In the TV series MASH, there was an episode featuring a statistician – an Army officer who predicted how many men would be killed or wounded given the parameters of an upcoming battle. To the statistician character, it was all about numbers – “just business, nothing personal”, to invoke a line from a different seventies production. To surgeon Hawkeye Pierce, the character who had to try to patch together the actual men behind the numbers, is was in fact personal.
At the end of the episode, losing his temper at the statistician, after showing the geek through the operating room, Pierce yells “the thing I hate about you isn’t that you’re good at your job. I hate you for liking it so much”.
I have a similar reaction to people who try to boil all human behavior down into numbers, statistics and analytical models.
If blogs existed 50-60 years ago, a story like this would be accompanied by a photo like this. Good thing this is 2016, right?
Now, before you launch into some misguided jape about conservatives hating science, remember – part of my day job is, well, boiling down human behavior into numbers, stats and patterns. A bigger part, at least for me, is finding the qualitative answer behind the numbers.
But I digress. Among the many joys of this past election – the potential for a safe SCOTUS, a solid cabinet, no Hillary, no leasing of US foreign policy to the Saudis and Qataris – was the complete collapse of analytics in predicting (and, via our media, shaping) this past election.
The analytical models for both sides pointed to a Clinton victory, albeit not a runaway. The Clinton campaign and super PACs had several of the most highly regarded polling firms in the Democratic Party, yet in the places that ended up mattering, very little if any polling was done. So while 2016 wasn’t a victory for traditional polling, it certainly took a lot of the luster from analytics. In the end, big data mattered very little.
While tinkering with stats can be fun, I’ve long loathed notion that all of human behavior can be boiled down into numbers. And I’ll admit, the schadenfreud when the geeks fail to do so is glorious.
They didn’t figure in that The Donald may be a very smart marketing man. Think Wisconsin. As discussed before…..the week before the election, the Democrats sent Chelsea to Eau Claire, where she underwhelmed about 200 upper-middle class white Prius-driving Democrats at a sedated event in a hotel conference room. That was about it for the Clinton campaign that week in Wisconsin.
Then Trump held a loud energetic rally before 3,000+ (full capacity) at UW-Eau Claire arena (and other places also). Even the anti-Republican Eau Claire Leader-Telegram gave it fair coverage.
I wonder if there were intangible items also…like several hundred Democrats outside the Trump rally screaming nasty obscenities and threats to the families lining up to enter the Trump event. I know non-political blue-collar types who went to that rally. The protesters may have made them solid Republicans.
“Every age has its own outlook. It is specially good at seeing certain truths and specially liable to make certain mistakes.”
-C.S. Lewis
One of the “certain mistakes” frequently made in our own age is to believe that people are machines. Complicated machines, but machines nonetheless. It has never been proven that human thought, aka language, is a product of the natural world or that human thought is capable of representing the natural universe accurately.
The reason the analytics failed is that they are all using garbage data. The reason it is garbage is that Americans en masse have decided to stop cooperating with the data takers.
If you could understand that phenomenon you could understand a lot about what’s going on in the country. But how are ya gonna collect any data on the phenomenon?
If I don’t recognize the name or number on the caller ID on the home phone, it goes to voicemail. If the caller doesn’t leave a message, I know I didn’t waste my time.
Predictions based on models are as good as the assumptions built into them. What economic model predicted the 2008 recession? Or the preceding housing bubble? In retrospect it’s easy to see what happened. One of the things that infuriates me about climate”science” is that its proponents claim to be able to accurately forecast earth ‘s climate 50 years from now based on the concentration of a single trace gas (CO2 is 0.035 % of our atmosphere and is a much weaker greenhouse gas than water vapor, which exists in much higher amounts). It’s scientific hubris at its worst. Science is better when explaining the past but even then it’s far from perfect. When I was a kid the conventional wisdom of why dinosaurs were extinct was pure Darwinian selection. The tiny mammals ate dinosaur eggs and wiped them out. Only later was the mass extinction blamed on collision with an asteroid. Oops. And in my field of medicine, there were many dogmatic beliefs upended by skeptical researchers. One example: I remember listening to the chief of cardiology at a prominent Twin Cities hospital announce in no uncertain terms the first thing to give a patient with chest pain was intravenous lidocaine. A few years later further research disproved that and showed that outcomes were worse when lidocaine was given. So show me a person who says science about anything is settled and I’ll show you a complete idiot.
In 2008 a lot of us in the Center, Center-Right and Right asked: Beyond Community Organize, what has Obama ever done? The answer, often, was that he ran a successful campaign for the nomination and later for the Presidency. Snicker we did, but based on the 2016 results, there is something for having an instinct in how to manage yourself and your campaign for maximum advantage.
When I read after the election that the Hillary spent more in Omaha than in Michigan, Pennsylvania and Wisconsin combined and that the reasoning was that they didn’t want the Trumps to get the idea that those states were in play, what could one do other than laugh at her hubris? Then to read that her ground forces – the supposedly “uncoordinated” union and other aligned advocacy groups on the ground who wanted to go to Milwaukee or Detroit to GOTV for Hillary and were ORDERED to stay put because Hillary was doing some kind of Jedi mind trick so Trump would realize he couldn’t win, you recognize that she either doesn’t listen to bad news or doesn’t (as our genial host might put it) ‘pack the gear’ to respond to a negative situation. And that her GOTV effort in itself was such a muddle that she wound up getting out more Trump voters. In the end Obama proved right – running a successful campaign is kind of a big fu***** deal (H/T Joe Biden). Shame that KellyAnne Conway doesn’t get more credit for the glass ceiling she shattered managing one.
Seflores,
If Kellyanne had been a DemonRAT pulling off the same feat, she would be nominated for Queen of the world.
Nothing wrong with the mathematics, but GiGo always applies.
Mitch, I don’t think anyone this side of Asimov’s Trilogy has ever argued that you can reduce it all to numbers. The comments above about the quality of the data today apply to survey data, but many of us have access to data from actual observed behavior (even when the person doesn’t know they are observed.) And the behaviors do show patterns from which you make inferences. Inferences occasionally can become predictions, and predictions can come true, but ’tis many a slip twixt the cup and the lip; even the best forecasters get them wrong.
Not all of us missed the buildup of debt in the mid-2000s, but without any historical precedent to a derivative of derivative securities (a “derivative-squared” in the industry, seriously) we had no way to forecast how the impact would play out because the firms themselves did not know their own exposure.
You had three possible forecasts: it won’t be a problem; it will be a problem but manageable; we don’t know what will happen because we’ve never seen this before. If you’re a commercial forecaster, just know it’s hard to get paid for the third option.
moneyball is fine when confined to baseball/sports. Human behavior is way too unpredictable which is why despite being closest, Nate Silver still said Hillary had a better than 7 in 10 chance of becoming madame president
king you really need to drop by and comment more
The excellent movie “The Big Short” told the story of people who saw the crash coming, tried desperately to warn the regulators, and eventually made a fortune off of the debacle.
If the people who lost everything in 2008 by going long were acting rationally, and the people who made a mint by going short in 2008 were acting rationally, where does that leave us?
Life is a long lesson in humility.
J.M. Barrie
Pingback: In The Mailbox: 12.29.16 : The Other McCain
The rational people said that if the problem is as bad as you say it is why hasn’t it crashed already? Same argument people are making about the national debt.
People have been saying the debt his going to destroy us for 40 years. And it hasn’t yet. So the argument must be wrong.
What do you think King?
Suppose the US government, somehow, had a 20 trillion $ surplus — money in the bank — instead of 20 trillion $ deficit. What would that mean? It’s just markers, isn’t it? It wouldn’t mean that there was 20 trillion $ worth of goods and services out there, sitting around, idle and unused, until the US government loosened the purse strings and spent the money.
With so much money to lend, wouldn’t the result be interest rates down to, say, 1%?.
It’s interesting that Trump’s nominee for OMB, Congressman Mulvaney believes that the national debt is the biggest problem facing the country. First, balancing budget and paying down debt will be highly deflationary. Second, the stock market is near 20k because the debt is at 20tr. Neither Trump nor the people around him don’t seem to know much about Alexander Hamilton, who created vast financial wealth out of worthless Revolutionary War IOU’s by pointing out to the population that every debt is also an asset, especially if it’s tied to a guaranteed revenue stream called… federal taxes. It’s very dangerous to believe that taxes are antagonistic to financial wealth, instead of simply being the other side of the coin.
“Life is a long lesson in humility.”
So when are you going to get the message, Emery Incognito?
I prefer to lose an argument and learn something than win an argument and finish with the same ideas I had when I started.
Well life is just one big elementary classroom for you, isn’t it SSOLSEmery? You have that going for you, which is good, but maybe you should consider applying what you learn; the view of the world is better when your nose isn’t perpetually planted in the corner.
Re-post because it seems appropriate!
There, see how easy it is to get this peckerwood to reply to me over and over and over again? Can’t… stop… laughing
Say Swiftee, it’s time to feed your sparkle unicorn again.
https://memegenerator.net/instance/11580288
SSOLSEmery always includes proper attribution to his own amphigory…LMAO
Scott Adams addresses Glerbal Werming in a predictably sensible fashion. He acknowledges both sides have good points, and lets the rabble argue it out among themselves. Hilarity ensues….
http://blog.dilbert.com/post/155142928411/best-arguments-for-and-against-climate-model
SSOLSEmery dreaming of me feeding my sparkle pony…
http://tinypic.com/r/2qi7c40/9
Does your wife practice yoga?
http://cdn.randomfunnypicture.com/wp2/wp-content/uploads/2010/01/yoga-instructor-grope1.jpg
wow shits getting ugly
*grabs popcorn and waits for alt-right swiftees response*
Fine, I am overdue for my “1 in 20” serious comment. I only make serious comments maybe 5% of the time. This will be one of them. Attention Democrats: everything that happens from here on out is the result of you decision to nominate Hillary Clinton and nothing else. OK back to my regular character.
That photo looks like your sons reporting to prison to serve their time for molesting little kids, SSOLSEmery.
“This will be one of them. Attention Democrats: everything that happens from here on out is the result of you decision to nominate Hillary Clinton and nothing else.”
But they had no choice, Emery. Sanders would have been a disastrous candidate, a second George McGovern. Sanders’ economic ideas, rarely examined closely, were kooky, “Port Huron Statement” New Left crap that no one, other than Bernie Sanders, believes in any more. None of the candidates other than Sanders managed to win a single primary. O’Malley retired from the field after winning < 1% at the Iowa caucuses. Biden, I think, might have won the EC against Trump, but I have heard that Biden in his past presidential runs has been a horrible candidate. He made unforced errors, spent money in the wrong place, etc. In any case, the Clintons sucked the air out of the room.
The GOP bench was deeper than the Democrat bench, but none of the GOP candidate, not even "outsider" Cruz, were willing to challenge the GOP donor class on the two issues where Trump was willing to challenge the GOP donor class: immigration and free trade.
The GOP donor class, BTW, tends to be billionaire businessmen and business organizations, not social conservatives.
It made me laugh, which is all that matters
All-right JC: On the bright side, Trump and his cabinet are not owned by anybody. It is hard to overstate how unusual that is these days. It is very much a positive thing. On the other hand, nobody knows what Trump actually believes or will actually try to do. Biggest concerns are (a) conflicts of interest related to his personal wealth and (b) use of executive power to retaliate against personal enemies. I view the latter as the most serious because it is consistent with his personality and history as I understand them. I highly suspect the Council of Economic Advisors is going to be irrelevant to economic policy. Trump will be listening to Ross and Mnuchin, not Kudlow. The Kudlow thing is just to make people scream. None of this is going to be clear for another year or two. Which is part of what makes all the howling so funny.
First, balancing budget and paying down debt will be highly deflationary. Second, the stock market is near 20k because the debt is at 20tr.
Some people, after making even one of these two inobvious assertions, would make at least a trivial effort, a hand-wave at backing them up. But no, having been made, these unsubstantiated assertions are then used to confidently draw conclusions. Perhaps these too are questionable, but that sort of depends on the the predicates.
Trump’s annual deficits will be 2 to 3 times larger than $600B. Trump’s own plan states this. Do you think interest rates are increasing because the markets expect less government spending?
One thing I’ve noted in many, not all, prognostications is that a lot of them are really “searching for a hypothesis”–tweaking the data until you get a statistically significant result. Add that to the phenomenon of “not realizing your measurement influences the data”, and it’s a recipe for disaster.