Wimbledon Picks 2012 (Explanation)

With the first week of Wimbledon now a thing of the past, I figured I should actually give my reasoning for my Wimbledon picks.

First, I went with my gut (over data because I didn't have the time) when picking the men's and women's sleepers - players outside the top 10 to go the furthest.  I picked John Isner and Venus Williams who both lost first round.  Guess I shouldn't listen to my gut feeling when making picks anymore.

So let's get to the picks that I actually put some thought into.  I only looked at one stat when choosing my women's champion.  Here are the results of the Williams sisters (both singles and doubles) at Wimbledon since 2000.

Williams Sisters Wimbledon Results 

Year	Doubles Result	Venus Singles	Serena Singles
2000	WON	WON	SF
2001	3RD	WON	QF
2002	WON	RUP	WON
2003	3RD	RUP	WON
2004	-	2RD	RUP
2005	-	WON	3RD
2006	-	3RD	-
2007	2RD	WON	QF
2008	WON	WON	RUP
2009	WON	RUP	WON
2010	QF	QF	WON
2011	-	4RD	4RD

Since 2000, every year that the Williams sisters have played doubles at Wimbledon (highlighted), one of them also wins the singles title (8/8, with both Williams reaching the Final 4 times).  Notice that it doesn't matter how they perform in the doubles, as long as they are playing.  The 4 years that the Williams did not play doubles, often due to one sister being injured, only once did one of them win the singles title (Venus in 2005).  

Serena and Venus entered to play doubles this year (they are currently in the 2nd round).  Looking at past results, the probability that a Williams wins the single title given they play doubles is 1.  So all I had to do was choose between Serena or Venus winning.  I went with Serena because she is the higher ranked and Venus is coming back from injuries.  As of this post, Serena has won her first 3 singles matches and is still alive.

For the men's, I looked at the winner's seed and previous year's performance since 2003 (Federer's first Wimbledon win).

Men's Wimbledon Champions 

Year	Winner	Seed	Previous Result
2003	Roger Federer	4	1st
2004	Roger Federer	1	WON
2005	Roger Federer	1	WON
2006	Roger Federer	1	WON
2007	Roger Federer	1	WON
2008	Rafael Nadal	2	RUP
2009	Roger Federer	2*	RUP
2010	Rafael Nadal	2	-
2011	Novak Djokovic	2	SF

* Federer was the 2 seed, but he was technically the top seed after Nadal withdrew as the number 1 seed with injury before the 2009 tournament began.

Because Federer, Nadal and Djokovic have won 28 of the last 29 major titles, I have eliminated the rest of the field.  I eliminated Federer because it has been one of the top 2 seeds to win every year since 2004.  I went with Djokovic over Nadal because the defending champion has won 6 of the last 9 years (if I consider Federer the defending champ in 2009 since Nadal withdrew and Nadal the defending champ in 2010 since he had not defended his title.**)  Also, Federer and Nadal each defended their first Wimbledon title by winning the following year that they played.  This gives me belief that Djokovic will continue this trend and win his second straight Wimbledon title.  

My men's pick is looking even better after Nadal lost in the second round.  (Still, I'll be cheering for Roger Federer to win the title as he is my favorite player.)

** Under this argument, both Federer and Nadal would have been considered defending champions in 2010, so you could change the fraction to 6/10 defending champs winning. 

MIT Sloan Sports Analytics Conference

I came across a great article today (via Simply Statistics) describing the largest sports analytics (i.e. statistics) conference in the world, held at MIT.  A few points that I found most interesting:

1. The guy that they highlight (Kirk Goldsberry) created the plot of the week last week.
2. They mentioned that they had a panel on tennis analytics for the first time this year, including Pete Sampras and Roger Federer's coach Paul Annacone and former player Todd Martin.  Two great quotes:

On why tennis analytics is lacking: "There's no shared service," says Martin.  "This isn't a team sport with a $500,000 budget for analytics."

"[Analytics] should be a strength of our game.  Tennis is a game of patterns." - Craig O'Shannessy

This conference could be worth a vacation next year.  Might depend on whether I have one or two great blog posts to report on.

Plot of the Week 8

I've been dreaming up some cool plots to show off on my blog, but as I've mentioned before, thinks are a little crazy for me right now.  So this week I am borrowing another neat visualization that I found on the web.

http://thepowerrank.com/visual/NCAA_Tournament_Predictions

This is an interactive plot: put your mouse over your favorite team to see their probability of winning it all (based on this group's predictive model).  Or put your mouse over a specific game to see the probability that each team will win that game.  This visualization relies on java, so if its not working for you, blame your web browser!

Wimbledon 2012 predictions

Wimbledon starts on Monday and I wanted to announce my picks (I was 2/2 for the French Open).  

I am going to follow ESPN's format for their picks (which you can see here):

Men's winner: Novak Djokovic

Men's sleeper (someone outside of the top 10 to go the furthest): John Isner

Men's toughest road (top 10 player to lose earliest): Janko Tipsarevic

Women's winner: Serena Williams

Women's sleeper: Venus Williams

Women's toughest road: Samantha Stosur

Things are very busy for me, so I do not have the time to justify my picks right now - but I hope to later in the week.

Plot of the Week 7

Sorry for the delay in posting.  I started my new postdoc position yesterday and am still trying to get settled in.  Here is an interesting chart showing the NBA players with the lowest shooting percentage in different areas of the court.  Enjoy!

 http://courtvisionanalytics.com/behold-the-worst-shooters-of-2012/

Pissing off an NFL player

Chicago Bears cornerback Charles Tillman does not like your pro-Packers math!

1. None of my homework or exam questions ever elicited this type of response.
2. Question on her next hw assignment: If 100 NFL players attempted this stat problem, how many would answer correctly?  

French Open Picks: Follow Up

A few weeks ago, I made my picks for the 2012 French Open champions (Part 1 and Part 2).  Using my statistical wizardry (ok, more like logic and process of elimination), I correctly picked Rafael Nadal as the men's champion and Maria Sharapova as the women's.

Compare this to the ESPN expert picks: 11 of the 12 correctly picked Nadal, but only Chris Evert picked Sharapova.  Maybe if I keep this up for Wimbledon and the US Open, ESPN will start knocking on my door.

Plot of the Week 6

Yesterday, Union Rags won the 2012 Belmont Stakes.  For this week's edition of Plot of the Week, I wanted to look at the winning times at the Belmont Stakes (since 1926, when the current distance of 1.5 miles was established).

 The red circles designate that the betting favorite won, and the blue +'s show the horses that won the Triple Crown in that year.  The green and magenta lines show the average and median winning time, respectively.

Some interesting takeaways:

1. The record time is 144 seconds, run by Secretariat in 1973.  No horse has gotten within 2 seconds of this winning time.
2. There is an overall downward trend in time, although winning times seem to have leveled off in the past few decades.
3. There were 2 horses that won the Triple Crown but were not favorites at the Belmont Stakes.
4. The past 3 Belmont Stakes winners have had slower than average times.
5. The betting favorite has not fared well lately - only 7 of the favorites have won since 1980.

Now let's look at the betting odds of the winning horse.

1. Obviously, all of the favorites that won had low betting odds. 
2. The 2 horses that won the Triple Crown but were not the betting favorites still had very low betting odds.
3. There have been many more long-shot winners (> 20:1 odds) in the past 15 years (5) than all previous years (3).  I don't know if this has to do with the format of the race (more horses allowed to run) or there is greater disparity in horse racing today.

Did I miss anything?

Check My Stats

Anthony Davis rocking a "Check My Stats" t-shirt.  Image borrowed from following link.

Robinson not buying Davis hype

1. Where can I get this shirt?

2. I'm thinking that if Robinson wears a shirt reading "Numbers Don't Lie", another NBA prospect should wear one saying "Torture Numbers and They'll Confess to Anything", a quote from Gregg Easterbrook.  Maybe someone who had an underwhelming college season -  Harrison Barnes?