Republican New Math

We are witnessing truthiness in action here....

Their math is, unfortunately, used by the touch-screen voting machines.

Surely you mean 'diebold new math'...

I woke up to that exchange yesterday, and boy did it make my head hurt.

While he phrased it imperfectly, Karl's point is that he thinks the polls he's got, and the rules he's using to handicap them, are more accurate than the ones the media is mostly talking about. If I had to bet, I'd say he's probably right. Karl's is a data-driven operation, and he's had great success with it. I bet they're going to keep the house and the senate.

would it be hyper-pedantic, or just wrong, to point out that they're not even talking about "math"...it's arithmetic!

I heard that interview the other day. I couldn't believe what I was hearing...The arrogance displayed by Rove is staggering. Of course he won't share his data so they can actually have a productive conversation...He needs the secret gnosis to preserve his smug and superior attitude.

Reminds of the hybrid South Park episode..."cloud of smug.."

While I'm no fan of Mr. Rove, I've got to agree with steve s on this one. Rove's just saying that he's run a different set of calculations than Mr. Seigel has and that he likes his computations better. You have to disambiguate the technical meaning of "math" from some of its more colloquial definitions. By "math", Rove clearly meant "result of some computations/arithmetic" in the same sense that people mean when they say "do the math". They almost always mean "crunch the numbers". And by '"the" math', he clearly means "the best numbers (in his opinion)" not some fancy new "Rebulican Mathematics".

Also keep in mind that it's hard for anyone to be perfectly articulate in extemporaneous speaking. I doubt Rove would have written what he said in response to a written question, but it's easy enough to misspeak during an oral interview.

The only kind of math that matters on November 7th is Diebold's.

billb and steve s:

I agree with your interpretations of what Karl Rove meant by "math". However, I feel he needs to validate his invisible definition. After all, they could be using "math" based upon bones tossed onto a plate or the patterns in cow dung.

Even if their "math" is based upon rigorous probability and statistics, I still don't trust them to get it correct.

Hypergeeky response:

http://www.tbm.tudelft.nl/webstaf/gertjanl/dacosta.pdf
"Multiply Modal Extensions to Da Costa's C_n, 1 <= n <= omega, Logical Relativism, and the Imaginary"
by Gert-Jan C. Lokhorst
The Journal of Non-Classical Logic 5, nr.2 (1989), 7-22.

Some people will love the metalogic equations, others might enjoy the narratives of Introduction (pp.1-2); Section 3.1 "Vasil'ev's imaginary logics and worlds" (pp.5-6), Section 3.2 "The Logic of Belief" (pp.6-7); Section 3.3 "Logical Relativism" (pp.7-8); and the conclusion.

I cannot watch Emperor Bush II and his minions without an Anthropological or Fictional stance. Science Fiction (per the Cranky Book Meme thread here) very much depends on imaginary worlds, some with very different Math. Think also of Greg Egan's "Luminous" [1998] and "Permutation City" [1994].

Bush and Rove are aliens from a cosmos with different math, utterly beyond our multiverse. Explains a lot.

I don't get it, Astrochicken. Rove says he has some polling data that is better than what Seigel has been seeing, and that his number cruching puts the Repubs on top. You imply that he is due criticism because he won't show the world his data and methods, but this is politics not science. He's entitled to run his own polls and be confident in them and not need to share that data with anyone else. There's no reason for him to feel compelled to give away any compettitive advantage by revealing his sources before the election.

Yeah sure, Rove has secret sources which are the real sources, there's a Silent Majority, and Bush has a Secret Plan to End the War. CREEPS, all of them; what goes around comes around and history repeats.

At intrade.com, shares of CT.SENATE06.DEM are going for one-third of the price of HOUSE.GOP.2006, yet the Kossacks don't seem to have thrown in the towel on Lamont. Are they using alien math, too?

It's not merely the Diebold math - as if that were not bad enough. It is also the gerrymandered redistricing, the pre-election disenfranchisement campaigns, and the selective vote "spoilage", just to name a few under-reported issues. See Greg Palast's recent article for a litany of the types of calculations that would give folks such as Rove confidence now. He does indeed have access to considerably more "reality-based" information about how the election will be executed than we do.

It is also the gerrymandered redistricing...

This right here is what I consider the worst. Every single one of the 435 house seats is up for reelection this year, yet at least 374 of those are not considered remotely competitive (and probably more are not *really* competitive). That seems to run contrary to the notion of people being able to enact democratic change.

Come on people, Rove is just being an intelligent politico. It would be suicide for him to come out and say "Wow, it looks like we're going to get killed". He knows that for some stupid reason, Americans are more likely to turn out to vote if they think their guy is going to win. He's not a scientist, he's a football coach making a halftime speech to a team that is losing.

As far as the gerrymandering issue, we have a nice scientific test of that, since one can't gerrymander the Senate. If the Dems take the Senate but not the House, especially if the margin is a large one, then the gerrymandering argument looks strong. OTOH, if the Dems take the House and not the Senate, as looks most likely, then we can at least breath easier on that issue.

Of course Rove has every right to say, however clumsily, that he's got better data and analysis than the public and press. Being a political strategist, he should be expected to shade everything he says in some way that he thinks will benefit his candidates.

But what was his tone like in the interview? In the transcript he sounds downright belligerent. It's curiously consistent with the Rove image as an utter partisan, as if he were expressing his contempt for Siegel, NPR, and NPR listeners and playing to the secondary crowd of people who might hear an excerpt played back by Limbaugh or Hannity or someone else who shills for the Republicans.

If you listened he was belligerent, too. Some would say rude.

As we've learned countless times from this administration, reality doesn't matter. If Iraq is bad, they say it's great. (If it happened to be great, they'd also say it's great.) Repeat it over and over. And over. Never admit failure is coming. Never admit it's possible. Never admit it happened and continues to happen. If someone counters with facts, counter with loudness. If someone questions the numbers, counter with belligerence.

Rove also likes to maintain his image as an enigma. "I've got access to knowledge from sources you've never imagined could exist. That's all you need to know. Now go feel sorry for yourself. I'm Karl Rove, bitch!"

Yes, I prefer to be non-associative with Rove, and non-commutative too for that matter (Lex Diebold).

Surely if Rove had any decent polling data which showed that the Republicans weren't going to get spanked, it would at least get onto Fox News?

Ron Avitzur's comments suggest that the GOP is in the position of the subjects of medieval witch tests: The only way to prove their innocence is to die. Well, I guess they should have thought of that before they started turning people into newts!

billb:

Of course he doesn't have to divulge his methods. However, as a citizen I demand that he does in order to attain credibility in his assertions. I demand that from the entire administration.

Will it happen? No.

Astrochicken:

What's the difference between a demand and a request?

Nat Whilik:

If I demand something from you, I will be upset if that demand is not satisfied. If I request something something from you, I'm giving you the option to do nothing for me.

I demand that the current administration be transparent and provide a method to their madness. They do not. Therefore I am upset.

I think the point here is not that he means something patently false (mathematics somehow operates differently inside GOP headquarters--flashback to Life, the Universe and Everything and the Starship Bistromath), but rather that he's babbling, insisting on using words that don't mean what he's trying to communicate. Is he being deliberately obfuscative? Or is it that he just can't construct a clear sentence? Idiot or evil genius? It doesn't matter, both are evilly (oops, freudian slip) equally repugnant.

Especially repugnant to a mathematician (or at least, to my mindset as a mathematician) is Mr. Rove's insistance on making new uses of words to express things that English already has a perfectly good way to express. If he had replaced "math" with "methods" or "results" or "statistics" he would have been perfectly correct. Not only that, but when the interviewer tries to point out that he was babbling--(to paraphrase) "I'm afraid you aren't entitled to your own math, any more than you're entitled to your own physics, your own economics (where the economy is booming), or to otherwise live in your own little universe"--he responds with deliberate condescention.

"What's the difference between a demand and a request?"

That reminds me of something I read once, about how strangely children parse what they observe. A youngster reportedly listened to two adults telling a series of jokes with format: "What's the difference between a [widget] and a [wadget] and [boff]?"

The child asked:

"What is the difference between a crocodile?"

The adults were baffled. The child answered his/her own riddle:

"Because it swims faster than it walks!"

... and laughed.

I remember being very confused in the 1950s by "Tom Swifties" -- as it turned out, because I did not know waht an adverb was.

There's the old Math joke:

Prove that the crocodile is longer than it is wide.

Lemma 1. The crocodile is longer than it is green:
Let's look at the crocodile. It is long on the top and on the bottom, but it is green only on the top. Therefore, the crocodile is longer than it is green.

Lemma 2. The crocodile is greener than it is wide:
Let's look at the crocodile. It is green along its length and width, but it is wide only along its width. Therefore, the crocodile is greener than it is wide.

From Lemma 1 and Lemma 2 we conclude that the crocodile is longer than it is wide.

I suppose that this whole email would be clearer if expressed in Category Theory. John Baez emailed me to say again how good a job Mark C. Chu-Carroll is doing here. To paraphrase, the Cat Theory of Computation here is the easier part, but that's the hardest and most important part to learn!

very strange: if sixty-eight varied-quality polls of sample size 500 to 1000 were to be culled, would that produce a "better" poll result with a smaller margin of error? even if one could somehow guarantee that the poll questions were the same, how can one even begin to deal with selection bias and other self-serving polling practices? mr rove, having lost his keys in the dark, needs sixty-eight streetlamps to look under...

Astrochicken: I demand that you give me a million dollars!

As long as Rove is spending the RNC's or individual candidates' money to do his own polling, you don't have any legal right (and probably no moral or ethical right) to a full answer to your demand. Do you make the same demands of James Carville? Does he answer them? Do you foam at the mouth (err...keyboard?) when he refuses?

lytefoot: I suppose you've never said a phrase including the words "do the math" when you really meant that you were going to do some calculations?

All: I sure Mr. Rove deserves his share of condemnation and criticism, but this, like Slate's regular Bushisms column, just seems like a petty attempt to take a cheap shot a public figure. I'd rather see a substantive conversation/debate on some real mathematical failures of this administration (or of any government's, past or present, mathematical failures) especially as they pertain to government policy. This kind of petty sniping seems beneath us all.

lytefoot: I suppose you've never said a phrase including the words "do the math" when you really meant that you were going to do some calculations?

Ah... but that's a legitimate English expression. It's part of the consensual illusion that is our language. I didn't make it up to obfuscate my meaning. The person with whom I'm conversing and I (assuming of course that they're also a native English speaker) have what amounts to a prior agreement that the phrase "do the math," an independent lexeme from the word "mathematics" when used in its formal sense, means "perform whatever calculations are necessary to this problem." Math is an overloaded word. This is acceptable.

On the other hand, when Mr. Rove says, "I have my own math," there is no prior consent to this overloading of the expression. Indeed, all the prior similar overloadings of this expression (e.g. "new math") seem to imply that he believes that the laws of mathematics don't apply to him. Certainly, the burden is on him to make his meaning clear.

This discussion threatens to stray into linguistics...

This kind of petty sniping seems beneath us all.

We are dealing with petty little men. Their petty absurdities can't be addressed in a rational manner, because there's no rationality to latch on to. It's impossible to debate rationally with someone who attacks an argument either by denying its premises ("I never said we would be greeted as liberators") or by insulting the opposition.

Petty ridicule is the only possible response.

What compells you to respond at all?

Lytefoot:

I gave no prior consent for you to spell "condescension" that way. I demand that you stop it!

Rove has The Math, Bush has The Google.

Both are cheap shots. Both are funny.

Republicans lose elections, entire field of Mathematics thrown into turmoil
Film at 11.

Bush called Rove his "Architect of Victory" in the 2000, 2002, 2004 elections. Does that mean that Republican Math-master Rove, in the 2006 elections, was the Architect of Defeat?

subthread on Thomas McDonough's fine novel "The Architects of Hyperspace" eliminated here, but you can Google it or find it on Amazon. Good title for this blog, though.

In an ideal world, the current polls should have no impact on the voter. Voting isn't a test where you try to guess what everyone else is going to do. Well, it isn't for me. If the polls show that Bush is ahead, but I don't want him in office, I still won't vote for him. Voting for the one who is not going to win is still a vote. Bush did not have the mandate of the people. That's what the voters said.

In the real world, people don't seem to like to vote for a loser. It's stupid, IMO.

The tricky spot is this: I want Nader, but if I don't vote for Gore, then we get Bush.

http://www.geocities.com/electionmodel/TruthIsAllFAQResponse.htm#MidTer…

The 2006 Mid-term Elections

This is the final of three pre-election articles I wrote with Michael Collins (autorank) and Alistair Thompson of SCOOP. The purpose was to quantify the risk of fraud in the 2006 Mid-terms. The analysis forecast that the Democrats would gain control of the House and Senate. It also indicated the House seats and Senate races where fraud was most likely to occur.

The vote fraud model projected that the Democrats would win at least 240 House seats, but lose 10-15 to fraud. It correctly forecast that they would gain control of the Senate by winning six GOP-held seats, although they barely won Virginia and Montana. The Democratic Tsunami overcame the fraud in the House and Senate.

There is no longer any doubt that the poll/vote discrepancies were caused by uncounted and switched votes. Evidence of fraud was once again found in the Final National Exit Poll which matched to the recorded vote count with the use of fallacious weightings. The 2006 Final "How Voted in 2004" weights were manipulated just like they were in the 2004 Final "How Voted in 2000".

In 2006, the weights were transformed from 47 Bush/ 45 Kerry at 7pm to 49 Bush/ 43 Kerry at 1pm the next day! This replicated the 41 Bush/39 Gore to impossible 43 Bush/ 37 Gore weight changes in 2004. The net effect of the change was to cut the Democratic margin in half -from 55-43% to 52-46%! Applying realistic weights to the 7pm NEP (using the 12:22am 2004 NEP) the Democratic margin becomes 56.7-42.1%, exactly matching the 120- Generic poll trend projection! Was it a coincidence or confirmation? You decide.

Here is a Comprehensive Mathematical Analysis of the 2004 and 2006 Elections

http://www.geocities.com/electionmodel/TruthIsAllFAQResponse.htm

For the record, I have three degrees in mathematics, including an M.S. in Applied Mathematics and an M.S. in Operations Research.

Over the past 30 years, I have developed engineering, financial and investment analysis models working in the defense/aerospace industry, on Wall Street for several major Investment Banks and as an idependent consultant specializing in financial analytical software applications.

This link contains mathematical analyses far more extensive than any of the articles written about the 2004 and 2006 elections.

The analysis fully debunks the naysayers who still believe that Bush won the election and peddle faith-based theories such as the "reluctant Bush responder" and "false recall".

Start off with this fact: In every presidential election, approximately three percent of total votes cast are never counted- at least 75% of them are democratic. That accounts for 1.5% of the exit poll discrepancy. The remaining 5% is due to electronic vote-switching. I give you Florida and Ohio to mull over.

The analyses are a result of two years of Excel model development.

It's all here: Regression, Optimization, Simulation, Sensitivity Analysis, pre and post election state and national polls - all leading to the same conclusion:

Kerry must have won by over 7 million votes with 52.5% of the vote.