I told you to keep an eye on these College Humor guys.
Comments, observations and thoughts from two bloggers on applied statistics, higher education and epidemiology. Joseph is an associate professor. Mark is a professional statistician and former math teacher.
Wednesday, December 16, 2015
Tuesday, December 15, 2015
In the Heart of the Sea – – another example of the strange economics of filmmaking technology
As previously mentioned, the digital revolution has increased the productivity of pretty much everybody involved in the filmmaking process but the greatest impact has been in the field of visual effects. The time and money required to get the impressive and impossible on the screen have decreased by orders of magnitude. Which makes it all the more strange that increasingly visual effects are causing long delays and huge cost overruns.
From Deadline (emphasis added):
From Deadline (emphasis added):
In fact, the biggest problem for ITHOTS was its lofty production cost. I understand that the cost originally started at $85M but swelled as the director and his crew contended with the challenges of shooting on the water (always costly), followed by VFX which was the primary reason why the film was delayed from its original March 13 date to December 11.
Monday, December 14, 2015
Distracted by the large flock of black swans
[I'm rushing this out and scheduling an early posting because, as previously suggested, Jonathan Chait is writing a very sharp series of post along these lines and I'm afraid he's is going to beat me to the punch.]
In recent years, a large part of the foundation of the GOP strategy has been the assumption that, if you get base voters angry enough and frightened enough, they will show up to vote (even in off year elections) and they will never vote for the Democrat (even when they really dislike the Republican candidate).
Capitalizing on that assumption has always been something of a balancing act, particularly when you constantly attack the legitimacy of the electoral system ("The system is rigged!" "The last election was stolen!" "Make sure to vote!"). With the advent of the Tea Party movement, it's gotten even more difficult to maintain that balance.
I don't want to get sucked into trying to guess what constitute reasonable probabilities here – – I'm just throwing out scenarios – – but it certainly does seem likely that, if he doesn't get the nomination and does not choose to run as an independent, Trump will still make trouble and things will get ugly.
Keep in mind, Trump's base started out as the birther movement. They came into this primed to see conspiracies against them. Now the RNC has given them what appears to be an actual conspiracy to focus on.
I don't think we can entirely rule out the possibility of Trump calling for a boycott of the vote to protest his treatment but even if it doesn't come to that, it seems probable that, should we see a great deal of bitterness and paranoia after the convention, the result will not help Republican turnout.
What kind of magnitude would we be talking about? It's still too early to say and even if it weren't, I wouldn't feel qualified to speculate, but it would be an interesting conversation to follow among political scientists.
At the very least, the possibility of something big happening down-ballot, though perhaps still not likely, is more likely than it was in the days before Trump.
In recent years, a large part of the foundation of the GOP strategy has been the assumption that, if you get base voters angry enough and frightened enough, they will show up to vote (even in off year elections) and they will never vote for the Democrat (even when they really dislike the Republican candidate).
Capitalizing on that assumption has always been something of a balancing act, particularly when you constantly attack the legitimacy of the electoral system ("The system is rigged!" "The last election was stolen!" "Make sure to vote!"). With the advent of the Tea Party movement, it's gotten even more difficult to maintain that balance.
I don't want to get sucked into trying to guess what constitute reasonable probabilities here – – I'm just throwing out scenarios – – but it certainly does seem likely that, if he doesn't get the nomination and does not choose to run as an independent, Trump will still make trouble and things will get ugly.
Keep in mind, Trump's base started out as the birther movement. They came into this primed to see conspiracies against them. Now the RNC has given them what appears to be an actual conspiracy to focus on.
I don't think we can entirely rule out the possibility of Trump calling for a boycott of the vote to protest his treatment but even if it doesn't come to that, it seems probable that, should we see a great deal of bitterness and paranoia after the convention, the result will not help Republican turnout.
What kind of magnitude would we be talking about? It's still too early to say and even if it weren't, I wouldn't feel qualified to speculate, but it would be an interesting conversation to follow among political scientists.
At the very least, the possibility of something big happening down-ballot, though perhaps still not likely, is more likely than it was in the days before Trump.
Friday, December 11, 2015
Damn you, Jonathan Chait
I've been meaning to write a post walking through the different scenarios for the GOP primary (working title: "I was distracted by the flock of black swans"):
I. Trump gets the nomination
II. Trump doesn't and instead
A. Fades away and/or decides to play nice (that second one just got even less likely)
B. Runs as an independent
C. Causes as much trouble as possible
Unfortunately, while I've been procrastinating, Chait has been writing, producing a string of really sharp posts that are on track to anticipate all of the major points I wanted to make. He already gave us the definitive brief analysis of Scenario I and today he covers most of what I wanted to say about II.B.
Chait is the only news analyst I've seen so far who is seriously thinking through the GOP's dilemma: they can't afford to keep Trump but they can't afford to lose his supporters. The Republican model requires very high turn-out from the base. Pissing off a large segment of those voters could have significant consequences.
Chait actually goes so far as to suggest that the party's best way out of this dilemma might be the much-feared third party run:
I. Trump gets the nomination
II. Trump doesn't and instead
A. Fades away and/or decides to play nice (that second one just got even less likely)
B. Runs as an independent
C. Causes as much trouble as possible
Unfortunately, while I've been procrastinating, Chait has been writing, producing a string of really sharp posts that are on track to anticipate all of the major points I wanted to make. He already gave us the definitive brief analysis of Scenario I and today he covers most of what I wanted to say about II.B.
Chait is the only news analyst I've seen so far who is seriously thinking through the GOP's dilemma: they can't afford to keep Trump but they can't afford to lose his supporters. The Republican model requires very high turn-out from the base. Pissing off a large segment of those voters could have significant consequences.
Chait actually goes so far as to suggest that the party's best way out of this dilemma might be the much-feared third party run:
2. A Trump independent candidacy would have down-ballot benefits for the party. Trump would split apart the Republican vote at the presidential level, but the socially conservative white working-class voters who turn out to vote for him would overwhelmingly pull the lever for Republicans in Congress (and in state elections). The deepest risk Republicans face is the prospect of an electoral wipeout that puts its control of Congress at risk. An independent Trump candidacy would close off such a prospect.
"Confessions of a Republican," Johnson, 1964
I don't want to draw too strong an analogy here -- this is a very different time --but there certainly are parallels. Compare this with the wording of some recent Brooks' columns on the Donald. Of course, the acid test for Brooks will come if Trump or Cruz gets the nomination.
Thursday, December 10, 2015
Brooks on a good day
Critics of David Brooks (myself included) have a tendency to focus solely on the stumbles, those times when the misdirection fails, the shtick grows thin , and the factual misstatements becoome simply too blatant to ignore.
This does a disservice to Brooks, but more importantly, it undercuts the effectiveness of the criticisms. Even if you do not like David Brooks (particularly if you don't like David Brooks) you should, from time to time, check out what he does on a good day.
Last Friday's column provides a good example. The central thesis is, at best, disingenuous – – though Trump remains a longshot, you cannot completely rule out the possibility of a nomination – – but once you get past the basic dishonesty of the title, the rest of the article is about as smooth and well executed as anyone could hope for.
Take the opening analogy:
The rest of the piece follows very much in the same thing. Reasonable, thoughtful and scholarly, making good points and citing the right people, from Nate Silver to Montaigne. No one is better at projecting a calm, professorial tone than Brooks, even when he's whistling past the graveyard.
This does a disservice to Brooks, but more importantly, it undercuts the effectiveness of the criticisms. Even if you do not like David Brooks (particularly if you don't like David Brooks) you should, from time to time, check out what he does on a good day.
Last Friday's column provides a good example. The central thesis is, at best, disingenuous – – though Trump remains a longshot, you cannot completely rule out the possibility of a nomination – – but once you get past the basic dishonesty of the title, the rest of the article is about as smooth and well executed as anyone could hope for.
Take the opening analogy:
A little while ago I went rug shopping. Four rugs were laid out on the floor and among them was one with a pink motif that was dazzlingly beautiful. It was complex and sophisticated. If you had asked me at that moment which rug I wanted, I would have said the pink one.The rug story is simple and accessible, but it does a good job capturing the underlying dynamic. Home furnishings are definitely an area where most of us tend to initially gravitate toward the flashy before having second thoughts and opting for the more tasteful. It is not at all unreasonable to suggest that voting might follow a similar pattern .
This conviction lasted about five minutes. But then my mentality flipped and I started asking some questions. Would the furniture go with this rug? Would this rug clash with the wall hangings? Would I get tired of its electric vibrancy?
Suddenly a subtler and more prosaic blue rug grabbed center stage. The rugs had not changed, but suddenly I wanted the blue rug. The pink rug had done an excellent job of being eye-popping on its own. The blue rug was doing an excellent job of being a rug I could enjoy living with.
The rest of the piece follows very much in the same thing. Reasonable, thoughtful and scholarly, making good points and citing the right people, from Nate Silver to Montaigne. No one is better at projecting a calm, professorial tone than Brooks, even when he's whistling past the graveyard.
Wednesday, December 9, 2015
Whose budget plans are more realistic, Trump or Kasich? (this is not a rhetorical question)*
Nor is it one with an obvious answer, at least not according Jonathan Chait, who takes apart all of the candidate's claims, starting with his habit of taking credit for the balanced budgets of the nineties:
The key element of the Kasich myth is the 1997 Balanced Budget Act, which he credits with producing surpluses in the 1990s. “I balanced the budget in Washington as a chief architect,” he claimed at the last Republican debate, echoing a frequent boast. Kasich’s iteration of his origin story is almost a pure inversion of fiscal reality.
...
Kasich, in other words, opposed the two main laws that created the balanced budget in the 1990s, and supported one that had nothing to do with it. He continues to maintain that he would oppose any tax increase, even a budget deal that combined a 10-1 ratio of spending cuts to tax hikes.
It is worth taking a moment to note that most of the same publications and often the same journalists who labeled Al Gore a self-aggrandizing fabulist for making entirely accurate statements about supporting the early funding of the internet have largely allowed Kasich's fables to go unchallenged.
Building upon his almost entirely imagined record as mastermind of the 1990s budget surplus, Kasich touts what he and his press clippings call his “plan” to balance the budget in eight years. In actuality, it is not a plan at all. Kasich has a bunch of numbers for spending, but he does not say what he would do to arrive at those numbers. For instance, he would freeze all domestic discretionary spending, a wide catch-all category of general federal spending on scientific research, infrastructure, law enforcement, and many other things. This spending has absorbed deep cuts for several years — so deep, in fact, that Republicans in Congress have had trouble funding tolerable savings and compromised on a plan to cancel out some of the additional cutting. Kasich proposes to hold spending on this category constant in nominal dollars, which means that, accounting for population growth and inflation, services will have to be reduced every year. Kasich does not specify how he would allocate those service cuts.
...
Balancing that off is Kasich’s plan to cut taxes. There is not yet an official score for the cost of Kasich’s plan, a fact that by itself nullifies the campaign's claim to have a plan to balance the budget. You can't bring revenues and outlays into line if you have no idea what revenue levels will be. Imagine a business claiming it has a target date for breaking even, and then conceding it has no idea whatsoever what its earnings will be.
By the eyeball test, the scale of the revenue lost by Kasich's tax cuts will be absolutely massive. Kasich would cut the top tax rate to 28 percent from its current 39.6 percent rate. He would cut the capital gains tax rate from 25 percent to 15 percent, cut the estate tax rate from 40 percent to zero, cut the business tax rate from 35 percent to 25 percent, and allow businesses to immediately write off the full cost of all investments — a tax cut for the rich of a scale never before seen in American history. Kasich would also expand the Earned Income Tax Credit for the working poor, which is nice, though it further raises questions as to where he will find the trillions of dollars in savings to pay for all this. Kasich’s campaign tells me that he believes deep tax cuts will encourage faster growth, undeterred by the clear past failure of his beliefs about tax rates and growth.
In sum, it is inaccurate to say Kasich has a plan to balance the budget. It would be accurate to say that he is promising to eliminate the deficit, but he has a plan to dramatically increase it, at least if you define plan to mean the actual change to taxes and spending that he has specified.
As we've frequently noted before, today's journalists love not just narratives, but specifically simplistic, cliched narratives. Once they have latched onto one of these stories, they will go to great lengths to maintain it. This goes beyond selection bias. Many, perhaps most, reporters and editors will actively play down those facts that contradict conventional wisdom and play up or even invent facts that support it (not to name any names).
The press has long since made its casting decisions when it comes to Kasich. Frank, mature, moderate, dull but reliable. Nothing he says or does at this point is going to change the spin. He can lay out a supply-sider vision so extreme it would have both Reagan and Kemp spinning in their graves. He can insist the science is undecided on global warming. He can propose a new federal agency to spread Judeo-Christian values throughout the world as a way to combat the Islamic State. None of this will matter.
That's how typecasting works.
* I wrote this a few days a few days ago before Trump upped the crazy yet again. I might have approached this differently today and probably used a different title, but I still stand by the main points.
Tuesday, December 8, 2015
Note to self:
Start checking Michael Hiltzik's column on a regular basis.
Solving the inversion crisis: How the U.S. can keep companies at home
What George F. Will gets wrong about the progressive income tax
And following up on the last post...
The attack on climate change scientists continues in Washington
The war on data
I've been sitting on this for a while, trying to decide how best to weave it into one of our ongoing threads about the increasingly blatant attacks on important scientific research, about how the conservative movement's experiment in information management has broken down causing the propaganda and conspiracy theories intended for the foot soldiers to propagate among the generals, about the ever clearer absurdity of the establishment press's insistence that both parties are equally to blame.
But there is simply too much here. Just read it.
The House science committee is worse than the Benghazi committee by David Roberts
But there is simply too much here. Just read it.
The House science committee is worse than the Benghazi committee by David Roberts
Monday, December 7, 2015
This American Life had two thirds of a great episode on marketing
Don't get me wrong. The whole episode (569: Put a Bow on It) is very good, but only two of the three segments concern branding and marketing. The first explores the process behind developing the increasingly bizarre junk food hybrids that are coming to dominate the industry.
Zoe Chace
There are a few reasons that guys like these are churning through these food mashups right now. One big one is fast food is losing market share to places like Chipotle, Panera, more upscale, healthier. So a way for fast food to compete is to go in the other direction-- downscale, greasier, sell to their core customers, 18 to 34-year-old guys. Though industry analysts told me it's nearly as many women as men.
And of course, there's money to be made in selling a sandwich that makes people want to take a picture of themselves while they eat it, but only to a point. The question is, will they eat it twice? The Double Down, you know the one where the chicken is the bun, as groundbreaking as it was, it didn't sell that great after people tried it once. Brad, Bruce, Mark, and Eric say it's too expensive to roll out a new product that you'd never order twice.
Newscaster
This is a taco that's the talk of the town.
Zoe Chace
What they want is something that food industry people say Taco Bell did better than anybody in 2012, when they released that taco whose shell was a Dorito.
Newscaster
It's what one marketing consultant calls a marriage made in belly busting heaven. Doritos, the Super Bowl brand that helped turn America into a nation of chunky chip munchers, providing a nacho cheese flavored shell.
Zoe Chace
The Doritos Locos Taco sold and sold and sold and sold-- $375 million in its first year. This is an amazing year for Taco Bell. Every sandwich that arrives on our plates here in Hardee's test kitchen, that is the goal.
From a marketing standpoint, the most interesting part was the way the experts considered both the appeal of the food and the salability of the concept.
Even more fascinating from a marketing perspective is the third segment. TAL called up the best advertising and PR people in the business and asked them what they would propose if Volkswagen had engaged them to rehabilitate its badly damaged brand. There is a Jack in the Box connection with both segments, but surprisingly it's stronger in the third (strong enough that you might want to check out the videos before listening to the episode).
Friday, December 4, 2015
Obscure comic book heroes battle (the idea of) cost disease
Before we get started, one big caveat: the piece I'm linking to seems to be based on rumors and speculation. Back in the Nineteenth Century, newspaper stories often started with the disclaimer "Important if true." For stories like this, we probably start with "interesting if true."
These days, the explanation de jour for cost spirals is...
As mentioned before, technology has improved productivity in film and video by orders of magnitude. These changes have affected every part of the industry but the biggest jump has been in what used to be called special effects. Ray Harryhausen's standard answer to the question "can you do ______" was that he could do anything if given the time and the money, and that was largely true. What has changed is mainly speed and cost.
Here's the weird part, though. The budgets are usually spiraling out of control not because studios are spending that much more on people like writers (whose productivity hasn't grown that much), but because they are pouring money into those areas where productivity has exploded.
Assuming both the rumor and the speculation are sound (and there's lots of other evidence that Hollywood has a growing problem with budget spirals), this raises some perplexing questions.
Fans of the DC Television Universe might have some reason to worry. While the creation of the first season of Legends of Tomorrow is underway, it has come out that the CW may have overextended themselves in regards to budget. It seems that the first season of the Legends of Tomorrow is more expensive than the network originally anticipated. Intel from Bleeding Cool is now claiming that the CW may be nixing the idea of a second season, knowing that show will likely continue to become more expensive as it grows. Yikes.
...
What is it about Legends of Tomorrow that makes it so much more expensive than its predecessor? I can only imagine the multitude of visual and special effects are what is taking such a toll on its budget. Afterall, in a show that revolves around time travel, and where each protagonist has a unique set of superpowers, the effects team must have their hands full. Maybe having one superpowered lead in both Arrow and The Flash allowed for a more budget-friendly production, rather than having to stretch funds across a baker's dozen worth of heroes.
These days, the explanation de jour for cost spirals is...
Baumol's cost disease (also known as the Baumol Effect) is a phenomenon described by William J. Baumol and William G. Bowen in the 1960s. It involves a rise of salaries in jobs that have experienced no increase of labor productivity in response to rising salaries in other jobs which did experience such labor productivity growth. This seemingly goes against the theory in classical economics that wages are closely tied to labor productivity changes.If there's an economist in house, I'd greatly appreciate a knowledgeable take on this, but it would seem we should a disproportionate amount of money going to the people who have had the smallest gains in productivity.
As mentioned before, technology has improved productivity in film and video by orders of magnitude. These changes have affected every part of the industry but the biggest jump has been in what used to be called special effects. Ray Harryhausen's standard answer to the question "can you do ______" was that he could do anything if given the time and the money, and that was largely true. What has changed is mainly speed and cost.
Here's the weird part, though. The budgets are usually spiraling out of control not because studios are spending that much more on people like writers (whose productivity hasn't grown that much), but because they are pouring money into those areas where productivity has exploded.
Thursday, December 3, 2015
XKCD polio
In addition to making an important point about an important world health initiative, Randall Munroe does a beautiful job satirizing the common impulse to abandon the effective for the trendy.
Wednesday, December 2, 2015
Argument by outlier
Perhaps "argument" is going a bit far -- the ideas don't really cohere enough to qualify for that -- but you have to grant the outlier part.
Kevin Carey has just discovered that people who stack up normal debt getting a bachelor's, then take out more loans to go to law school and don't graduate, then take out more loans to go to grad school and don't graduate (and also use student loans for other things like child care), then get a life-threatening autoimmune disease, then get screwed in a divorce and then take a string of low paying jobs and make a number of really bad financial decisions can rack up quite a tab.
We should probably stop here for now. There is nothing instructive in the case of Carey's example, Liz Kelley. The only lesson one can reasonably draw here is that people with bad luck and incredibly bad judgement tend to fare poorly in life.
Joseph and I are planning to come back to the rest of Carey's hopelessly muddled reasoning later. That's not to say that everything he says is wrong (life is not a knight/knave puzzle). The system is horrible, just not necessarily for the reasons he gives. Subsidizing higher education through student loans creates all sorts of perverse incentives, it sends too much money to undeserving institutions and it allows a few people like Kelley to spend far too much taxpayer money. It isn't difficult to argue against this system, but if the NYT Upshot wants to maintain its reputation, the paper should probably find someone other than Carey to make the case.
Kevin Carey has just discovered that people who stack up normal debt getting a bachelor's, then take out more loans to go to law school and don't graduate, then take out more loans to go to grad school and don't graduate (and also use student loans for other things like child care), then get a life-threatening autoimmune disease, then get screwed in a divorce and then take a string of low paying jobs and make a number of really bad financial decisions can rack up quite a tab.
We should probably stop here for now. There is nothing instructive in the case of Carey's example, Liz Kelley. The only lesson one can reasonably draw here is that people with bad luck and incredibly bad judgement tend to fare poorly in life.
Joseph and I are planning to come back to the rest of Carey's hopelessly muddled reasoning later. That's not to say that everything he says is wrong (life is not a knight/knave puzzle). The system is horrible, just not necessarily for the reasons he gives. Subsidizing higher education through student loans creates all sorts of perverse incentives, it sends too much money to undeserving institutions and it allows a few people like Kelley to spend far too much taxpayer money. It isn't difficult to argue against this system, but if the NYT Upshot wants to maintain its reputation, the paper should probably find someone other than Carey to make the case.
Tuesday, December 1, 2015
Kruzno -- the beta
Back when I was teaching secondary mathematics, I made extensive use of games and puzzles. In addition to building a pretty good collection, I also developed a few of my own. Of the board games for the, the one that played the best and got the most positive feedback was a hexagonal checkers variant called Kruzno.
A while back, I decided to put together a beta version of the game and see what kind of response it got. I commissioned a well qualified and quite reasonable professional to do the art. I had a local company produce a small run of boards and boxes, but I decided to forgo custom pieces and instead stick with chessmen which were available off-the-shelf (it is far easier to do a small inexpensive batch of paper products then plastic products).
The project has since been pushed to the back burner, at least until I have the time to put together a mobile version of the game. In the meantime, I decided to set up a fulfillment-by-Amazon store (which is a surprisingly low maintenance option if you want to try some small scale retailing).
I hate to oversell a beta, but I am quite proud of the game and besides, every house should have a hex board. I will be going up the abstract strategy thread again both here and at the teaching site, so if any of the upcoming discussions make you curious, drop by the Amazon page and take a look.
I haven't posted this in a while...
... which is a shame because there's a useful lesson here: even the guy with an axe draws the line somewhere.
Subscribe to:
Posts (Atom)