It's Alive!!

My beloved Triage appears to be alive and well, but in stealth mode. I've not actually had the pleasure of meeting it. But the Times yesterday kind of let the cat out of the bag. If this isn't a description of Triage driven campaign building, I'd be hard pressed to do better. What's galling is that the Democrats happily ignored the issues in 2010, and have set us on the road to permanent minority rule. Just like a South American junta.

Issenberg, a he by the way, is publishing a book, "The Victory Lab: The Secret Science of Winning Campaigns". Alas, I wasn't a source. No royalties for me.

Over the last decade, almost entirely out of view, campaigns have modernized their techniques in such a way that nearly every member of the political press now lacks the specialized expertise to interpret what's going on. Campaign professionals have developed a new conceptual framework for understanding what moves votes. It's as if restaurant critics remained oblivious to a generation's worth of new chefs' tools and techniques and persisted in describing every dish that came out of the kitchen as either "grilled" or "broiled."

My first serious post college position was in Washington, DC (for the Civil Service Commission, which no longer exists; bet you didn't know that), in a group titled Office of Analytic Methods. I was the economist/econometrician, while one of the other worker bees was AbD in psychometrics. For reasons not yet discussed, I've long viewed any study prefixed psych- with suspicion; a means for the venal to manipulate the naive'. Advertising, "Mad Men" style, is the archetype. Eventually, we got "The Selling of the President 1968". Some 40 years on, and the number crunching has gotten evermore convoluted, possibly more sophisticated.

Oh, did I mention that I wandered into my local Barnes & Noble to see what's newish in the data/stat world? Yes, yes I did. And what to my wondering eyes did appear but "R For Dummies". I suppose that R will become the next Excel: any knucklehead will feel empowered to play math stat in the office, just as Excel empowered cube monkeys to self-identify as financial analysts. And we now know what that produced.

Campaigns have borrowed techniques from the social sciences, including behavioral psychology and statistical modeling. They have access to private collections of data and from their analysis of it have been able to reach empirical, if tentative, conclusions about what works and what doesn't.

And to quote my humble self, from Triage:

There is, available to the apparatchiks, both public data (the FEC here in the States) and data developed by their own organization. This latter data is, amorphously, expenditures (the source data that ends up at the FEC; their own they have, but opposition data must wait for FEC and may well not be sufficiently timely) and outcomes; perhaps simple polling results; perhaps some focus group results; perhaps some name recognition surveys. Social network data mining is also big these days (although I've not done enough research to know for sure that this could be a data source for outcomes).

Issenberg throws in the towel:

Breathless, and often fact-free, stories about "data mining" and "microtargeting" soon became plentiful. But few journalists had access to any of the campaigns' data, or even much understanding of the statistical techniques they used. We found ourselves at the mercy of self-promoting consultants who described how they were changing politics by ignoring stodgy old demographics and instead pinpointing voters according to their lifestyles. We played along, guilelessly imputing new mythic powers to microtargeting. In many retellings, data analysis became the reason George W. Bush was re-elected.

There has been, in the wake of Ryan's perversion, hand wringing from some of the press that fact-checking (which effort draws the ire of the Right Wingnuts, not surprisingly) in the face of such lying will take too much effort to police. The message is that Right Wingnuts will send a tsunami of falsehood, much never exposed as forcefully as the assaults. "Swift Boats" 24/7. Welcome to the new Gulag.

Indeed, the telling numbers wouldn't be polls but the individual probability scores that Mr. Obama's targeters developed (and update weekly) to predict how likely each voter in the country is to support him.

As Triage described, high granularity data, external to the campaigns can be used. One of the not so secret secrets in the quant world is that private databases exist, for a fee, to very fine detail. As you bend the mind, so you bend the finger on the voting lever.

But particularly in a polarized race like this one, where fewer than one-tenth of voters are moving between candidates, the most advanced thinking inside a campaign is just as likely to focus on fine-tuning statistical models to refine vote counts and improve techniques for efficiently identifying and mobilizing existing supporters.

So, we find:

...Mr. Romney deployed statistical models to track Iowa supporters and current vote counts for his rivals. It amounted to a largely invisible 21st-century upgrade to the traditional infrastructure of offices, phone banks and staff that most journalists visualized when they tossed around the term "organization."

Back in the 1980s I applied to, and was accepted into, the American University (the one in Washington, DC) Economic Journalism graduate program. For various reasons, I didn't get to go (you know who you are). I still have, more so recently with the advent of R in particular, much regret that I wasn't able to watch and participate in this evolution.

Are You Crackin' Wise with Me, Palooka?

Among quant types, there's been an on-going battle twixt the frequentists (among whom, Your 'Umble Servent) and the Bayesians. To the unfamiliar: frequentists have all faith and credit in the observed data while Bayesians accept that investigators must have some prior knowledge of the situation, and this knowledge ought not be wasted, but utilized in the analysis.

In the words of the Wiki the point of Bayesian: "...the posterior probability of a random event or an uncertain proposition is the conditional probability that is assigned after the relevant evidence is taken into account."

That "relevant evidence", typically called the prior (again, the Wiki): "A prior is often the purely subjective assessment of an experienced expert."

Since the prior is fuzzy, shall we say, frequentists often use a somewhat derogatory phrase for the process of finding one. So, imagine the chuckle induced when I ran across this turn of phrase from a recent posting: "... a different sampler for sampling from their posteriors." Sometimes, an editor is worth the few ducats they receive.

Too Much of a Good Thing

There's the great tag line from Scotty on "Star Trek", "I can nah make it go faster, Cap'n". Turns out, he never said exactly that. But, as mentioned in recent musing, the IT/computer world is facing a surfeit of power and a shortage of useful purpose for that power. As the saying goes, It's the Distribution, Stupid.

Back in the Goode Olde Days, Intel and MicroSoft had the symbiotic good fortune to satisfy each other's need for more of what the other had to offer. To the extent that Office owns the Fortune X00 offices, the PC turnover will continue for a while. Whether MicroSoft can find a way to chew up more (parallel) cycles remains to be seen.

But what is now apparent is that the mobile phone world, surprise!!, has entered that Twilight Zone. Here's a new review of an LG phone, the Optimus 4X HD.

Honestly, Tegra 3 hasn't done anything for me that OMAP4 and Exynos 4210 weren't already able to do just fine. So while it's awesome that quad-cores have come to phones, I'm not certain that it'll change your smartphone usage patterns significantly unless you have a specific need for a ton of compute horsepower.

If you read the whole review, there's creepy crawlies afoot for both hardware vendors (parts is parts), software (Google and MicroSoft), and phone assemblers (Apple is just that, too).

The only way the O4X HD significantly changed my usage patterns, actually, were related to battery life. Or, to be more accurate, the lack of it. It's a phone that's pretty brutal on battery, between the quad-core and the 4.7" IPS display, so I wasn't expecting anything great to begin with. But connected to 3G, I was averaging roughly 24 hours of *standby* time. That means screen off, sync off, everything off - just with the phone sitting there doing nothing.

Having just gotten an LG Lucid (one of the freebies on the Verizon upgrade), yeah LG phones seem to be battery shy. And for someone who just wants to make a few damn phone calls...

It makes one wonder how long planned obsolescence can succeed? Who, with what, will be the next Great Cycle Sink for mobile phones? I recall a scene from a later "X-Files" episode where Mulder is seen walking, wearing his trenchcoat. You hear a phone ring, and Mulder pulls one of these up to his ear.

(Dr. Martin Cooper of Motorola made the first private handheld mobile phone call on a larger prototype model in 1973. This is a reenactment in 2007, Wiki)

As the current notion of what the purpose of a mobile entertainment device morphs, how soon will we be lugging things that big around? Hell, an iPad has nearly the same bulk.

The original Razr, while not always bullet proof, is still the best design for making mobile phone calls.

Teddy Roosevelt's Revenge

Now that the planet has read up as much as it wants about the Apple-Samsung case, here comes another drop in the ocean.

Does anyone remember the iPhone, as it was called in 2007? I do. It was the dumbest phone I'd ever seen. Bar none. The current phone design at the time was the flip phone, narrow compact, yet able to span the ear/mouth distance quite nicely when open and in use. This iPhone drek looked like the then current iPod chassis. Here's the page for the iPhone, here's the page for the then current iPod. Notice the similarity? So, what's so innovative about the iPhone? The hardware, but that's not developed by Apple, but by the hardware companies.

From the Wiki:

Most touchscreen patents were filed during the 1970s and 1980s and have expired. Touchscreen component manufacturing and product design are no longer encumbered by royalties or legalities with regard to patents and the use of touchscreen-enabled displays is widespread.

The equally silly BlackBerry of 2007 can be seen here.

The dimensions of the BlackBerry: 107 × 51 × 15 mm
The dimensions of the iPhone: 115 x 61 x 11.6 mm

Should RIM have sued? You betcha. So far as I can find, they never did. Too bad. This case makes it clear that the US patent process is so broken, one could patent a ham sandwich. It's a farce. On balance, the USPTO gets money for granting patents; it's biased towards granting an application for a ham sandwich. The details are convoluted, and AIA makes them moreso (IMHO), but the overall effect is that applicants are buying approval. Much of the (left wing?) criticism of the patent regime is the bias to approve; the rest lies with the law which empowers silly patentability. Not everyone agrees that "Apple has a patent on a rectangle with round corners"; some split hairs and some wave their hands, but the result is that Apple has such authority. Does it make sense to grant such authority?

Not exactly the Square Deal that the Rough Rider had in mind. Here's a picture of what's happened to copyright authority over the years (from the Wiki entry).

Note that the 1909 (beige) version from Roosevelt is about 50% of what it is today, effectively infinite.

Body Snatchers

You have to watch this. So creepy it's cool.

Money Makes the World Go Round

It's been a while since I've vented about the silliness of the DNC in turning down the Triage application; in particular the mapping part. What to my wondering eyes did appear, but a US map half done in today's Times. It's half done, since it displays the input, $$$, but doesn't show the output, it seems. Again, in Triage terms, output would be some measure of effectiveness. Can be done, as Triage demonstrated.

To be fair, it may not even be the Times who did the work. The source cite is "Campaign Media Analysis Group at Kantar Media", but doesn't say whether Kantar provided only the raw numbers, or both the numbers and the map. Since the Times is well known to use R for its data graphics, my guess is the former. The states are graphed with their blue/red leaning (without indicating whether the lean was measured correlatively with the spends), so a time series of the map would be closer to the real time thrust of Triage (googleVis supports such graph animation, sort of; R-bloggers has a number of posts on point). Even so, bravo.

Mine Is Bigger Than Yours

Likely the most significant (not so secret) secret in quants is that Size Matters. Big Men on Campus think so, too. Both have about the same result: the innocents get screwed (or substitute for the euphemism).

Alzheimer's turns out to be intractable so far, and this has had a depressing effect on drug trials. Nothing has worked, so far. Well, may be something does work.

Today's reports about Lilly's failure, not the first in recent time, reveals that Lilly is trying the sample size bigger ploy. The snippets are from the NYT.

Lilly said, however, that when the results of the two trials were combined, creating a larger sample size, there was a statistically significant slowing of the decline in cognition.

Being somewhat addicted to cop shows since adolescence, one of the standard lines from same: "A DA can get a ham sandwich indicted, if he wants to." In the world of quants, the analogue goes, "A quant can find a significant difference between ham sandwiches with a large enough sample size." In drug development, clinical trials can be time consuming and expensive, so individual trials tend to be powered to the smallest sample size indicated by previous data, and the assumed magnitude of difference between the drug and either placebo or some standard of care. What Lilly tried was to pool data from separate trials.

Pooled data is acceptable to math stats, but the requirements are pretty strict, principally with regard to variance within and between trials. Here's an historical criticism.

Lilly also went the way of post-hoc sub-group analysis, another not universally accepted gambit.

What this means for the drug's future is still unclear. The effect of a drug on a subgroup of patients in a trial is typically not sufficient grounds for a drug to be approved without further clinical trials involving just that subgroup.

And finally, the coup de grace,

Also left unclear Friday was how big the effect on cognition was -- whether it would be meaningful for patients or merely meet some statistical test.

A biostat can make the slightest difference look colossal, given a large enough sample size. When does a ham sandwich look like a BLT? Measure enough of them, and you'll be able to prove it.