communication. community. cognition.
Posts tagged Math
Game the System
Dec 3rd
(How Ike became the #1 Twitter Elite of Planet Earth)
We just came out of an election season where there were multiple ways to keep score:
- Popular vote
- Electoral vote
- States carried
In the above instance, we have a constitutional mandate that tells us which one matters. Life is often more fuzzy than that. Take the college football standings in the Big-12 South, where three teams finished with seven wins and a single loss.
- Texas beat Oklahoma 45-35
- Texas Tech beat Texas 39-33
- Oklahoma beat Texas Tech 65-21
If you spin them head-to-head, you just keep rolling in a circle. The Texas fans think their win ought to count for “more,” because it happened on a neutral field. The Oklahoma fans say Poppycock (or something equally rustic, and likely demeaning to both cowboys and cows), that they lost to Texas within the state of Texas. Also, Oklahoma appears to win in both overall strength-of-schedule and in point differential.
The Texas Tech coach, realizing his team was too far out of the conversation to matter, said his team’s higher graduation rate ought to break the tie.
Instead, the division championship came down to which team was highest ranked by an arcane formula involving two separate polling efforts, six computers algorithms, a field mouse and SCUBA gear. No one knows, we just trust the result. Or complain about it.
Laughter, the best disinfectant
Hidden processes are at the heart of how one can claim victory in something few understand. Which is exactly what Matt Bacak did earlier this week.
Mr. Bacak – in a fit of self-promotion, powerfully promoted his wares as a marketing guru with a press release touting his score on a service called Twitter Grader. He claimed to have risen to the rank of #3 in the Atlanta Twitter Elite. And every bit of it was true.
What is also true is that on the very same day, I truthfully proclaimed myself the #1 Twitter Elite of Planet Earth.
Unlike Mr. Bacak, I won’t sell you any expensive secrets. Here’s how it’s done.
- Find a service that offers some type of rankings.
- Figure out what impresses that service.
- Emulate, taking as many shortcuts as you can.
Holes in the System
Here are the secrets Matt won’t tell you.
First, Twitter Grader rewards people who appear to have more “followers” than they actually “follow”. There are many users of the service who pride themselves on maintaining as even a ratio as possible. So Matt added as many as he could – waited for the reciprocal follow-back – then did a mass dump of his “following” list. As of yesterday, he was being followed by over 1900 people, but he was only following 32. The ratio makes you appear more influential than you are.
Second, the location definition for “Twitter Elite in ________” is arbitrary and has no heirarchy. If you put “Muncie, IN” as your location, you will only be compared to those with a match. Switch to just “Muncie” and you’re in a different pool. It’s entirely possible that you could be far-and-away the TwitterKing of Muncie, yet not show up in a search for the state of Indiana. I chose “Planet Earth” for my location. I could have claimed the whole galaxy. It doesn’t matter.
Other than changing my location to “Planet Earth” for a day, I did nothing to game the system.
Tempest in a Twitpot
This is worthy of more than a laugh. There are a couple of object lessons here.
I can’t feel very sorry for Matt. Either he knew precisely what he was doing and ought to be ashamed for promoting it, or he has no clue and ought to be prevented from drawing consulting fees from businesses that don’t know any better.
Additionally, this is a call to feed your inner skeptic.
- When someone makes a claim, ask for the proof.
- When you see the proof, ask for the innards of the algorithm.
- When you don’t understand the algorithm, ask for an explanation.
- When you don’t understand the explanation, run.
Does this mean I need to quit relying on the almighty Google? That depends. Am I satisfied with the results? Mostly. Is it costing me anything? No. But at least I’m putting the algorithm to the test.
And you should, too.
[tags]Ike Pigott, Occam’s RazR, algorithm, search, marketing, BCS[/tags]
Look For the Twist
Mar 27th
Steve Harden was one of my best friends in high school and college. We were in many of the same science and math classes, although he had a very strong artistic streak. When it came time to get serious about actually graduating, Steve declared a double-major in Chemistry and Art.
One of my other friends (who had a penchant for asking rather snotty questions) posed the following: “What are you going to do with that? Draw illustrations for science textbooks?”
Steve answered “No. I may go the other direction, and get involved with art restoration.”
I’ve always thought the most interesting niches develop at the intersections of different disciplines. That’s where the concepts of one dovetail with the uncertain problems of the other and reveal a new way to solve. Fermat’s Last Theorem tied mathematicians up in knots for centuries, until a topologist translated the problem into his field and attacked it in a new way.
While the intersections can provide insight, you must be careful not to jump to hasty conclusions. That intersection that you view from directly above might just be an overpass. In Steve’s case, it was the chemist’s knowledge that enriched the art, rather 
than the artist’s touch helping the chemist communicate.
Are you making unnecessary assumptions about which road’s on top?
[tags]Ike Pigott, Occam’s RazR, science, communications, philosophy, art, Fermat, mathematics, topology[/tags]
Five Reasons Why Financial News Is Bogus
Feb 21st
“The Dow dipped sharply today amid rumors that Britney Spears would get out of rehab a week early.”
Well, why not? It’s about as correct as anything else the financial pundits spew.
One of the functions of feeding a 24/7 news cycle is creating content where no verifiable news exists. Analysis has its place, but that piece of real estate should be very small if the news editors had any sense of shame. Stocks go up, stocks go down. Trading is either heavier or lighter than expected. But the vast majority of those measurable effects are the product of chaos, non-linear dynamics, and random flux. The way these analysts share their 20/20 hindsight is the height of windbaggery, but it is also dangerous in many ways:
-
It divorces causes from effects
There are very real causes for some of the major trades and influences on a stock market. Unfortunately, many of them exist beneath our observations. External factors that are important but counter-intuitive get no attention.
-
It celebrates causation
It’s easy to come up with some big news of the day and tie it to an economic outcome. Maybe it’s investors who are jittery about a natural disaster, or news of a political scandal. Anything in the headlines can contribute to the “watercooler effect,”
even if the Wall Street stuffed shirts only drink bottled water. These faux explanations have the same predictive value as the winner of the Super Bowl has on the markets. -
It promotes fear in the marketplace
As a reporter, I hated those days when the Dow Industrials took a sudden plunge. It meant a newsroom scramble to tell people what to do with their 401-K plans. The people assigned to the story could neither name a single stock on the Dow Industrials, nor how many stocks there were in the index, nor that the “K” stands for Keough. It creates a shark mentality, that if you aren’t actively buying and selling and swimming forward you’ll somehow drown.
-
It promotes a false sense of assurance about our ability to know
The manner of these presentations is always sure and omniscient. No analyst has ever turned to the camera and said “Beats the hell out of me, Jack.” Since the numbers are mostly arcane, unknowable, or unpredictable, it’s nice to have a tour guide who can make sense of them. It lulls the public to sleep, thinking there are in fact geniuses who can make sense of these things. Things like climate change, for instance. -
It is non-predictive and useless
We’ve got much better data on the markets than we do planetary climate. If we’re so damned sure the computer models are on the money (assuming the right observations), why hasn’t the same fuzzy-logic been used to make some egghead dizzyingly wealthy? Because the models stink. They smugly pronounce the whys and wherefores as if they knew it all along that morning. And they did. Pick some significant political event or speech, circle it, and then attribute it at the end of the day to the ______ in the index. (Insert either “rise” or “fall” in the blank.) This is how they write the story before they know the final score. It’s just a matter of reverse engineering.
Intelligence is a measure of your ability to acquire and use knowledge. Wisdom is a measure of your willingness to accept your limits. The financial airwaves are filled with very smart fools — empty heads in tailored suits selling fairy tales to fill the time between commercials.
[tags]Ike Pigott, Occam’s RazR, finance, news, markets, broadcasting, Super Bowl, statistics, chaos, nonlinear dynamics[/tags]
Fractions
Nov 9th
As of the time of this post, I’ve been out of television for 1390 days. Haven’t really missed it much. But it’s high time I start sharing some of my experiences. I will, under the banner:
One of my pet peeves was the general lack of math skills in a newsroom. That goes for anything involving statistics or finance. I used to joke that “Two out of every three television news anchors can’t handle simple fractions… that’s almost half.”
Well, here’s what my wife overheard yesterday morning (names and stations being withheld to protect the innumerate):
“A new study shows that 20 percent of all homeless people are veterans. That’s one out of every four.”
Forget the Writers’ strike. You just can’t make this stuff up.
[tags]Ike Pigott, Occam’s RazR, broadcasting, mathematics[/tags]
I am a mathematician
Jul 11th
…or at least more so than the guy who claims to be one in the latest “Screw-Big-Oil” e-mails that is wafting past my inbox.
This one has THIS IS NOT THE ‘DON’T BUY’ GAS FOR ONE DAY, BUT IT WILL SHOW YOU HOW WE CAN GET GAS BACK DOWN TO $1.30 PER GALLON in it. It also includes this rich statement:
This was sent by a retired Coca Cola executive. It came from one of his engineer buddies who retired from Halliburton.
At least we now know the real reason Pepsi has been shut out of Tikrit.
Actually, this isn’t going to be another debunking e-mail – David and Barbara Mikkelson have already handled that at the Urban Legend Reference Pages. My beef is with the bogus call to authority:
I am sending this note to 30 people. If each of us send it to at least ten more (30 x 10 = 300) … and those 300 send it to at least ten more (300 x 10 = 3,000)…and so on, by the time the message reaches the sixth group of people, we will have reached over THREE MILLION consumers.
If those three million get excited and pass this on to ten friends each, then 30 million people will have been contacted! If it goes one level further, you guessed it….. THREE HUNDRED MILLION PEOPLE!!!
Again, all you have to do is send this to 10 people. That’s all!
(If you don’t understand how we can reach 300 million and all you have to do is send this to 10 people…. Well, let’s face it, you just aren’t a mathematician. But I am . so trust me on this one.
Trust you? Hah!
A real mathematician wouldn’t have made elementary mistakes – like confusing possibilities with probabilities. Go ahead – send an e-mail to 10 friends, and ask them to send it to ten more. What are the odds that you have many of the same friends?
If we assume that there are 300-million people in the United States, and
if we assume none of them are children, and
if we assume that nobody has multiple e-mail addresses, and
if we assume there is no duplication in the lists,
then maybe I’ll buy this explanation.
The thing that really chaps me is how often we fall for bogus claims to authority. “I am a mathematician, so trust me on this one.” It’s a classic fallacy of informal logic, to attempt to belittle or berate the reasoning of another. It is also related, loosely, to a weak ad hominem argument: you’re not as smart as I am, so you must be wrong.
There’s also something very, very wrong about a person who issues an unsolicited promise. “Trust me.” You know, I never indicated that I didn’t trust you. Yet you proffer an explanation. Psychologically, you’re telling me a great deal about your trustworthiness, or lack thereof. (citation: The Gift of Fear, by Gavin de Becker.)
(standing on soapbox) – I get a little testy when people try using the appeal to authority, and in a bullying fashion. Most often, I see it on forums and blogs pertaining to climate change. I’m not going to waste my time digging through data and reports that are prefaced with “The debate is over, and all the experts agree.” Well, as a matter of fact, they don’t agree. And don’t start your argument with “Dr. Gasbag is a shill for Big Oil.” That’s an ad hominem if I’ve ever heard one.
You want to frustrate those who abuse the appeal to authority? Just calmly ask questions. And when the answers come couched as attacks or informal bullying techniques, politely ask the question again.
“Trust me. I’m a psychologist.”
[tags]Ike Pigott, Occam’s RazR, mathematics, economics, argumentation, rhetoric, psychology, The Gift of Fear, Gavin de Becker, Urban Legends[/tags]
A Prime Time Mystery
May 24th
{Scientists still don’t know why the cicadas emerge every 17 years.}
That’s not an exact quote from ABC News reporter Claire Shipman, but it is fairly close. On this morning’s GMA, she filed a report about the emergence of “Brood XIII,” a noisy event to be sure. For a moment, I was baffled as to why she’d refer to the 17-year cycle as a mystery, especially since the answer is based in math and common sense. (And was the subject of several articles by the late Stephen Jay Gould as far back as 30 years ago.)
It’s simply a matter of evolution and adaptation. It’s a survival strategy. Burst forth in giant numbers, and predators will never be able to eat your species to extinction. However, if you emerge from your slumber every year, your predators can adapt their numbers to enjoy a regular lunch. If you pop up every other year, you’ll be a meal for every predator with an even-year cycle of its own.
Over the generations, those cicadas that emerged on a cycle that was a prime number were the least likely to bump across a predator with cycle that might sync up (and wipe out the brood.) Which explains why the two largest brood cycles are now 17 years and 13 years – two somewhat large primes that don’t fit any easy multiples.
What I find interesting would be the eventual intersection of those two brood cycles, which happens every 221 years. There’s no real guarantee that both would pop up within the same window, but it is possible (broods make their noise, make their babies, and die off within 30 days or so.) The 13-year Brood hit in 2004, with the 17-year Brood firing up now. Here is the breakdown of future years: More >
