April 8, 2011

Nate Silver does the math.

"There is no evidence that Waukesha County’s revised vote count is unusually high, whereas there is some evidence that its original vote count was unusually low."

Massive detail at the NYT link.

33 comments:

PaulV said...

Can the left trust even a known liberal?

edutcher said...

Waiting for one of the Lefties to intone, "Figures lie and liars figure".

PS Love this tagline from PowerLine, "Whoever wins the judicial election, I think we can at this point declare Althouse a winner".

Yes, Mr deMille, the lady is definitely ready for her closeup.

Maguro said...

OMG, the Koch Brothers have gotten to Nate Silver, too. End times must be near.

chickelit said...

Waukersha County trumps Dane County!

Richard Dolan said...

Regression analyses never fail to impress, but in this case it is a lot of statistical sophistication serving no purpose. When the vice-chair of the county Dem party in Waukesha took the microphone at the press conference yesterday and said that she was completely satisfied that the numbers and explanation were correct, the 'stole the election' meme died.

It was interesting that all of the players in Waukesha were middle aged (or above) women. They may have their partisan loyalties but they also knew that those have to take second place to their public duties. To those who expect the worst from political actors (and are rarely disappointed), that display of plain-spoken midwestern rectitude was quite impressive. It could well serve as a model for the would-be post-partisans in DC.

MadisonMan said...

I see a lot of complaints that the number of votes is just enough that an automatic recount is unnecessary, as if that's some kind of weird thing.

Sometimes, coincidences to happen.

MadisonMan said...

ugh. DO happen, not to happen.

Quayle said...

All those erudite New York Times' readers are calling it fraud-like.

gadfly said...

Looks Like Nate Silver spent a whole bunch of time analyzing nothing. The "incompetence" of Waukesha County came about because the AP was not notified of Brookfield's vote count.

The proper number was immediately posted online by the county. So when does the AP become an official party in an election and if they are, how come they haven't been called out to defend why they failed to verify the numbers that they received electronically?

The Drill SGT said...

MadisonMan said...
Sometimes, coincidences to happen.


I don't know about you science types, but us beady eyed killers have a hypothesis for that.

once is happenstance
twice is coincidence
three times is enemy action :)

or the shorter, Marine version:

Shit happens

PaulV said...

Since the town vote count was posted online at 12:30 am the night of the election any coincidence was that the rest of the state was 50/50. The 7500 votes were a known known.

Scrutineer said...

MadisonMan - I see a lot of complaints that the number of votes is just enough that an automatic recount is unnecessary, as if that's some kind of weird thing.

That vote total was reported by the city of Brookfield on the night of the election. It hasn't changed since then. The county clerk just hadn't included it in her numbers.

http://brookfield-wi.patch.com/articles/additional-votes-found-in-waukesha-county-canvassing

miller said...

Oh, the meme hasn't died - it's growing. Sadly.

I'm not convinced that mankind is rational.

WV: Flaterei, the Valkyrie of effusive praise.

miller said...

Sometimes human error is simply error and not maliciousness.

Brian O'Connell said...

A lot of the suspicious commentary on this topic seems completely unnecessary (well either that or willful).

This is no murky question on whether certain ballots were counted or not. As far as we know, all ballots were counted- up to the usual standard anyway. At issue here is the much easier to verify question of whether all the local vote totals were included in the overall total. Surely that's something that can be checked in 5 minutes- or hell, an hour if the software is especially annoying.

Why would we need Nate Silver's statistical analysis? It might be useful as a double-check, but surely it wouldn't be difficult to verify that the revised totals are actually the sum of the local ones, and that the old totals were not?

Fprawl said...

Media outlets want the numbers right away so they can scoop the competition.
In our town, churches provide poll watchers to phone in results to the 3 stations in exchange for commercials.
Sounds like she was collating numbers for that unofficial duty, and screwed up.
I like graphs, but a recount of that county and certification by a Democratic Secretary of State will go a long way to restoring sanity.

May said...

The one thing that strikes me about Nate's post was the demand for the clerk to resign. That seems harsh for the error of sending the wrong version of a file to a news organization. The error didn't effect the actual election, it just got the wrong set of partisans elated and despondent.

I make simple errors like this all the time, and yet I'm still employed.

Bob Morant said...

All the conspiracy theorists posting on Madison.com seem to overlook what the numbers would have looked like were this error not detected. They all seem to think it would be just dandy if the City of Brookfield failed to cast a single vote. Since it's a conservative county, this would not raise a red flag at all for them.

miller said...

More to the point, the city itself reported the totals and they were available online at the time (and even now).

The error was not in the way the votes were stored or saved or sent.

The error was in a hand-tally application.

Go ahead and pay for a recount.

The people who are squabbling about this are so stupid and partisan I think my despair meter is at 11.

rhhardin said...

It is hard to excuse the mistake

It's a software mistake in the user interface, failing to warn about data imported but not saved.

C.T. said...

Rhiel World View links to an AOL site called Madison Patch which actually published the Brookfield numbers on election night.

http://www.riehlworldview.com/carnivorous_conservative/2011/04/brookfield-city-results-were-reported-accurately-on-election-night.html

"It turns out the now not so mysterious big bump of over 7,000 votes for Prosser in the Wisconsin Supreme Court race were accounted for all along.

A contemporaneous local media report on election night, linked further down, proves it. The AP might not have gotten them, but the media, in one form, or another, most assuredly did.

Here they are apparently directly from Schmidt in Brookfield in all their glory, as reported, more or less, at the time by local media."

http://brookfield-wi.patch.com/articles/brookfield-gives-prosser-nearly-11k-votes

Brookfield Gives Prosser Nearly 11K Votes

Incumbent State Supreme Court Justice David Prosser gets 10,859 votes from city residents or 76 percent against JoAnne Kloppenburg.
By Lisa Sink | Email the author | April 6, 2011

Brookfield Gives Prosser Nearly 11K Votes

"As expected, Brookfield city voters ran up a good turnout in the state Supreme Court race and gave incumbent Justice David Prosser nearly 11,000 votes. Unofficial, unaudited results showed 76 percent of city residents who voted picked Prosser, with 24 percent voting for challenger JoAnne Kloppenburg."

Alex said...

ugh. DO happen, not to happen.

You mean shit happens to you guys.

Fred Drinkwater said...

Holographic ballots ( or similar, voter-only identifiable physical objects). An auditable trail.

Hell's bells. The frenzied rush to all-electronic voting is driven only by the desire of the media to have instant results (or instant counting problems, that can sell papers). A mature polis would not want that; a mature polis would want correct results in a reasonable time. Too bad we don't have one.

Someone in a previous thread remarked on how they had hoped vote counting was handled by adults. A forlorn hope. All political process is driven by the base desires of the political class, no matter what they say in public.

AJ Lynch said...

I was impressed that Althouse, a onetime lefty, was so adept at math and formulating estimates and trends and predictive ratios etc.

Most lefties are bad at that stuff.

Tom said...

May we assume that every time the NYT publishes a correction, they fired the person responsible for the mistake.

Kirby Olson said...

I wonder if anybody else remembers the Garnett Mimms hit, Cry Cry Baby, that briefly lit the charts in about what 1964? Janis Joplin later covered it, but Garnett Mimms had such a wonderful range in his voice. Mimms is still alive, but I bet he can't still sing like he did then. What an athletic, operatic voice for lovely rock. We should turn it on for all our Democratic friends who have come home crying that they lost again:

http://www.youtube.com/watch?v=47OXAqbv4cU

Gene said...

Nate Silver says that when Kathy Nicholaus held a press conference to announce that her initial vote tally was wrong, she might have been "expected" to resign.

Why?

Because Nate Silver's wildest hopes for a Kloppenburg victory were raised and then cruelly dashed?

Sliver is a reporter. What's he doing becoming so emotionally invested in a Democratic victory anyway?

PackerBronco said...

Silver spent 2 hours running all sorts of calculations when all he needed to examine what the simple question: "Is it remotely reasonably the city of Brookfield didn't cast a single vote in the election?"

Uh, no Nate, it's not. And all those graphs you produced showed that you have too much time on your hands and just wanted to play with your statistics software.

Talk about missing the whole blooming forest for the trees!!

Fen said...

Media outlets want the numbers right away so they can inform Democrats how many votes they need to steal

/fixed

Lisa said...

The woman who committed this 'human error' has done this at least three times before... always to the benefit of Republicans. She was involved in a significant scandal in which many people went to jail. She worked for the Republican candidate.

How can anyone trust her with this kind of record and this kind of partisan interest?

PaulV said...

Bitter much Lisa? Have you no concept of presumption of innocence? Small minded people like you are scary. Why blame a person for a computer error even though your hopes were wrong?

kent said...

Sliver is a reporter. What's he doing becoming so emotionally invested in a Democratic victory anyway?

First time reading Silver, I take it...? ;)

Fen said...

How can anyone trust her with this kind of record and this kind of partisan interest?

Right, because giving bad information to the AP is a hanging offense. Thats all that happened. The votes were counted at the county level, the accurate totals were sent to the Secretary of State.