = "death", col = yellow)

####Donald Trump and Melania Trump
add_seg(35,58,46,69,end="",col = blue)

####Demi Moore and Freddy Moore
add_seg(17,29,22,34, end = "divorce", col = hot_pink , 
     lty = 'dashed')

####Demi Moore and Ashton Kutcher
add_seg(42,27,51,35,end = "divorce", col = hot_pink)


add_seg <- function(x1,y1,x2,y2,end="",...){
     segments(x1,y1,x2,y2,lwd=1.5,...) #plot line segments
     points(x1,y1,pch=16,...)          #plot left endpoint
     #add endpoint
     if (end == "divorce") points(x2,y2, pch = 17,...)
          else (
               if (end == "death") points(x2,y2, pch = 15,...)

celeb_legends <- function(){
     #empty plot
     plot(1, type="n", axes=FALSE, xlab="", ylab="")
     #main legend
          c("Hugh Hefner and Crystal Harris",
          "Hugh Hefner and Kimberley Conrad",
          "Woody Allen and Soon-Yi Previn",
          "Sun Myung and Hak Ja Han Moon",
          "Donald and Melania Trump",
          "Demi Moore and Freddy Moore",
          "Demi Moore and Ashton Kutcher"), 
          lty=c(1,2,1,1,1,2,1), lwd=2,
          inset = .02, bty="n")
     #endpoints legend
          c("marriage", "divorce","death"), 
          col=hot_pink, pch = c(16,17,15), 
          inset = .02, bty = "n")


OkCupid – 6 Weeks In

OKC Pic.jpg

Six weeks ago, I signed up for OkCupid. It’s my first attempt at online dating and also the first time I’ve actively tried to date.

Here’s how the numbers stack up:

Number of first messages I received:………..61
Number of first messages I sent:……………….0
Number of first messages I responded to:..8    (13%)
Number of first messages I received
that were one word long: ………………………….12    (20%)
Number of one word long first
messages I responded to:…………………………..1    (8%)
Total number of likes I got: ……………154
Total number of likes I gave: ………….4
Total number of 1st dates: …………….3
Total number of 2nd dates: ……………3
Total number of 3rd dates: …………….1
Total number of 4th dates: …………….1
Total number of 5th dates: …………….1
Total number of 6th dates: …………….1

Total number of one word long first messages that I responded to
that resulted in a 6th date: ………………….1  (100%)

So far, online dating has been surprisingly easy. I went on dates with three entirely decent guys. I’ve had no bad dates and one is going unexpectedly well! It’s still really early and anything could happen, but I’m excited to see where it leads.

Where are all the single men?

According to an interactive map from Jonathan Soma, they’re everywhere.

Single men outnumber single women across the country up until age 35.



From age 35 onward, the balance starts shifting toward more single women until about age 50, where single women strongly outnumber the men.


Play around with the interactive version here and see for yourself. You can adjust the slider to change the age range.

Note: From what I can tell ‘single’ here means ‘never been married’, rather than ‘not currently in a relationship’.