There are very few things in life more boring than hearing someone talk about their fantasy football team….so feel free to tune out now.

Good grief has my first foray in to FF been a disaster….but an excellent example of how picking your data points can change the results.

For those of you not well versed in standard FF setups, each week your team plays another team in your league.  Your teams do not reflect real NFL teams, but rather new teams composed of existing players.  Your record is determined by how often you beat the team you’re playing that week.

In my league of 8 (run by the AVIs son, btw), I’m dead last.

However, I’m 3rd for points scored this year.  In fact, I’ve scored only 30 points less than the first place team….and 343 points more than the team directly ahead of me in the standings.   My problem of course is how many points my opponents score when they play me.  I have had more points scored against me than any other team by almost 100 points for the season.

Sigh.

This week is the first round of the playoffs, and I’m projected to lose yet again.  While I love the stats part of fantasy sports, it’s really much better to be lucky than good.

http://embed.ted.com/talks/ben_goldacre_what_doctors_don_t_know_about_the_drugs_they_prescribe.html Apparently I can’t get Ben Goldacre’s new book until February, so I only have this TED talk to hold me over until then.

Best quote “If I conducted one study, and withheld half the data points from that one study, you would rightfully accuse me of research fraud.  And yet for some reason, if somebody conducts ten studies, but only publish the 5 that give the result they want, we don’t consider that misconduct.”

Blogger tells me this post will be my 200th, so it seemed like a good time to go a little meta and reflect on my own statistics since I went live on March 21st.

Most popular posts:

#1 My most popular post didn’t surprise me, it was the one where I correlated retraction rates in scientific journals with conservatives decreasing trust in science.  That one got linked to/reposted on quite a few blogs, so it didn’t surprise me too much.

#2 The second most popular is a little strange, I still haven’t figured out what key words keep leading people to my 4th of July post….most of that’s just a repost from the Census Bureau.

#3 My third most popular post is my feelings on the application of Title IX to STEM professions.  It’s pretty funny because that’s the only post I’ve ever done that my brother ever got actively upset at me over, and it ended up as required reading for a class at a community college in California on gender issues.  I considered emailing the professor to ask what the discussion around it was, but I wasn’t sure I wanted the answer to that question.

The rest:

#4 5 Rules for Reading Scientific Papers Online
#5 Soviet Propaganda, Infographic Style
#6 Arguments and Discussions, learning the rules
#7 Mission Statement
#8 Rule 6D
#9 Are Republicans Stupid?
#10 Rule #6

#9 makes me laugh because “are republicans stupid” is actually the most popular search that brings people to my blog (excluding searches for my blog in particular)…I don’t think that post gives them what they’re looking for.  Relatedly “gas prices the day bush took office” also brings me some traffic.

I’m happy to report 4% of my traffic comes from Linux users (stay strong my friends!)

Most popular countries are:
USA
Russia
Canada
Iceland
France
UK
Indonesia
Ukraine
Germany
Australia

I suspect most of the Russia traffic is spammers, and probably Ukraine as well.  Not sure about the rest.

The correlation between the number of posts I put up in a month and the amount of traffic I get is .68, but it drops to .53 if I exclude March as a partial month.

I’d be interested to hear any thoughts on this, and as always any directions for the future!  Thank you all for making this an entertaining 200 posts, and I look forward to the next 200!

The Assistant Village Idiot is doing a series on fashionable politics. 

I had a rather entertaining post about doomsday prep all set for today, but then I looked at the calendar.

Six years ago today, at a trivia night, I got introduced to a guy who perfectly complemented my trivia strengths and weaknesses.  I knew I must get this delightful person (who could always remember who was in what movie, or what musician did what song and when) to be on my trivia team for as long as possible.  When he beat me at Trivial Pursuit, I knew I had to marry him.

The chances of love at first sight are small, and the chances of finding someone who would put up with someone who does literature searches and statistical breakdowns of optimal household and relationship management are even smaller, so I feel pretty darn lucky to have him in my life.

While the genders are reversed, I like the way this video puts it (linked to in case the embedding doesn’t work).
http://videosift.com/widget.js?video=226713&width=540&comments=15&minimized=1
Put more simply (from Andrew Gelman‘s whole post on the topic):
You are perfect; I’d make no substitutions
You remind me of my favorite distributions
With a shape and a scale that I find reliable
You’re as comforting as a two parameter Weibull
When I ask you a question and hope you answer truly
You speak as clearly as a draw from a Bernoulli
Your love of adventure is most influential
Just like the constant hazard of an exponential.
With so many moments, all full of fun,
You always integrate perfectly to one.


Love you honey!

It occurred to me recently that it’s a bit odd that most of my “real world” exposure to research comes in the form of the variety of clinical trials that go on around me on a regular basis, and yet I rarely comment on clinical trials.

This is probably because most clinical trials are a little dense to get through, and the results tend to be less interesting to people (it turns out the reuptake limitations actually weren’t as dramatic as they made them out to be!) and there’s rarely much media involvement to mix things up.

Anyway, I heard a tidbit recently about when to be suspicious of results of clinical trials that I thought I’d pass along.

In any trial assessing a new treatment/drug/etc vs a placebo, you would expect to see more dropouts in the “treated” arm of the study.  This, of course, is because most drugs/treatments have very real side effects that will bother people and cause them to drop out.  Therefore, if you see a trial where the dropout rate is higher in the placebo arm, you should be suspicious.  Placebo studies should almost always be blinded for the patients (and ideally for the providers), but if significantly more of those in the placebo arm drop out, you know this has gone wrong.  Patients don’t keep showing up if they know they’re not actually getting treated with anything…and once we’ve established that the patients know which arm of the study they’re in, the results become much less reliable.

I thought that was an interesting tidbit to keep in mind.

Do you like Johnny Cash?  Do you like data visualizations? Ever wondered how far he travels in “I’ve been everywhere man?”

The answer is 181075 kilometers.

Thank you internet.

FYI, I’m done with 75% of my Christmas shopping.  Still have to get a tree though.

For those of you not done yet, I’ll help you out with what to give me.   Here’s a whole list!  

And for my little genius baby, I’m thinking this “Outlier” bodysuit would be perfect….or perhaps a stuffed normal distribution?

Alright, enough shopping.  Need some entertainment?  Try the “thanks textbooks” tumblr. Featuring the best of the worst problems/examples/etc in textbooks.  Highlights in the commentary include “I’m less concerned with the question, “What does the scale read?”  and more concerned with the question, “Why the hell are we lubricating a hamster?”  and “Who has a “favorite” orange?  How long have you had this orange that you’ve bonded with it so much?  Who has an equation to calculate the weight of an orange?Is it your favorite because it happens to weigh nine pounds!?”

I wanted to put up a brain teaser yesterday, but the little one got his first cold.  Baby coughs are sad.

Anyway, one of the more famous statistical brain teasers is the birthday problem.  There are a few variations, but essentially the question goes something like this:

You’re at a party with 23 guests, including you.  What are the chances that  two people there have the same birthday?

The trick of course is that no one has to have a specific birth date, so the answer is not 23/366, but instead around 50% (interestingly, if the party were 50 people, it goes up to 97%).    For a further explanation, see here.

What’s interesting about this problem is that you have to assume every birth date is equally likely…which of course isn’t true.  I’ve written before about uneven distribution of birthdays in the US, due in part to scheduled c-sections or induced labor.  Anyway, I saw an interesting heat map today of birthday distributions from the Daily Viz, which is what got me thinking about the brain teaser.

 To note, this chart was made from a list of ranked birthdates, which is here.

I was a little struck by this, because I was thinking about how terrible I am at estimating things like this on my own.  The most common birthday in my circle of friends/family is Halloween.  The first week in April has the birth dates of my mother, sister and husband.  Neither of those time frames are overly popular within the general population, although I’d guess the difference between “most popular” and “least popular” are relatively small.  It was a good reminder that those I spend the most time with are not terribly representative of the population in general, on average.