Site Meter

Wednesday, October 31, 2012

Is there a Real Clear Politics Method (

I have been looking at different poll aggregators a lot.  I like Real Clear Politics, mostly because they don't use LOESS but more generally because they use simple averages.  I like simple and transparent.  The one question is which polls to average.

My guess as to their rule would was: no internet polls (definitely the case) and no partisan polls (counting Rasmussen as non partisan because that's what Scott Rasmussen says).

I was wrong.  I now see that they have no rule.  Sometimes they include PPP and some times they don't.

for Ohio yes

Nation wide no

One of the national PPP polls which they ignore is a tracking poll. There is no risk of selective publication. There is no justification for ignoring it while reporting other PPP polls.

The choice isn't between silly rules (LOESS smoothing and trend extrapolating) fancy rules (Nate Silver) or simple rules.  It is between silly, Silver and arbitrary.

Sunday, October 28, 2012

Polling Aggregator Obsession

The 2012 Presidential is so close just so close that not only do different polls show different results but different averages of polls.

OK it is clear that extremely fancy models such as the fivethirtyeight model which attempt to assign undecideds and use state data to estimate the national vote and vice versa are different (generally I don't        
like fancy but Nate Silver has a track record).  But simple smoothers are also different.

Partly it is the choice of polls to include yes no on web based or partisan polls. is a simple average but they don't include partisan polls or web based polls (and use self reported partisanship so Rasmussen is included but not PPP ooops).

But part of it is the smoothing algorithm.  both www.talkingpointsmemo and use loess smoothers which fit a constant and a coefficient on time using weights which decline away from time t and report the fitted value for t.  I think this means they are too quick to extrapolate trends.  This means that one outlier can convince the computer that the polls have been trending for a candidate.

My example the odd case of TalkingPointsMemo and Michigan.  The program has Ohio leans Obama and Michigan a tossup.  This is due to one poll by Foster-McCallum (in serious contention for the bitterly fought prize for worst pollster).  Without them, the computer estimates that Obama is 3.9% ahead.  That is enough for TPM to call the state leans Obama and give more than 269 electoral votes in Obama leaning states.

Here is the graph with Foster-McCallum

Here is the graph without them

This is extreme, because Foster/McCallum had a Florida poll with an absurd sample which earned the coveted and very rare double asterix for not included in the average because of an editorial decision.

I think such decisions should be made pollster by pollster not poll by poll.  But totally aside from that a sensible smoothing algorithm shouldn't be so sensitive to one poll when 10 October polls are available all of which polled after the first TV debate.

Notice how the poll pulls down estimates of the state of the race before it's sample began.  This feature of the smoother makes it hard to see if a shift is due to an event.

I say even simpler is better than Loess.

Saturday, October 27, 2012

Comment on Nyhan on Mittmentum

Brendan Nyhan autopsies the mittmentum narrative here.

I comment.

Thank you for this very good post.  I had been puzzled by the "bad narrative about momentum" narrative on progressive blogs (largely because I interact with the rest of humanity largely by reading progressive blogs).  You demonstrate that there was indeed a mittmentum narrative.

I'd like to add some thoughts.  First, aside from the general tendency to seek facts which confirm the story we want to tell, people have huge amounts of trouble with time series which have no momentum.  This has been demonstrated by psychologists.

The experiment is to show people a random walk -- a time series in which the changes are uncorrelated so the best forecast of where it will end up is always exactly the current level.  Subjects just can't resist perceiving mean reversion or  momentum. These are opposite.  Mean reversion is called "a bounce" in the campaign literature.  The idea is that a recent change shall partly fade away and things will go back to the way they were.    So immediately after the first debate, there was much discussion of whether Romney's gains were a bounce -- destined to vanish.  Momentum is the opposite, people see a trend and extrapolate it so they think a shift will not just last but grow stronger over time.  Basically people always see one or the other when they are shown data with neither.

These opposite errors are linked.  The idea that a random series of changes shouldn't move in the same direction for a while (so changes are perceived as bounce) makes people think something funny is going on when they happen to be in the same direction for a while (we see a new trend).   Notably, since polls contain sampling error which is independent across polls, polling averages contain independent changes and give the sort of series which we just can't mentally accept.

Tracking polls make things trickier still.  Because they are moving averages, the will have trends even if overall public opinion doesn't.  Say the Gallup poll will tend to move in the same direction for a week, because it is a weekly moving average.  It is psychologically hard to see something trend for a week and not extrapolate the trend.  This is true even if one knows that the data are a weekly moving average.

I don't know votamatic (I check Jackman, Talking Points Memo, fivethirtyeight ,and Real Clear Politics uh often).  Jackman and the TPM group add to the problem.  Their smoothed average is LOESS which means they calculate the value for t as follows: estimate a time trend with data with weights which decline for polls further from t then report the fitted value for time t.  this means that they extrapolate trends.  The approach assumes that there is momentum.  Then people extrapolate the trend in the extrapolated trend.    I  have checked and confirmed that dropping a medium old poll with huge Romney - Obama can cause the current estimate of Romney minus Obama to go *up*.  I try and try to find such cases and have found one or two IIRC think the McLaughlin for Allen Virginia poll was one (the web user clicking buttons experiment depends on what other data are used so can't trivially be reproduced using tools at the sites).

There is also just a lag in reporting.  Total webaholic political junkies consider early October ancient history.  The ink, paper and TV reliant not so much.  This post stresses how a 10 day old pattern was noted 3 days after it was confidently asserted that there was no such pattern.  7 days  has not always a huge amount of time.

Finally there is spin frankly reported as spin.  A whole lot of the momentum stories quoted (often anonymous) Romney campaign staff claiming they had momentum.  It is just a fact that they made those claims.  Importantly the Obama campaign didn't push back (Ezra Klein claims this and he would know).  Reporters just don't report that while he said and she said the same thing, the data show something else.   I think many consider reporting the mutually denied facts to be unethical -- they feel they can't contest Romney campaign claims more than the Obama campaign does.  Now I have no idea why anyone granted anonymity to flacks pushing the campaign's narrative and I don't like the he said she said approach.  But it is not the same as perceiving momentum and reporting that there is momentum.   They said Romney flacks said Romney had momentum, because this is just what happened.  We have a problem, but it is not in our common psychology but in the manipulability (sp) of reporting based on   simple rules.

Anyway sorry for the overlong comment on your excellent post.

Friday, October 26, 2012

Comment on House Effects

Simon Jackman publishes his estimates of pollster house effects.

I comment

I think the post illustrates the fact (which you stress) that estimated house effects are uh estimates with standard errors.  This is made very clear by the fact that the estimated house effects are quite different for Pulse Opinion Research and Rasmussen.  IIRC they are two different names for the exact same pollster (Pulse when the poll is commissioned Rasmussen when they do it on their own with the business logic of getting publicity).

I also think a calculation for Pulse or Rasmussen would be interesting (and fairer than fair to Rasmussen since it would include 11 polls with a small house effect which they present under a separate name).  The a Rasmussen under any other name would smell as sweet estimate will be very close to the Rasmussen called Rasmussen effect, since there are only 11 Rasmussen called Pulse polls).

[Now I can't help wondering about Rasmussen's business plan.  The  poll and publish the results for free to get publicity strategy might be a big mistake if the publicity is bad publicity because the published polls show they are biased (we have to wait for the election to estimate bias rather than house effect but Rasmussen rep minus dem - actual election Rem minus dem averaged 3.8 % in 2010).  I think however that it is a brilliant strategy.  Rasmussen has become famous and much loved by many, because of the huge Rasmussen Republican house effect. I googled Rasmussen and "most accurate pollster" and found a huge number of recent 2012 hits to definite assertions made *after 2010* that Rasmussen is the most accurate.  This in spite of Rasmussen having a huge mean squared error (and as noted a huge mean error) in 2010 (and 2000).  Clearly there are many people who equate "largest Republican House effect" and "most accurate".  A reputation for being reliably Republican is clearly extremely valuable to  Rasmussen.  It is clear that there are many people who sincerely equate pro-Republican with accurate, unbiased, fair and balanced.

So it might hurt them to include the Pulse results and get the estimate Rasmussen house effect down. But it's still the right thing to do.

I think it is clear that the huge Gallup house effect is the result of an admirable determination to not fiddle the numbers.  Gallup is still using the likely voter filter they developed decades ago (at least the description seems the same to me -- it might be subtly changed in a way which is numerically very important but passed my this seems familiar vague memory screen).  They developed an excellent record of getting it about right with that filter.  Then in 2010 they missed the actual Republican - Democrat average over congressional races by 9%.  It is very clear that they were worried about the LV filter in 2010, because they began publishing alternative LV estimates using the same sort of LV filter as others (basically if you say you are likely to vote you are counted as a likely voter).

It seems that something about the population has changed.  I don't know what.  My guess is that the Gallup LV filter always gave a sample which was older than the population of people who actually voted.  This would not be a huge problem back when the LV filter gave good forecasts of results, because there wasn't a huge difference in partisan preferences by age.  In 2008 and 2010 there were huge differences (older = more Republican).

Then Gravis Marketing the other pollster beyond Rasmussen.  I am quite sure they are accutely embarrassed by their house effect.  This is because a Gravis Marketing employee spent an amazing amount of time and pixels in the comment thread to a post of mine about their house effect.  One explanation of the Rasmussen house effect is that they try to poll all in one day.  Others choose phone numbers and then call and call day after day (for 3 or 4 days) till they get an answer.  All in one day pollsters over-sample people who are home a lot.  Note that the first Gravis polls were conducted all in one day and then they called for 2 days.  This was clearly (first person claim) an effort to avoid a bias which might be showing up as a Republican House effect.

Over on the other side, I just don't get Zogby.  The real phone based Zogby poll is perfectly respectable.  The web based poll Zogby Interactive now renamed JZ analytics is notoriously absurdly unreliable.  This has to be terrible publicity for Zogby.  Using just initials and not a name is not enough to hide the association from, well the people who might hire a pollster. It isn't even a strategy to get Democratic organizations to commission reliably Democratic slanted polls to affect the narrative.  Newsmax is a major consumer of JZ analytics polls.  I just can't imagine the thinking of John Zogby or whoever runs Newsmax (I type whoever as I assume there is a human being involved although I can't claim that Newsmax editorial policy has clearly passed the Turing test).  ]

I think some arithmetic assistance might be useful.  You show the house effects for Obama's vote share and stress that all estimates are in two party terms.  I think it would be helpful if you explained that the house effect for Obama minus Romney is twice as large (I am presenting myself as arithmetically sophisticated but I typed "slightly less than" because I mis-remembered "two party vote share" as "percent support").  Yes some readers will be insulted to have arithmetic explained to them, but others will be helped.

Note: Huffdweebs have a word limit so this is longer than my comment there.  I let it all hang out here.  The stuff that isn't there is in [].

Wednesday, October 24, 2012

Comment on Greenwald

Ouch.  I admire Glenn Greenwald and don't like writing or posting this.  It is a comment on 3 words in this column.  Please read it before reading my comment.

I think that you are unfair to Klein.  I think your argument is based on a false dichotomy between "indiscriminate" and discriminate.  Klein isn't advocating killing all Pakistani 4 year olds or even killing Pakistani 4 year olds chosen at random.  You assert that he is advocating exactly that.

You do not recognize a distinction between killing, say, only uniformed soldiers of an army which has actually invaded one's country (I think you accept this) and killing people at random.

The claim that deaths are collateral to a legitimate military aim is not an automatic absolution from all alleged culpability.  However, it is also not a completely irrelevant claim as you assert.  

I think I am being totally 100% fair to you.  when you assert that Klein's moral vision is identical to that of terrorists, you assert that there is no difference at all whatsoever between taking the risk that civilians will be killed and deliberately killing civilians.   You assert that the acting with the certainty that the course of action will imply the death of innocent civilians is equivilant to "indiscriminate" killing.  Thus you assert that anyone who allows cars to be driven is a terrorist.   If the only choices are to be a terrorist and to avoid civilian deaths at all cost, then we can do nothing but avoid death.  Airlines are terrorist organizations because one in a million flights ends in a crash.

I have called you (with absolutely no irony whatsoever) the George Orwell of our day.  However, this column is consistent with the assumption that you are a moral idiot.  Did you read what you wrote ? (I ask for information, as I honestly suspect that you didn't (and admit that I plan to post this comment without reading it).  Did you think at all about the  definition of  "indiscriminate" before typing it or before posting ?

I stress that my criticism of this article has nothing to do with drones.  Let me stipulate that current US policy is monstrously criminal.  Under that assumption, I assert that your use of "indiscriminate" and "terrorist" is not at all serious.  It demonstrates contempt for language and logic and morally important distinctions.

Monday, October 22, 2012

comment on Kessler

Glenn Kessler has been doing his job lately.  He has been calling Romney lies lies.
But he does feel the need for Ballance.  So in his latest he recalls giving the Obama campaign 4 Pinocchios for a claim about Romney and Bain which was supported by massive documentation.

I think on the particular issue of when Romney stopped managing Bain, Kessler fell for Romney lies and just won't admit he was wrong.  I also think he feels that he must Ballance his criticisms of Romney.  The reference to Bain and outsourcing has nothing whatever to do with the ostensible topic of the current article.

Kessler is also clearly wrong on the substance.  Also the June fact check which he brings up only for Ballance bases his decision on what "seems".  That is his word.  He couldn't have made it clearer that his so called fact checking is an expression of hunches and opinions.

I give Kessler 4 Posties for arrogant refusal to admit he is fallible, stubborness and Ballance (I know that the number of criticisms isn't clearly four but they seem roughly four in number to me).

I put my angry comment here where Kessler won't read it, because I don't want to irritate him.  It is clear that criticism angers him and I honestly hope that he will write an opinion column (you know on the op-ed pages where it belongs) about how Romney is amazingly dishonest.

my comment

You just won't let your stubborn insistence that official SEC documents don't matter go will you.  I had decided to forget about your Bain fact check gross error, but you have chosen to re-repeat it.  I quote from your 4 pinocchio fact check

..." the Obama campaign supplied reams of additional SEC documents regarding Romney’s *ownership* in Bain after he left for the Olympics, most of which we had examined previously when we first looked at this question. The campaign also supplied SEC documents showing that two of these companies, Modus and SMTC, as well as one called Stream International (a predecessor of Modus), earned money in part by helping other companies subcontract work overseas. Some of this business predated Romney’s departure from Bain, but thus far it *seems* a slim case for this particular ad."

(emphasis mine)

So you concede that the basic claim in the ad is accurate then give it 4 Pinocchios on the basis of "seems".  I recall my very first angry comment here.  It was related to the Obama campaign and Bain and your use of the word "seems."  You claim to be a fact checker.  You just can't use the word "seems".  If you don't know, do more research.  Your 4 Pinocchios for Obama are, in your own word, an expression of opinion.  I think fact checking is very important.  I suggest you consider that it might be and stop writing about what "seems".

I also note your pathetically totally utterly dishonest use of the word "ownership."  Romney's continued ownership of Bain when he was in Utah for the Olimpics has never been contested.  The issue on which you incorrrectly asserted that the Obama campaign had made a false claim (making the error because you didn't check publicly available documents) was whether he was a manager..  The SEC documents list him as CEO and President.  They constitute proof that he was a manager when you said he wasn't, that the Romney campaign lied and you fell for their lie.

You decided that your hunches count for more than official documents by citing an anonymous legal source.  That was a plain violation of Washington Post policy.  You did not explain why you granted anonymity.  I can't imagine a justification.  I think that you must name your source.  I strongly suspect that you didn't, because the source is obviously grossly biased in the direction pleasing to you, because you just won't admit your error.

You asked the Romney campaign if Romney attended board meetings (as sole shareholder and chairman of the board at the time. when they refused to answer (tacitly conceeding that he had)  you said that moving forward your writing on the matter would be different as you would consider claims on a "case by case basis".  The June fact check which you, not I, brought up proves that, after you were forced to admit you were totally wrong, you went right back to the same claims from pure stubborness.

Everyone makes mistakes.  Your refusal to accept that you do to threatens your standing as a fact checker.

Friday, October 19, 2012

Not so Rude Comment on Pollack

I love this video

I agree that it is very odd for Romney to brag about balancing a budget with a 1.5 billion dollar federal bailout. But what exactly does it mean for a manager of a firm to balance a budget ? I’d say the very minimum is to keep the firms you control out of bankruptcy. By this standard Romney is not a budget balancer at all. At Bain he made huge amounts of money relying on limited liability. Several Bain controlled firms went bankrupt. Those were budgets under Romney’s control (sole shareholder CEO and all that) which were as un balanced as budgets can be.
I know I am mixing up budgets and balance sheets. I am doing that to bend over backwards to try to meet Romney half way (now that the upper hand is on the other foot). Really a corporations budget isn’t balanced if it issues debt. Unbalanced budgets are the key to private equity. Romney’s whole career as a businessman is based on understanding that an obsession with avoiding debt is costly.
On this he is consistent in practice as well as being consistently dishonest, since as you note, he proposes a huge increase in the US budget deficit. The approach of loading an entity up with debt and not worrying about what happens if it can’t pay has worked very well for him so far, so why shouldn’t he stick with it_
I’m very sorry I wrote an outraged comment on your post critiquing #rateloweringbasebroadening (RLBB)for not critiquing it completely enough. In this video you make the key point that extremely high income people will gain money they sure don’t need from RLBB as they just don’t have deductions on the order of their income (importantly this refers to Romney style RL and maybe a bit of BB (but no details) and things are different if the BB includes raising taxes on capital income and capital gains).
On Gas prices, I note that US consumption is a large fraction of world consumption. A US gas tax probably would cause lower petroleum prices in the medium run (not immediately only after people trade in their SUVs for cars). It makes no sense for a country to pretend it is tiny when it is large. Laissez faire is optimal only when price taking is optimal. In the world petroleum market, the US acts as a monopsony which has no interest in maximizing profits. It’s as if Saudi Arabia ignored the effect of their exports on prices. Of course you know this (you praised Mankiw on the gas tax). Fleet economy standards may be a silly way to do it based on the US obsession with low gas prices, but I suspect that Obama’s policies are causing lower petroleum prices and will eventually cause significantly lower petroleum prices. “There is nothing [more] a US President can do” is true given and only given political limits.

OK so enough being responsible. Now a dumb joke about a smart guy -- a smart Pollack joke
Love the denunciation of rate cutting and base broadening.  You are the most effective populist since  President  Jackson, Pollack.

Wednesday, October 17, 2012

Severe Bipolarism Disorder

I think that yesterdays debate went OK for Obama.  Today Chris Van Hollen e-mailed asking for money

Wasn't it just about the day before yesterday that the risk was we would be convinced it was lost already ?

My party suffers from bipolarism disorder.

Also far from me to insult my mother's representative (my father's too) but why exactly is the chairman of the DCCC worried about overconfidence ?  The big big 0.5% lead in the generic ballot ? Or is he maybe writing to me about the presidential so I will read the e-mail, but really thinking about the congressional ?  That is his job.

Just When I thought it was safe to get back in the Sample pool

I just spent about 10 minutes at dailykos reading and thinking about the results of YouGov polls (main thought well it's not a real poll but it's pleasant reading or ouch but hey it's just YouGov crap).  Little did I know how crappy they were.   Then the excellent Steve Singiser explained

that YouGov massive release of data from two dozen states. However, and this is worth noting, these are not traditional polls. They are "callbacks" of people who were surveyed online last month. As such, they are useful to look at how those folks have shifted their preferences since September, but the caveat that these are not unique samples should be remembered.

I conclude that the people at YouGov don't read my blogs and missed my criticism of CNN/ORC panel polls on debates.  That's not the way it should be done, or rather, the headline result of the second wave is not worth reporting.  Repeat interviews are very useful. The point is to see how people have changed their minds and not to mistake fluctuations in sampling error as changes of opinion.  An unavoidable problem is that having stated an opinion in the past may affect what people say (maybe they don't want to seem to be flip floppers or oppositely maybe they want to prove they are open minded and react to new evidence).

But there is no way no way that the second wave of a panel is going to be a representative sample.  Not gonna happen.

Trying to undo the damage by weighting would work if all White women age 25  to 40 with college degrees were exactly identical with the same opinions about everything.  They aren/t I've checked (the only almost universal opinion is "no thank you,  I would not like to engage in sexual intercourse with Robert Waldmann" (actually I really haven't asked that question of a large enough sample for reliable statistical inference (I am very shy in real life))).

I didn't watch the Townhall Presidential Debate

Actually I didn't watch the first 2012 debate either-- too much tension. This time much too much tension -- I didn't even watch my twitter stream.   The left blogosphere is delighted sure that Obama demolished Romney  In contrast the CBS instapoll showed a narrow 37 to 30 win.

I try to avoid making predictions but I will predict 2 things.  First CNN/ORC will have a panel poll in which they interviewed people before the debated and by now probably have interviewed the subset who agreed to be interviewed and actually picked up the phone and answered too.  Second more of the re-interviewed few (order of 3 to 4 hundred I predict) will say Romney won than will say Obama won.  I predict this because I am a fool and will never learn that all my predictions are wrong so I should keep my blogger window closed and be thought a fool rather than typing and removing all doubt.  Oh and also because the re-interviewed few are older than those who don't agree to be re-interviewed or can't be contacted or just change their mind about whether to bother answering the questions.  A panel is very useful, but the second wave is not likely to be a representative sample (and CNN/ORC can't weight to match the debate audience and get numbers out fast -- before not after Nielson gets their audience ratings out).

Update: my almost perfect record of wrong predictions (except I nailed the 2010 Senate election) is untainted.  CNN?ORC gets a narrow Obama win (and by the exactly exact same margin as CBS)

Forty-six percent of registered voters who tuned in to Tuesday night's town hall debate in Hempstead, N.Y. declared President Barack Obama the winner, according to a snap poll from CNN and ORC International.
The poll shows that 39 percent of voters who watched the second debate said Mitt Romney topped the president.  Obama's 7-point edge over Romney is within the poll's margin of error.

When will I ever learn to not make predictions in public ?  I'd type "never" but that would be a prediction made in public.
end update

OK the big debate event seems to have been the moderator Crowley saying Romney's claim of fact was false and Obama's true.  The argument was about when Obama publicly called the terrorist attack on our Benghazi consulate terrorism (correct answer -- the day after the attack when he first publicly spoke about it).  Here I think Romney's problem is that he wasn't lying.  He said he wanted the dispute of fact on the record because Obama first called it terrorism 2 weeks after the attack.

Clearly Romney must have believed this, otherwise he wouldn't have stressed the disagreement and made a specific claim.

Now this sure isn't like Ford saying Poland is free (I did watch that debate).   It absolutely provides no evidence that Romney is a liar.  This makes it unusual, since he usually at least dissembles and lies very often (much more often than Nixon if I correctly recall campaigns when I was 7 and 11 years old).

What went wrong with the #Romneystrength through deceit strategy ?  How did he let himself make a clear specific verifiable claim about recent events which captured the nations attention ?

I think partly the problem is the conservabubble eco chamber.  In Republimerica it is well known that Obama blamed the guy who made the crappy anti Mohamed film and claimed a long planned al Qaeda attack was second degree murder.  An alternate reality is useful -- it contains the fired up base which actually votes and maybe convinces friends and relatives.  But it is unwise for a candidate to live in the paranoiland as he has to venture out into the reality based reality to debate.

Another problem is falling for their own spin.  Republicans and the Romney campaign have been strategically claiming (aka lying) that there is something scandalous about the fact that Obama and the CIA didn't know all about the planning of the attack the instant it happened (frankly I personally guessed that it was a long planned attack and had nothing to do with the crappy film and little to do with violent protests in Egypt -- but  I didn't say that in public because uh search up for "remove all doubt").  This is an absurd dishonest argument, but uh method acting uh.  Even Romney gains by convincing himself of his lies.*  Arguing that not knowing everything instantly is a cover up is dangerous.  It is hard to keep a strategic (lying) distortion of reality strategic.  The Romney campaign convinced Romney of a falsehood by trying to cleverly mislead.

Oh also arrogance.  No not mine Romney's.  He was sure that he was as expert on what Obama said when as Obama was.  A lot of his success in debates (also in the primaries) is based on his projecting confidence.  It makes his claims more credible.  More importantly people want that in a president.  It is costly when it turns out that it is vanity and not justified confidence.  I think Romney has become to used to being in a position of power where his word is decisive even if he doesn't know what he is talking about.  I also think that Obama does not have this problem in spite of living in the White House for 3 years and 8 months.

So what's with Obama ?  How does he keep his ego under control ?  Add this to the long standing unanswered question of how the hell does he keep his pants on when hot babes are begging to screw him.  He's very confident and he is willing to listen and learn.  How did that happen ? Are there any other people like him ?

 *not much I think he can lie without any of the usual symptoms, that is I think he is a psychopath (talk about removing all doubt that I am a fool -- I am now making a psychological diagnosis based on what others saw on TV when I wasn't watching).

Tuesday, October 16, 2012

De Vries reconsiders views on genetics

The second person to discover the laws of genetics (after Mendel who was forgotten) Hugo De Vries tended to over estimate the rate of mutation.  The fictional character Pieter De Vries enjoyed inflicting pain.

Now William de Vries inflicts great pain on Mitt Romney by noting how different he is from his father.

Walter De Vries, who worked for the senior Mr. Romney throughout the 1960s, wrote that Mitt Romney’s bid for the White House was “a far cry from the kind of campaign and conduct, as a public servant, I saw during the seven years I worked in George Romney’s campaigns and served him as governor.” 
“While it seems that Mitt would say and do anything to close a deal – or an election,” he wrote, “George Romney’s strength as a politician and public officeholder was his ability and determination to develop and hold consistent policy positions over his life.” 

I am doing my best to  avoid sectarian bigotry and resist the temptation to suggest the Mitt try to posthumously convert his father to the church of later day flip floppers.

And failing.

DemoSpam reaching point of diminishing returns

I had already deleted most of my daily e-mails from Democrats asking for campaign contributions.  I don't reply that all candidates to whom I donate lose -- I'm a jinx an hex you don't want my bucks.

But then I saw the perfect case showing they are stepping on each others cyber feet

I can't help imagining them fighting over who gets to type on a PC at DNC headquarters -- a key battleground indeed.

Pollster Strategies

Some time ago I noted the marked Republican house effects of pollsters which had the words "marketing" or "strategies" in their name.  There is some legitimate evidence on this pattern from Missouri -- in the huffington Post pollster aggregate front page every single poll by pollsters whose name includes "strategies" has Akin ahead while every single other poll has McCaskill ahead (yes Rasmussen too). Wow. I'm not claiming that the "strategies" pollsters are deliberately slanting as part of their strategies.  I'm just insinuating real hard.

Monday, October 15, 2012

Is this the Slatiest Slate article ever ?

Sex Is Cheap

Why young men have the upper hand in bed, even when they're failing in life.

Key passage 

But just as critical is the fact that a significant number of young men are faring rather badly in life, and are thus skewing the dating pool. It's not that the overall gender ratio in this country is out of whack; it's that there's a growing imbalance between the number of successful young women and successful young men. As a result, in many of the places where young people typically meet—on college campuses, in religious congregations, in cities that draw large numbers of twentysomethings—women outnumber men by significant margins. (In one Manhattan ZIP code, for example, women account for 63 percent of 22-year-olds.)

Yes the alleged topic is young men who are "failing in life" yet "successful" is equated with "typical".  OK Mark let me give you a hint.  Typically young people are not in college.  Many never go (many never graduate from High School).  It doesn't last forever either (note typical post secondary education is not a completed 4 year BA plus maybe more).  

The young men who aren't univsersity students or yuppies just don't exist in the article even though it is supposed to be about them !  You don't read Slate, Slate won't report on how you mate.

OK OK I'm bitter because youth is wasted on the wrong lazy irresponsible males (I am the lazy irresponsible male who I care most about).   

Also the "in bed" in the title makes a promise which is not kept by the article.  The article is mostly about the claim that young men have the upper hand in relationships which basically amounts to noting that most relationships of young people don't involve marriage ceremonies.  Notably the possibility that women today don't want to commit when they are just as young as their grandmothers were when those grand*mothers wanted to commit isn't mentioned.  The possibility that more and more women continuing studies and almost all not only working but caring about their job prospects might maybe make them less eager to marry young is not even considered for one phrase let alone a full sentence.  It is just assumed that women everywhere and always want maximum commitment right now (am I allowed to claim it is sexist to assume that alllll women are obsessed with marriage ?).  More commitment oriented than contemporary young men and just as commitment oriented as young women were in the 1950s (back when my mom was turning down one wedding proposal after another).

But the "upper hand in bed" part is just "as my colleagues and I discovered in our interviews, striking numbers of young women are participating in unwanted sex—either particular acts they dislike or more frequent intercourse than they'd prefer or mimicking porn."  Well that was discrete.  I can't claim that Regnerus is mimicking porn in that sentence.  I will not admit that I find the passage titilating and am experiencing a strong prurient interest in the hard detailed data, but only because it is so obvious that I see no point in admitting it. 

Yes the thought of unwanted sex is horrible (we are not talking rape here but bargaining in which someone agrees to sex she doesn't want in exchange for ... what could be worth that).  I'm not 100% sure how men manage to have sex with someone who doesn't want it.  I would find it very hard --- uh no I mean very soft ("it" being my more sensitive 0.1% or, as @andrewrgoldman would say my "little  Freud")

* prefix added with extreme distress when I realize that the mothers were mostly too young to really get into the free love 60s. Mark Regnerus young man, if you make me feel this old again and I will write a really hostile blog post.

Obama Debate Performance Surpassed

He no longer has the record for the fastest fall

Felix Baumgartner broke the sound barrier ... making him the first skydiver to fall faster than the speed of sound

parachute not enough.  Fire up the Curiousity sky crane.

Simple answers to simple questions

@emilynussbaum @embeedub now I'm ADC and readers may be "onto something?" Doesn't that sound exactly like Stalinism?


Not so simple question. Can douchebaggery be raised to an art form ?

Se also "sprezzatura"

Comment on Singiser

Steve Singiser has a big post on likely voter polls vs registered voter polls up at dailykos (one of his favorite topics and one of mine).  I go wild with enthusiasm in comments.

I know you are busy, but I'd really like the exact words that the PPP machine says along with "unlikely" "vote" "hang" "up2 and "phone."

Given how everyone aggregates polls now, the magnitude of the error doesn't matter all that much.  I think the 50-50 favor Dems-Reps result suggests that it makes sense to look at likely voter polls.  But polls are aggregated to the average not the mean so what is the average LV poll error (note you don't need RV polls to calculate this).

LV minus RV Obama-Romney vs LV minus RV Mondale v Reagan.  Look a whole lot of the difference between voters and non voting adult is that voters are older.  I just can't doubt that the new huge LV-RV gap is due to the new huge age gradient of voting intentions with older people now much more likely to vote Republican (before a smaller difference and occasionally the elderly tilted Democratic).

LV filters might oversample the old compared to the young.  RV polls might over sample the young compared to the old.  It should be possible to compare LV filters to actual voting by looking at exit polls (maybe even actual data on votes compared to census data on the population).  This is a way in which things have changed.  It matters because pollsters have checked LV filters by looking at past forecasts minus outcomes.

One of the bees in my bonnet is the Gallup LV screen (total flop in 2010 after decades of working fine).  One question is "do you know where to go to vote".  The no answer has a very different significance on November 5  and on October 5 no ?  If it's Monday and someone claims they will go and vote on Tuesday but still doesn't know exactly where uh then maybe Gallup is right to take the claim with a grain of salt (3 of which grains gets one tossed out of the LV sample).  Another is have you voted there before which is certain to drop people who really will vote for the first time and also people who moved so strongly weeds out the young.  Notably the Gallup LV screen worked very well for the Congressional Generic ballot for decades then caused a total utterly humiliating forecast error of 9 (nine !) percent in 2010.

Average error of LV polls can be calculated for LV polls long before election day.  Yes there is the real shift in public opinion so one wants to average over many elections, but what we want to know *now* is whether polls released now are biased not whether the last poll before the election will be biased.

Sunday, October 14, 2012


Amazingly Massachusetts allowed me to vote by e-mail !!! I had to waive my right to a secret ballot.  I still can't show you my ballot.   Actually I swore I hadn't done so (the idea is that I am especially not allowed to show it to someone in exchange for money for having marked the ellipse he or she likes).   But wait, I still have my ballot -- I scanned it and sent the scan.  I'm not sure the law forbids me from showing someone my ballot after voting by e-mail (or fax but I hate faxing and that's a fact).  Not being sure it is allowed, I will just upload the very official electoraltronic transmission sheet.

Amazing no ?

Also just to go back to one of the many bees in my bonnet, what about the Gallup likely voter filter ?
It worked very well for decades then totally messed up in 2010.  Gallup began reporting results for two filters .. the traditional filter and one which counted more respondents as likely voters (if I understand it basically anyone who said they were going to vote was counted as a likely voter by the looser ad hoc invented during the campaign filter).  The traditional filter  congressional generic ballot poll was a disaster overestimating the Republican margin by 9%.  This was definitely one of the types of polls of which Gallup had been most proud as it had an amazing record of accuracy.

WTF happened ?  Basically I would guess that the filter has always lead to oversampling old people vs young. Back in the good old days when I was young, old people voted for Democrats about as much as young people did (and more than middle aged people did).  In 2010 not so much, so maybe a long lasting problem which didn't mess up the results in the past came home to roost and keep government hands off of its Medicare.

But I want to get deeper in the weeds.  The traditional Gallup likely voter filter consists of 7 questions with yeses required to get in the likely voter sample.  One question is IIRC "have you ever voted in your current polling place before?" and another is something like "do you know where it is?".  This last question obsessed me in 2010 as I guessed that it caused a bias which vanished by election day.  If someone doesn't know where to vote and it is election day their claims that they will certainly vote, are very exited and have been following the campaign must be taken with a grain of salt.  In early October not knowing exactly where to vote mostly means you haven't voted there before -- so people who have recently moved get two almost automatic no's as do people who will, in fact, vote for the first time.

The fact that the bias increased as election day approached meant my face shared the egg which was all over Gallups corporate visage.

But I am stubborn so I have a new theory.  The new theory is that people who will in fact vote count on being able to find out where on the day using google and google maps.  So I decided to see how long it would take me to find out where to vote in person were I in the USA.  It took me three whole minutes to find out that I would vote at 28 Sacramento St and get a google map showing me where that is.

I am quite sure that the last time I voted in person (same voting address as now) I went somewhere else.

So look it could be that someone plans to vote during their lunch break, has no clear idea where at 11:30 and actually votes.  I don't think it used to be that way.

Gallup is going to have to gallup much faster to keep up with all this newfangled technology.

Ruder comment on Portes

Comment on Portes banished here for excessive rudeness

Your definition of "the multiplier" is misleading.  

Originally "the multiplier" referred to the effect of an increase in autonomous (that is note really modeled) demand on output, but my point is not quite that pedantic.  It isn't just that you mean "the fiscal multiplier" when you type "the multiplier."  There are three fiscal multipliers in the IS-LM model, new Keynesian models and in reality. The effects on output of a government spending cut and of a lump sum tax increase are not the same in any existing theory or in the data. The difference is the balanced budget multiplier which is not zero in any existing model in which the deficit spending multiplier is not zero. 

By your definition, the balanced budget multiplier is meaningless as it is something divided by zero. However there is strong evidence that it is a positive number (well not super strong the direct evidence is the apparent effect of the tax financed war effort in Korea on US output, but different estimates of spending and tax cut multipliers are indirect evidence).  If your definition were the definition of the multiplier then Robert Lucas would have been right when he claimed that if infrastructure investment is financed by taxes then there is nothing for the multiplier to multiply. In fact he managed one of the few errors so extreme that it could cause me to fail a student in my introductory macro course (I am a notorious softy). This post contains that error.

The ratio of output increase to reduced taxes minus transfers is a second multiplier. Finally there is the ratio of output increase to deficit financed government consumption and investment which is the sum of the other two multipliers.

The claim that *the* multiplier multiplies the change in the budget deficit is a gross error in elementary macroeconomics. 

Of course you are not ignorant enough to believe that there is one and only one fiscal multiplier, but you are careless enough to help the ignorant remain ignorant. I think you should have left well enough alone -- it is better to provide no definition for your terms than to provide a grossly incorrect and utterly misleading definition which helps maintain widespread ignorance.

update:  I have a second objection

I have another objection.  You claim without making the case that Spain needs "structural reform, especially labour market reform."

My impression is that Spain enacted quite radical labour market reform starting decades ago.  In particular the extremely rigid Franco era firing restrictions were relaxed for newly hired workers by the Suarez government and then still more by the Gonzales government.  This reform in Spain was enacted long long before the Shroeder reforms in Germany.

The timing is very important, since all the reforms preserved employment protection for those employed at the time of the reform (IIRC).  Spain, unlike Germany, had a huge number of workers without strong employment security -- workers who had been hired since the reforms and not shifted to the protected employment relationship.

This made the extremely rapid increase in Spanish unemployment possible. 

One might also argue that what Spain needs is deregulation of banking so that foreign banks can open offices in Spain and so the international flow of capital is not hindered by the highly regulated protected Spanish banking system.  I don't think that you are one who would make that argument (it is a parody -- I've been in Spain and seen branches of UK based banks all over).  However, I honestly don't see why the argument for labour market deregulation is so much stronger than the argument that Spain wouldn't be in such trouble if only they had been able to attract foreign financial investment (again this is a parody-- I am not suggesting that you (or indeed any sentient being) would make that argument).

Why do you think that Spain needs further labour market reforms ?  It is an important question, yet you seem to consider the answer completely obvious ?

Saturday, October 13, 2012

Polls and Reporting: nitpicking meets grasping at straws

All aggregators are doing funny things by rounding first then subtracting.  This is a comment on Singiser, but it could be on any aggregator.

In the Gallup Tracker likely voter subsample, Romney is up 1.2%  (this link points to numbers which will change)  If one must round one should round to 1 not 2.

The standard report rounds Romney's lead up to 2 %
"NATIONAL (Gallup Tracking): Romney 49, Obama 47 (LV); "

In the IBD/Tipp tracker likely voter sample Obama is up 0.7.  If one must round one should round to 1 not zero.

IBD/TIPP 2012 Presidential Election
Daily Tracking Poll
Day 4: Oct. 12, 2012
Obama: +0.7Obama 46.4% | Romney 45.7%

Read More At IBD:

The standard report rounds Obama's lead down to 0%
"NATIONAL (IBD/TIPP Tracking): Obama 46, Romney 46"

The two average Romney up 0.25% not 1%

I often read "percents don't add to 100 because of rounding".  But somehow this is not OK when subtracting rather than rounding.

update: Singiser notes that (with the standard silly rounding) Romney's national poll lead for the day declined from 1.3% to 1%.  He notes that this is totally statistically insignificant and not worth wasting thousands of pixels to criticize as I do below.

It matters to people silly enough to try to back out daily results from trackers.  To be maximally silly I will use the rounded numbers released yesterday (actually day before yesterday here in Rome).

Gallup 10/11 minus Gallup 10/5 (silly rounded) comes up as a shocking Romney up by 1.4% (shockng as 10/5 should have been a great day for Romney) not an even more shocking 7%.

IBD/TIPP 10/11 minus IBS/TIPP 10/6 (silly rounded) comes out as Obama up 10.2 not 6.

The estimate for a normal day vs the two dread days after the debate is 4.4%  with half silly rounding not -0.5% with all silly rounding.  With silly rounding (both days for both pollsters) it looks as if the debate effect was permanent not a bounce. With silly rounding for the second latest day and not the latest day, it looks as if the debate effect was temporary.  This is all silly, because it is torturing the poor numbers to try to make them say something they don't know, but the point estimates are not small.

Friday, October 12, 2012

Longer Joe Biden

Biden “Let me have a chance to translate,” Biden said. “I was there with Ronald Reagan. He gave specifics in terms of tax expenditures.”

Bentsen: Senator, I served with Jack Kennedy, I knew Jack Kennedy, Jack Kennedy was a friend of mine. Senator, you're no Jack Kennedy.

Longer Biden: Congressman, I served with Jack Kennedy, I knew Jack Kennedy, Jack Kennedy was a friend of Tip, Congressma, you're no Ronald Reagan.


Actually speaking of Ronald Reagan, do you remember the second time he said "There you go again." in a debate ?  It was 1984.  It worked so well in 1980 v Carter that he said it again to Mondale.  Mondale smiled and pointed out that Reagan had said to to Carter when Carter claimed Reagan was going to propose Social Security cuts.  Then he smiled wider and noted that Reagan had cut Social Security.

Didn't someone tell Ryan he shouldn't compare himself to a beloved historical figure when debating Joe Biden who was there (I mean in DC not Kentucky) at the time ?

Omigod.  I honestly didnt' read the transcript down to the megawonderful part

“Jack Kennedy — lower tax rates increased growth,” Ryan said.
“Oh, now you’re Jack Kennedy?” Biden said.
Joe Joe Joe, he was asking for it. And at least you have to note that Kennedy signed a bill establishing a top marginal income tax rate of 70% and the economy boomed.   How about asking Ryan if he likes the Kennedy era tax code so much that he wants to re-enact it ?  

OK OK Biden did very well.

Wednesday, October 10, 2012

Return of the Gallup Likely Voter Filter

Just when you thought it was safe to get back in the sample pool, Gallup switched on their likely voter filter.

The results and that the headline result isn't the registered voters subsample of the tracking poll showing Obama ahead by 3% but the likely voter subsample showing Romney ahead by 2% and my loyal readers (if any) get the latest entry of the continuing saga Robert dumps on the Gallup likely voter filter.

First note that in 2010 Gallup's likely voter poll forecast Republican share minus Democratic share 9 points greater than the outcome ?  Second note that Gallup knew something was wrong and broke with tradition by also reporting a broader likely voter sample using a filter based only on stated voting intentions.

Second note that I slipped and made a prediction in 2010 (I was wrong as usual).  My problem with the traditional Gallup likely voter filter (described below) is that I think that well before the election it introduces bias which goes away as election day approaches.  So I predicted that Gallup R-D minus other pollsters R-D would decline as election day 2010 approached. Oooops.  The opposite happened.  But the filter just didn't work.

I actually checked to see if Gallup is using their traditional filter -- they are.

The calculation of likely voters is based on registered voters' responses to a seven-question series that -- with some revisions along the way -- Gallup has used since 1952 to calculate voters' likelihood of voting. In some years, such as 2008, there was only a marginal difference between the vote choices of registered voters and likely voters. In others, such as 1996, there was a much more substantial difference.

At this point, Romney voters are somewhat more likely to respond that they will definitely vote, that they have thought a lot about the election, and that they are more familiar with where people in their local area vote. These attitudes indicate that Romney at this juncture will benefit from higher turnout on Election Day among his supporters than will Obama. These patterns could change closer to Election Day as more voters become engaged or if Republicans' or Democrats' enthusiasm for voting is altered by campaign events.

Note the amusingly unusual grammar in my italics , but the bold. is the part that matters.  One of Gallup's 7 questions (5 yes answers are required to be considered a likely voter) is (something like) do you know where to vote ?  Clearly a no answer means something different in early October and early November.  If the election is the day after tomorrow people who say they will go to, vote but don't know where, are not very likely to actually vote (has happened to me but I'm flaky).  The same ignorance in early October is less of a signal of unlikeliness to vote.

The problem is that Gallup has evaluated the filter for decades mostly based on the gap between the last poll and the election.  I don't know if they would notice a bias which fades (especially if covered by true drift in public opinions towards Republicans which, I think, was typical for many of those elections as conservative Democrats reluctantly voted Republican).

As predicted by @Austen_Goolsbee (great economist but if you cut people's taxes don't try to prevent them from noticing) this was reported as a shift to Romney.

Obama would have been 0.15% ahead.

Monday, October 08, 2012

Historical Irony Alert

The Wall Street Journal (yes that Wall Street Journal) reports on Maoists challenging the ruthless exploiters of the working class who run the Communist Party of the People's Republic of China.  But that's not the ironic part.  The ironic part is that to learn about the Chinese New Left Wall Street Journal chief new left ideology reporter Brian Spengele talked to someone in Utah !?!?

Mr. Bo's takedown indicates consensus among Chinese leaders to push ahead with economic liberalization despite growing social problems, said Minqi Li, an economics professor at the University of Utah, who is aligned with new leftists in calling for more-egalitarian wealth distribution in China.
 I know that Utah is a red state but I didn't imagine that referred to the little red book.

History has long been a prankster, but used to have some respect for credibility.

OK seriously the People's Republic does have an unequal income distribution.  The Communist Party ruthlessly exploits workers on land ruthlessly expropriated from peasants and I wish professor Minqi Li all the best.

Sunday, October 07, 2012

Skewed CNN/ORC debate instapoll

Oh hell.  I want to put on record that I aim to try to calculated a Chi squared goodness of fit test for the CNN/ORC instant poll on the presidential debates based only on under 50/ over 50.

The problem is that the population is adult americans who watched the debate.  I don't know how to get data on who watched the debate, so I might not be able to do this even if I have the CNN/ORC internals.

Saturday, October 06, 2012

Morning in America ?

Bill McBride at Calculated Risk notes that the employment to prime age (25-54)  population ratio has increased since the trough (unlike the employment to working age (16-5) population ratio.  He gets a link for saying I am still prime age (if not for long).

I steal part of his graph (fair use) to note that the employment to prime age population ratio is now just about identical to what it was when Ronald Reagan was overwhelmingly re-elected in large part because it was morning in America (the unemployment rate was 0.5% lower then)

See also Krugman for some highly informative crunching of about three numbers (really 6 time series plus three sophisticated tools not available to most pundits -- multiplication, addition and division.  Paul Krugman is losing his patients with you guys and, if you don't pay attention, will use subtraction (with extreme prejudice)).

Thursday, October 04, 2012

Fact Checking Mitt Romney

Strictly subjunctive speaking: The CNN instant fact check seems similar to what it would be if they were in the tank for Romney.

Yes and if you take him at his word, our verdict on whether Richard Nixon is a crook, the verdict here is false.

But this guy said that Romney's claim to have a plan to guarantee coverage of pre-existing conditions was a flat out lie.  

Fehrnstrom said those who currently lack coverage because they have pre-existing conditions would need their states to implement their own laws — like Romney’s own Massachusetts health care law — that ban insurance company from discriminating against sick people.

Romney has to find some way to neutralize Fehrenstrom who was tougher on him that CNN ... and Obama were.

As often, I really suspect that Fehrenstrom has to hope that Romney doesn't keep his campaign staffing chart on his Etch a Sketch.  Also is he a Democratic mole ?

Wednesday, October 03, 2012

Daily Wanker Robert Stein

Brad shares his twitter stream. An excerpt

J. Bradford DeLong ‏@delong: @BobStein_FT Dems could defeat filibuster 7/7-8/1/09, 11/9/09, and 9/25/09-2/4/09. Did an awful lot of Senate business in those <5 mos="mos" p="p">
Robert Stein ‏@BobStein_FT: @delong Wrong. You're only counting Dems. Specter was a pro-O vote even before he switched. No reason not to count him.
J. Bradford DeLong ‏@delong: @BobStein_FT Don't start tweets "wrong"--especially when you r wrong. Dems had 58. Specter gives them 59. Franken seated on 7/7 makes 60
Robert Stein ‏@BobStein_FT: @delong it's not the length of time they're formally seated...u can anticipate & do legis spade work and wait for formal votes when they r.

So first Stein starts a contribution to a discussion of a "length of time" with "wrong".  Then when it is pointed out that his argument is totally wrong and based on shocking ignorance, he declares the topic irrelevant. 
Also Stein considers it obvious that Coleman couldn't possibly win an election although he had more votes on the initial count and after the recanvasing.  Or rather the time when it became clear that Franken would eventually be seated is irrelevant because ... shut up.
It gets worse

Robert Stein @BobStein_FT: @delong You are a fool. That rule itself can be repealed w/ 50 votes. Talk to someone who understands how Congress works.
Now Stein has resorted to noting that when newly seated the Senate doesn't have to adopt the old rules and can adopt brand new rules.  But wait the original topic was the importance of having at least 60 seats in the Senate.  When a new Senate convenes 50 senators and the vice President can declare that 50 votes are enough for cloture and that 30 seconds not 30 hours of floor time are required for cloture motions to mature.  In fact they could declare that one vote is enough for cloture if they wanted. 

The change a rule with 50 votes argument is not a discussion of the rules of the Senate.  It is a declaration that they are irrelevant provided 50 Senators and the vice President decide to make them irrelevant on the first day of a new Senate.  
What a wanker.
I think tht the reaction of someone who should be expert and displays his ignorance by changing the subject and insulting the person who noted his ignorance has earned a response beyond Brad's and mine.