Category Archives: Uncategorized
Yesterday I posted the probabilities of Biden winning each state in terms of a field goal distance. But I only did it for a few swing states, and I think this idea deserves more. So with the help of Mike Lopez’s post from last May, I put the field goal distances for each state on a football field. The distance in the plot is in terms of Biden winning the state and he is kicking towards the bottom of the plot into Trump’s end zone. The horizontal location of the state name is roughly alphabetical order with a few manual tweaks to make it look pretty. And the color of the state abbreviation is the projected vote share for that state.
So let’s look at an example of how to interpret this. Let’s take Ohio. According to fivethirtyeight, Biden has a 37.9% chance to win Ohio. The probability that an NFL kicker makes a 61.5 yard field goal is about 37.9% so Ohio is plotted just beyond the 50 yard line (51.5 yards + 10 for the end zone).
Here is the table of win probabilities and the corresponding field goal distances for each state.
Let’s hope Biden doesn’t double doink this election.
Code is available here.
UPDATE: Based on some suggestions that I have received from Ron Yurko.
Also, you can only post stuff related to politics on Thursdays?!?! That’s the strangest rule I have ever hear.
UPDATE: Based on suggestions from @alargewop.
UPDATE: Based on final suggestion by Ron Yurko.
I’ve converted FiveThirtyEight win probabilities into field goal yardages.
So for example Arizona – 54.75 means that the probability that Biden wins Arizona is about the same as a kicker in the NFL making a 54.75 yard field goal. Let’s look at the swing states in terms of how likely Biden is to win the state
- Arizona – 54.8
- Colorado – 40.5
- Florida – 52.8
- Georgia – 63.6
- Iowa – 65.4
- Maine (Statewide) – 45.5
- Michigan – 43.2
- Minnesota – 47.0
- Nevada – 45.5
- New Hampshire – 48.9
- New Mexico – 35.29
- North Carolina – 57.9
- Ohio – 61.5
- Pennsylvania – 49.4
- Virginia – 33.0
- Wisconsin – 48.9
The longest field goal in NFL history was 64 yards.
Biden winning Oregon is a 35.3 yard field goal
To put this in perspective, Biden winning California and Oregon are about 12 and 35.3 yard field goals respectively. Biden winning Indiana and Wyoming are about 85.4 and 102 yard field goals, respectively.
Finally, overall Biden’s chances of winning right now are about a 48.9 yard field goal. Sometimes kickers miss 48.9 yard field goals!
So, a while ago I tweeted out asking for podcast ideas. Didn’t really think I’d get anything too interesting back, but David Hess had a great suggestion: I go to a random wikipedia page and then just see where it goes. I think this is a brilliant idea, and I’m stealing it for myself. I’m calling this “Rabbit Hole” and it will be broadcast live on Twitch. It’s perfect for me because it takes absolutely no planning and I can just do it whenever. I did it for the first time last night. (Full video is here). Here is a recap of last nights episode.
We started here: Tideglusib
Tideglusib is a GSK-3 inhibitor. Which apparently has something to do with Bipolar disorder. This made me looks up why they changed the name from “manic depression” to “bipolar disorder”
That page led me to…..the Animaniacs.
Where I watched the 50 state capitals video and the Nations of the World video. In that second video they sing the lyric “both Yemens”. Apparently from 1967 to 1990 there were two Yemens! I had no idea!
Note: Animaniacs aired from 1993 to 1998. So the song was WRONG.
I then wondered who did the voice of Yakko. It’s a guy named Rob Paulsen. He was also the voice of the Teenage Mutant Ninja Turtle Raphael! He also the guy from this classic “Got Milk” commercial! (He was a Hamilton fan before it was cool……)
Then this picture happened and I needed to look up Anthrocon.
Here is the Fursuit parade from Anthrocon 2019. It’s…..something.
So then I started reading about furries, as one does…..and apparently, there is some actual academic work being done studying furry culture. I found and started reading this paper from the journal Animals and Society. Which led me here when I was trying to figure out the difference between pan-sexual and omni-sexual.
In that furry study, they were doing some statistical tests to see if there were differences in proportions of gay, straight, and bisexual people who are involved in furry culture. What they tested was if all the proportions were the same. I didn’t think that made sense and it would have been better to compare to the proportions in the general population. So I went and looked up what the proportions are in the general population. Interestingly, Americans tend to over estimate the size of the U.S. gay population. (Guess what it is an see how close you are. I wasn’t too far off, but I did over estimate it).
This led me to thinking about how you estimate the size of hard to reach populations and I remembered seeing work done in the past about leveraging social networks to estimate population size. Here is some work by Tyler McCormick on this exact topic: Network-Based Methods for Accessing Hard-to-Reach Populations Using Standard Surveys.
Then the baby started crying. I’m definitely doing this again. I had a lot of fun doing it. I don’t even care if you watch. But I’d prefer if you did.
This is a work in progress. Who else belongs on this list?
- Dr. Phil
- Clay Travis
- Warren Sharp
- Alex Berenson
- Jerry Falwell, Jr.
- Gwyneth Paltrow
- John Edward
- Jim Carey
- Ben Shapiro
- Sean Hannity
- Bill O’Reilley
- Jenny McCarthy
- Donald Trump
- Eric Trump
- Donald Trump, Jr.
- Ivanka Trump
- Jared Kushner
- Mike Pence
- Robert DiNero
- Robert F Kennedy, Jr.
- Practitioners of cupping
- Practitioners of acupuncture
- Flat Earthers
- Faith Healers
- Holocaust Deniers
- Men’s Rights Activists
- Proud Boys
- Richard Spencer
- Gavin McInnes
- Rush Limbaugh
- Alex Jones
- 9/11 Truthers
- Darrren Rovell
- Rovell, D’Nesh D’Souza
- Climate Change Deniers
- James Woods
- Chuck Woolery
- Laura Ingraham
- Tomi Lahren
- Shiva Ayyadurai
- Kevin Hassett
- Joe Walsh
- Richard Epstein
- James O’ Keefe
- Ryan Fournier
- Mike Cernovich
Group A, dubbed the “group of death” by basically anyone who closely follows sports analytics Catan players, saw @chuurveg win the first two games of the group making the third game nothing more than a victory lap for @chuurveg. Though he almost pulled off the sweep in round one, @sethwalder pulled out a victory in game 3 to advance to the round of 16.
@jacksoslow and @yuorme took all the wins in this group with the only other competitor finishing with a second @scott_bush.
@dangroob tore through group C posting 3 wins in his 3 games. @chibearsstats also advance on the back of 3 second place finishes. Fan favorite @statsinthewild disappointed throughout the first round. Needing a win in his third game to advance, all he could manage was 5 points and a tie for second. He’s going to have to answer a lot of questions about whether he even deserved to be in this tournament in the first place.
@cfbnate and @hausinthehouse advance out of this group in what was probably the most exciting group of the first round. Both of the first two games of this group saw ties for second at 9. After two games @stat_ron was sitting on 18 victory points with no wins. He then finished dead last in game three to not advance. His 23 victory points in the first round are the most of anyone who did not advance (tied with @kfloyd34 from group H).
Group E saw three different winners in three different games. @zbinner_NFLinj one the first game of the group and, like an injured NFL player referenced in his screen name, limped across the finish line to somehow win the group with only 22 victory points. @mtm1013 won game 2 and after his 2nd in game 1, he was very likely to advance to the next round barring some craziness in game 3. Game 3 saw @chiefsanalytics, whose Twitter name is not @arrowheadanalytics, went big in game 3 for the win, but it wasn’t enough to advance. Kansas City will just have to be happy with their Super Bowl win this year because the city won’t be bringing how a Settlers of Catan Championship this year.
Wow. What a wild ride this group was! @punkrockscience wins the group after putting up a 3 spot in game 1. Even more bizarre, she won the group with the lowest victory point total of the group (21, 22, 23, 23)! She is the only competitor to win their group and NOT have the most total victory points. Additionally, @sabinanalytics waited until the very last minute to advance by winning game 3 leaving game 1 winner @nflsharptampa as one of only two people to win a game and not advance (@chiefsanaltics is the other). @necesarypaper, who ultimately finished 4th in the group, was actually in really good shape to win the group going in to game 3 as a win in game 3 would have put him in first as he had finished second twice already and would have moved him into first place all alone with a win. @neccessarypaper ends their tournament as the only player to score 7 or more points in all three of their games and not advance out of the first round.
With one game left to go in Group G, @jPohlkampHart has already won the group, but that coveted second place is the last spot available for the round of 16 and any of the other three in the group (@msubbiah, @dtrainor4, and @nickwan) can still claim it. Game 3 will be win an advance for the three unqualified contenders. If @JPohlkamHart can win game 3, @msubbiah will advance unless @dtrainor can finish alone in 2nd AND make up the 3 victory deficit behind @msubbiah. @nickwan, the worlds greatest Catan streamer, needs a win and only a win to advance.
Last, and kind of least, group H featured @sethpartnow dominating the first two games of Group H, making the third game in the group a exhibition game for Seth. The second spot was up for grabs with a win an advance scenario for the other the other three in this group. Ultimately, @Hharris3419 took the other spot with a 10,9,9,6 win in the final game to move on.
Here is a summary of the first round (color is what place they finished in the group (gold, silver, bronze, and red)):
Round of 16
Group I features the winners of group A and B and he runners up in groups C and D.
Group J features the winners of group C and D and he runners up in groups A and B.
Group K features the winners of group E and F and he runners up in groups G and H.
Group L features the winners of group G and H and he runners up in groups E and F.
Here is the email that I just sent to Bob Pittman, CEO of iheartmedia.com, and Eric Shanks, Fox Sports CEO. I encourage you all to send emails to Bob and Eric expressing your dismay for giving Clay Travis a platform to spread this type of dangerous misinformation about coronavirus. You can find their emails here:
Bob and Eric,
I’m writing to you today in regards to Clay Travis. Recently, Clay has been writing articles (such as this: https://www.outkickthecoverage.com/coronavirus-infections-are-likely-to-peak-next-week/) related to Coronavirus that contain misinformation about the pandemic. Given his large readership, this is incredibly irresponsible of your companies to give him a platform to spread dangerous falsehoods about the disease. When he makes “hot takes” about sports, it doesn’t matter whether he is right or wrong. But this is literally a matter of life or death. As a result, I’m asking you to please stop giving Clay an outlet to spread misinformation about the worst crisis of our lifetimes. Let him write about politics or sports or whatever else he wants. But please stop giving him an outlet to spread his Coronavirus falsehoods.
For reference, I’ve written a blog post with everything that Clay got wrong in his article from March 18 where he seriously downplays the risk of this virus: https://statsinthewild.com/2020/03/30/clay-travis-said-another-stupid-thing-and-this-time-it-actually-matters/
Gregory J. Matthews, Ph.D.
Director of Data Science Program
Department of Mathematics and Statistics
Here is a tweet I saw today. It’s from April 7. It’s from Tomas Pueyo, the author is this medium post on Coronavirus from March 10. (Tomas Pueyo is not a public health expert). When I first read this, I thought, wait that doesn’t seem right and I tried to reproduce his numbers. To get to 36%, Pueyo is simply dividing 1.8/5 = 0.36. But is that right?
What Pueyo is trying to do here (I think) is calculate the conditional probability that someone dies given that they are in the ICU. That is:
By the definition of conditional probability, this is equal to:
So based on this calculation, Pueyo is implying that P(Death and ICU) = 0.018 and P(ICU) = 0.05. However, he is using an estimate of P(Death) rather than the correct probability of P(Death and ICU) in his numerator. While these numbers may be close, they are almost certainly not equal to each other:
The only way that these two numbers would be equal to each other is if P(ICU|Death) = 1. This seems unlikely. There are certainly people dying from Covid-19 that never make it into the ICU.
So even if Pueyo’s number are completely correct, this calculation is almost certainly an over estimate based on basic intro probability rules.
And I’d like to add two additional points:
- We don’t actually know what the mortality rate is of Covid-19 because we don’t have accurate counts of either the numerator (how many deaths) nor the denominator (how many cases) in a mortality rate calculation. So this number has a ton of error in it.
- We already know that age is a huge predictor in mortality rates for Covid-19, so to not include that in the calculation of probability of death of a specific individual of known age makes no sense.
So basically this entire calculation is meaningless. So my advice is let’s stop playing armchair epidemiologist, especially if you can’t even get very basic probability rules correct!