Friday, September 21, 2018

New Sports and Math Post On Darply and an Update on Trump's Approval Ratings

I have a new post on sports and statistics on the news site Darply. Featured prominently in this article is former Baltimore Ravens center John Urschel who retired from the NFL to pursue a PhD at MIT in Mathematics.  You can see another interview he recently gave on Amanpour and Company at the link below.  

John Urschel | Video | Amanpour: Walter Isaacson sits down with John Urschel, a rising star who walked away from a lucrative football career in the NFL to pursue a doctorate in mathematics at MIT.  

As a Steeler fan and a statistician I must say I'm glad he retired.


I also have an update on Trump's approval ratingsHis ratings have taken a 2%-3% dip in the events surrounding the convictions of Paul Manafort and Michael Cohen and Sen John McCain's funeral.  It's worth noting but still not in Nixon territory for impeachment to happen.  Things have been hectic for me lately but I will begin work on my 8th anniversary post.

**Related Posts**

Will Trump be Impeached? Damned if I know

Monday, September 10, 2018

An Update on Data Journalism and Darply

I've been forced to move back to Pennsylvania so it has been harder to find the time to post to this blog.  The eighth anniversary of this blog is coming up and I will be preparing the anniversary post.  I have found time to write an article on the new data journalism site Darply on hate group concentration in groups per million and Trump's approval rating at the state level.

The full article can be read here.  The website is now crowdfunding to increase it's reach if you care about supporting real news websites please consider supporting them here.

**Related Posts**

Hate Crimes in Pennsylvania Under Reported (and other states as well)

Friday, August 17, 2018

Trump is Popular on the Economy but not Foreign Policy

Much has been written about how Trump's popularity inched up from a low of 37% (according to the Real Clear Politics (RCP) poll average) in December 2017 to 43% in June of 2018.  It has stubbornly remained around 43% ever since as can be seen in the graph above.  His disapproval ratings have fluctuated between 52% and 54% over the same period.

Much less reported is Trump's approval ratings on the economy.  The RCP average of polls on this question is 50.8%, a slight majority.  The polls used to create this average over the last two months can be seen in the image above.  The approval ratings of these polls range from 49% to 55%.  The letter RV next to the sample size for the poll means that they limited their sample to registered voters.  The A next to the sample size means that all Americans were included in the sample.  There is no graph showing how this rating has changed over time but the few times I have looked at this average has been consistent.  His disapproval ratings on the economy average to 42% and range from 36% to 47%.  There aren't as many polls on this question as there are on his overall popularity.  

One poll that is absent from the above table in Rasmussen Reports.  They come out almost daily with overall approval ratings for Trump ranging from 46% to 50%.  They restrict their sample to likely voters (the only ones in the RCP average to do so) and their estimates are consistently the most generous to Trump.

There are even fewer polls asking about Trump's approval on Foreign Policy.  Not surprisingly the RCP average on this is lower than his overall approval rating and his approval rating on the economy at 40.7%.  Rasmussen's polls are not on this question either.  

For the generic congressional race polls the Democrats have a 6.8% lead in the RCP average.  Only Rasmussen limits their sample to likely voters the rest use registered voters on this question.  These numbers have been more volatile than the approval ratings for Trump.  Gerrymandering in many states gives the GOP an advantage in states where the two parties have an equal number of voters.  Pennsylvania just had it's congressional districts redrawn and it remains to be seen what impact it will have.

The Republicans running this year probably will stress the economy while Democrats should be stressing Trump's foreign policy as his approval ratings are weaker there. This doesn't mean that Democrats should ignore domestic/economic issues such as health care, immigration, climate change, and income inequality.  Foreign policy provides a fuller picture.

**Related Posts**

Will Trump be Impeached? Damned if I know

Friday, August 10, 2018

Bishop McCort's Class of '18 Also Has Over 40% NHS Membership

Back in June I posted about how my high school Alma Mater Bishop McCort had over 46% of its senior class in the National Honor Society (NHS) in its 2017 graduating class.  McCort, in it's alumni news letter, just published the list of the 2018 graduating class with NHS membership and the college that they would be attending.  I thought I would take a look the numbers for the 2018 class.

In the graph above shows that the percentage of seniors in the NHS was down slightly at 42.6% though still well above the years 2016 and 1985-1988.  The classes from the 80's (when I was there) had more students.  The years 2017-2018 also had a larger number of students (40) in the NHS than the other years considered.  Eyeballing the colleges that the class of 2018 are attending shows that most are attending local colleges and universities. There were no students going to an Ivy League school.  One student is attending Bucknell University who was not in the NHS.  I will look at that in my next post.


As previously stated the criterion for membership is more stringent now than it was in the 80's with a cumulative 96% grade average needed for membership.  When I was there a 3.3 grade point average was needed.  My Class just had it's 30th reunion.  I couldn't be there but the 35th will be here before I know it.  As far as I know most of them are fine whether they were in the NHS or not.

**Related Posts**

NHS Membership, Not School Year predicts Prestige in College Admission at McCort

Saturday, August 4, 2018

Wuthering Depths

Emily Bronte 1818-1848
I write here when I am inspired by something and when life does not get in the way.  This past Monday was the 200th anniversary of the birth of Emily Bronte, the author of the novel Wuthering Heights.  For this blog I considered the title Wuthering Depths as a tongue in cheek pun on the novel but went with CSI without Dead Bodies as the TV show was still popular when the blog began.  

The details of Bronte's early life can be seen in the image above.  She had a younger sister Charlotte who wrote Jane Eyre and Anne who wrote Agnes Grey.  She had a brother Branwell who struggled with alcoholism and his image was removed from the portrait above with his sisters.
Laurence Olivier as Heathcliff and Merle Oberon as Catherine in the most famous movie adaptation of Wuthering Heights from 1939

Emily was reserved and quiet but she knew the people around her well.  In Wuthering Heights (her only novel written shortly before she died in 1848) the main characters are Heathcliff, an orphan from Liverpool brought to the Moors by the father of Catherine Earnshaw, the other main character.  After the father dies Catherine's brother Linton shuns him and treats him like an indentured servant.  Heathcliff and Catherine have a complicated love affair where she marries a rich man and he becomes cruel and vindictive as a result.  The next generation on the Moors is able to overcome their hatred and be happy and the protagonists are finally reunited in death.

I first read the novel in high school.  It didn't grab me until a few years later as I sympathized with Heathcliff's struggles in a harsh landscape.  Yes he was a product of his environment where the right guidance might have produced a very different outcome.  We all struggle with these issues in life and we all need the right guidance to overcome.  The depths that this blog searches are the patterns that others might miss.  

**Related Posts**

Ambiguous Symbols in 2001:A Space Odyssey

Tuesday, July 24, 2018

Will Trump be Impeached? Damned if I know

Listening to the endless chatter about Trump and Russia on MSNBC I keep thinking about whether there is anything else going on?  The Thai soccer team has been rescued. A mass shooting happened in Toronto.  France won the World Cup for a second time.  Followers of Bernie Sanders have won some stunning victories in the primaries.  Trump is threatening Iran again. Oops that's about Trump again.

It's easy to get caught up in all this hysteria.  I used to write a lot more about Trump and the election when there was a lot more data to sift through.  Of my previous 13 posts this year, this is my second this year about Trump and Russia.  One post was about James Comey's interview on ABC that was only tangentially about Trump.  The rest were about hate groups and recent graduating classes from my high school.  My posts have tried to focus on other issues that are marginally related to Trump if at all.  The graphic above shows how fast Mueller has been cranking out indictments much faster compared to other recent high profile federal investigations. 

The graphic shows that the pace of indictments is coming faster than the others considered, even Watergate.  In this modern world we do not have access to the information that the Muller investigation and the Russians have.  One would assume that Mueller's team would not pursue these indictments without strong evidence to back it up.  Will all this lead to the end of Trump's Presidency?

I don't know about the evidence but I do know the numbers of impeachment.  A simple majority of 218 out of 435 members in congress is needed to impeach in the US House.  After that a two thirds majority of the Senate is needed to remove him from office.  

If the democrats gain control of the house they can impeach him but it will be difficult to gain the 2/3 majority (67 out of 100 members) needed to convict without a significant number of Republicans defecting.  Trump remains popular among Republicans with 88% popularity.  His summit with Putin and other debacles have had no effect on his ratings.  His approval ratings would have to enter Nixon territory of 24% for significan GOP defections to happen.  Watergate whistle blower John Dean recently said that Nixon might have survived if Fox News had existed.

**Related Posts**

How we see the world when we're young and old

Tuesday, July 10, 2018

Managing Predictions with Bayes Theorem

I have been reading Nate Silver's book The Signal and the Noise, about how hard it is to make predictions even when we have complete data. I tend to be a slow reader so I've been reading it for a while.   Finally I got to the chapter on his years as a professional poker player.  He talked about how players like him get overconfident if they start out winning.  Poker is a game of skill and luck.  The great player are lucky and good.  He argues is a thing that can be managed according to Bayes theorem.

In layman's terms, Bayes theorem says that people bring their preconceived notions to a situation and then update their notions based on what happens in that situation.  The image above shows the mathematical formula for the theorem.  If we know the probability of event B occurring given that event A has occurred, If we know the overall (or marginal) probabilities of events A and B occurring separately from each other we can calculate the probability of event A occurring given that B has occurred.  An example of this theorem is shown below.
The table above (that I used in my first post on data driven journalism) shows the 2016 primary wins for Clinton, Trump, and Sanders.  The probability of Trump winning a state given that Clinton has won in the other party's contest in that state is 25 out of 29 or 86%.  We can use Bayes Theorem to find the probability that Clinton wins the state given that Trump has won.

First we need the overall probability of Trump winning the state overall which is 37 out of 51 contests (DC is included) or 73%.  Next we need the probability of Clinton winning a state which is 29 out of 51 contests or 57%.  We can now plug those numbers into Bayes theorem as in the above image.  The percents were converted to decimals for computation sake.  

We update our knowledge of the probability of Trump winning given that Clinton won in the other party with the overall probability of Trump and Clinton winning to find the probability of Clinton winning given that Trump has won.  That number is 67% which is considerable lower than the original 83%.  This is because Trump was more likely to win on his party's side than Clinton was in her party.

We can calculate these probabilities easily after the contests are over but it is different making these predictions beforehand.  Nate Silver was wrong about how many primaries Bernie Sanders would win and who would win the general election after correctly predicting the winner in 2012.  How will he update his predicting model for the 2018 and 2020 elections?  Time will tell.

**Related Posts**

Don’t test me: Using Fisher’s exact test to unearth stories about statistical relationships (Repost)