Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Psychohistory and Nate Silver

30 views
Skip to first unread message

Philip Chee

unread,
Nov 7, 2012, 12:12:40 PM11/7/12
to
<http://www.forbes.com/sites/johnmcquaid/2012/11/07/three-lessons-from-the-nate-silver-controversy/>

[[The silly debate over Silver’s model showed a political, pundit and
media class at a classic horse-and-buggy meets internal combustion
engine (or ink-stained newspaper curmudgeons meet Internet, or, to get
really on-the-nose, Moneyball) moment: ignorance and fear masked by loud
harumphing. It seems like an amusing footnote now, but only days ago
Peggy Noonan was seriously arguing that a mild proliferation of Romney
yard signs she observed in liberal Northwest Washington, plus magical
vibrations of true Americanness emanating from all over, were signs of
imminent Romney victory. Other Silver critics seemed not to understand
basic statistical concepts. Others tried to change the subject. Some
political reporters simply ignored the existence of polling averages
altogether, pumping up the uncertainties of divergent poll results into
epistemological voids]]

Phil

--
Philip Chee <phi...@aleytys.pc.my>, <phili...@gmail.com>
http://flashblock.mozdev.org/ http://xsidebar.mozdev.org
Guard us from the she-wolf and the wolf, and guard us from the thief,
oh Night, and so be good for us to pass.

David V. Loewe, Jr

unread,
Nov 8, 2012, 11:05:23 AM11/8/12
to
On Thu, 08 Nov 2012 01:12:40, Philip Chee <phi...@aleytys.pc.my> wrote:

><http://www.forbes.com/sites/johnmcquaid/2012/11/07/three-lessons-from-the-nate-silver-controversy/>
>
>[[The silly debate over Silver’s model showed a political, pundit and
>media class at a classic horse-and-buggy meets internal combustion
>engine (or ink-stained newspaper curmudgeons meet Internet, or, to get
>really on-the-nose, Moneyball) moment: ignorance and fear masked by loud
>harumphing. It seems like an amusing footnote now, but only days ago
>Peggy Noonan was seriously arguing that a mild proliferation of Romney
>yard signs she observed in liberal Northwest Washington, plus magical
>vibrations of true Americanness emanating from all over, were signs of
>imminent Romney victory. Other Silver critics seemed not to understand
>basic statistical concepts. Others tried to change the subject. Some
>political reporters simply ignored the existence of polling averages
>altogether, pumping up the uncertainties of divergent poll results into
>epistemological voids]]

(Not having read the article, but sallying forth anyway...)

A lot of the criticism of Silver revolved around two things.

1) Unlike PECOTA, the formula isn't out there, so it can't be rigorously
examined by other people who are the political equivalent of
sabermetricians (and several sabermetricians did critiques of Silver in
the final days).

2) The disconnect between the polls - especially between the state polls
and the national polls.

Of further interest is the increasing lack of response to the polls.
--
"Short human words were like trying to lift water with a knife."
- Robert A. Heinlein

David Goldfarb

unread,
Nov 8, 2012, 9:00:28 PM11/8/12
to
In article <phln98pm5e7kfscca...@4ax.com>,
David V. Loewe, Jr <dave...@charter.net> wrote:
>A lot of the criticism of Silver revolved around two things.
>
>1) Unlike PECOTA, the formula isn't out there, so it can't be rigorously
>examined by other people who are the political equivalent of
>sabermetricians (and several sabermetricians did critiques of Silver in
>the final days).
>
>2) The disconnect between the polls - especially between the state polls
>and the national polls.
>
>Of further interest is the increasing lack of response to the polls.

And yet Silver's projections appear to have been extremely accurate.

--
David Goldfarb |"Why are you persecuting me? What have I ever
goldf...@gmail.com | done to you?"
gold...@ocf.berkeley.edu | "You tried to kill me, and destroy this
| entire planet."
|"Apart from that!" -- Doctor Who, "Boom Town"

Dorothy J Heydt

unread,
Nov 9, 2012, 10:26:32 AM11/9/12
to
In article <MD76w...@kithrup.com>,
David Goldfarb <goldf...@gmail.com> wrote:
>In article <phln98pm5e7kfscca...@4ax.com>,
>David V. Loewe, Jr <dave...@charter.net> wrote:
>>A lot of the criticism of Silver revolved around two things.
>>
>>1) Unlike PECOTA, the formula isn't out there, so it can't be rigorously
>>examined by other people who are the political equivalent of
>>sabermetricians (and several sabermetricians did critiques of Silver in
>>the final days).
>>
>>2) The disconnect between the polls - especially between the state polls
>>and the national polls.
>>
>>Of further interest is the increasing lack of response to the polls.
>
>And yet Silver's projections appear to have been extremely accurate.

So he must be doing something right, yes?

I had to google "sabermetrician" and came up with a lot of
references to *baseball.* Still, I suppose a technique developed
to analyze one topic can be applied to another. I guess.

--
Dorothy J. Heydt
Vallejo, California
djheydt at gmail dot com
Should you wish to email me, you'd better use the gmail edress.
Kithrup's all spammy and hotmail's been hacked.

David Loewe, Jr.

unread,
Nov 9, 2012, 10:35:19 AM11/9/12
to
On Fri, 9 Nov 2012 02:00:28, gold...@ocf.berkeley.edu (David Goldfarb)
wrote:

>David V. Loewe, Jr <dave...@charter.net> wrote:

>>A lot of the criticism of Silver revolved around two things.
>>
>>1) Unlike PECOTA, the formula isn't out there, so it can't be rigorously
>>examined by other people who are the political equivalent of
>>sabermetricians (and several sabermetricians did critiques of Silver in
>>the final days).
>>
>>2) The disconnect between the polls - especially between the state polls
>>and the national polls.
>>
>>Of further interest is the increasing lack of response to the polls.
>
>And yet Silver's projections appear to have been extremely accurate.

Well, on the second point, he was correct. When there are two options
and you choose the right one, you look a lot better.

As I said, these were all concerns *beforehand*.

And remember this, his models aren't perfect predictors (as he would be
the first to point out). He gave Sharron Angle a 75% chance of beating
Harry Reid in 2010.
--
"Go ahead and hate your neighbor, go ahead and cheat a friend
Do it in the name of heaven and you can justify it in the end
There won't be any trumpets blowing come the Judgement Day
And on the bloody morning after....One Tin Soldier rides away."
One Tin Soldier - The Legend Of Billy Jack

D. Glenn Arthur Jr.

unread,
Nov 9, 2012, 7:41:18 PM11/9/12
to
In article <MD888...@kithrup.com>,
Dorothy J Heydt <djh...@kithrup.com> wrote:
>I had to google "sabermetrician" and came up with a lot of
>references to *baseball.* Still, I suppose a technique developed
>to analyze one topic can be applied to another. I guess.

If I remember correctly, Silver himself _was_ a baseball satistician,
who had the thought, "a technique developed to analyze one topic can
be applied to another," and started applying sports-statistics
technique to politic. So it's no accident that folks keep making
comparisons to Sabermetrics or that you found mostly references to
baseball. AFAICT, many of his naysayers don't want to believe that
such a tool can be reapplied this way.
--
D. Glenn Arthur Jr./The Human Vibrator, dgl...@panix.com
Due to hand/wrist problems my newsreading time varies so I may miss followups.
"Being a _man_ means knowing that one has a choice not to act like a 'man'."
http://www.dglenn.org/ http://dglenn.dreamwidth.org

David Loewe, Jr.

unread,
Nov 9, 2012, 8:06:44 PM11/9/12
to
On Fri, 9 Nov 2012 15:26:32, djh...@kithrup.com (Dorothy J Heydt)
wrote:

>David Goldfarb <goldf...@gmail.com> wrote:
>>David V. Loewe, Jr <dave...@charter.net> wrote:

>>>A lot of the criticism of Silver revolved around two things.
>>>
>>>1) Unlike PECOTA, the formula isn't out there, so it can't be rigorously
>>>examined by other people who are the political equivalent of
>>>sabermetricians (and several sabermetricians did critiques of Silver in
>>>the final days).
>>>
>>>2) The disconnect between the polls - especially between the state polls
>>>and the national polls.
>>>
>>>Of further interest is the increasing lack of response to the polls.
>>
>>And yet Silver's projections appear to have been extremely accurate.
>
>So he must be doing something right, yes?
>
>I had to google "sabermetrician" and came up with a lot of
>references to *baseball.* Still, I suppose a technique developed
>to analyze one topic can be applied to another. I guess.

Silver was the person who formulated a respected baseball statistical
predictive tool known as PECOTA.

<http://en.wikipedia.org/wiki/PECOTA>

It has a good record, but has had some misses as well - most notably
Matt Wieters' rookie season (thought to be caused by overvaluing the
strength of certain minor leagues Wieters played in the year before -
which is why there were questions about which polls Silver was weighting
as more valuable).
--
"Small wonder that human beings in general disgust me."
Fred J. McCall in <qoil6vog8pbc1m252...@4ax.com>

David V. Loewe, Jr

unread,
Nov 9, 2012, 10:22:02 PM11/9/12
to
On Sat, 10 Nov 2012 00:41:18, dgl...@panix.com (D. Glenn Arthur Jr.)
wrote:

>Dorothy J Heydt <djh...@kithrup.com> wrote:

>>I had to google "sabermetrician" and came up with a lot of
>>references to *baseball.* Still, I suppose a technique developed
>>to analyze one topic can be applied to another. I guess.
>
>If I remember correctly, Silver himself _was_ a baseball satistician,
>who had the thought, "a technique developed to analyze one topic can
>be applied to another," and started applying sports-statistics
>technique to politic. So it's no accident that folks keep making
>comparisons to Sabermetrics or that you found mostly references to
>baseball. AFAICT, many of his naysayers don't want to believe that
>such a tool can be reapplied this way.

And some of them are fellow sabermetricians - as I pointed out in my
first post to this thread.
--
"You don't win a war by dying for your country, you win a war by
making the other poor bastard die for his country."
- George Smith Patton, Jr.

David Goldfarb

unread,
Nov 10, 2012, 2:28:16 AM11/10/12
to
In article <mf8q9816juimt0m44...@4ax.com>,
David Loewe, Jr. <dlo...@mindspring.com> wrote:
>>And yet Silver's projections appear to have been extremely accurate.
>
>Well, on the second point, he was correct. When there are two options
>and you choose the right one, you look a lot better.

He got a lot more right than just "who will win" -- he correctly called
every single one of fifty states, right down to "Florida will be very
close but will more likely go Democratic".

--
David Goldfarb |"Oh, no! They've all become giant Swiss
goldf...@gmail.com | lederhosen-clad dancing yodelers!"
gold...@ocf.berkeley.edu | -- Animaniacs

David V. Loewe, Jr

unread,
Nov 10, 2012, 9:03:07 AM11/10/12
to
On Sat, 10 Nov 2012 07:28:16, gold...@ocf.berkeley.edu (David Goldfarb)
wrote:

>David Loewe, Jr. <dlo...@mindspring.com> wrote:

>>>And yet Silver's projections appear to have been extremely accurate.
>>
>>Well, on the second point, he was correct. When there are two options
>>and you choose the right one, you look a lot better.
>
>He got a lot more right than just "who will win"

No one said otherwise.

>-- he correctly called
>every single one of fifty states, right down to "Florida will be very
>close but will more likely go Democratic".

And he missed rather spectacularly in 2010, as I noted in a bit that you
edited out.
--
"To every man upon this earth
Death cometh soon or late.
And how can man die better
Than facing fearful odds,
For the ashes of his fathers,
And the temples of his gods,"
- Thomas B. Macaulay

David Goldfarb

unread,
Nov 10, 2012, 10:00:31 AM11/10/12
to
In article <eins989qsv5nppt1a...@4ax.com>,
David V. Loewe, Jr <dave...@charter.net> wrote:
>On Sat, 10 Nov 2012 07:28:16, gold...@ocf.berkeley.edu (David Goldfarb)
>wrote:
>
>>David Loewe, Jr. <dlo...@mindspring.com> wrote:
>
>>>>And yet Silver's projections appear to have been extremely accurate.
>>>
>>>Well, on the second point, he was correct. When there are two options
>>>and you choose the right one, you look a lot better.
>>
>>He got a lot more right than just "who will win"
>
>No one said otherwise.

It sounded like you did.

--
David Goldfarb |From the fortune cookie file:
goldf...@gmail.com |"You think that is a secret, but it never has
gold...@ocf.berkeley.edu | been one."

Mark Zenier

unread,
Nov 9, 2012, 2:51:27 PM11/9/12
to
In article <MD76w...@kithrup.com>,
David Goldfarb <goldf...@gmail.com> wrote:
>In article <phln98pm5e7kfscca...@4ax.com>,
>David V. Loewe, Jr <dave...@charter.net> wrote:
>>A lot of the criticism of Silver revolved around two things.
>>
>>1) Unlike PECOTA, the formula isn't out there, so it can't be rigorously
>>examined by other people who are the political equivalent of
>>sabermetricians (and several sabermetricians did critiques of Silver in
>>the final days).
>>
>>2) The disconnect between the polls - especially between the state polls
>>and the national polls.
>>
>>Of further interest is the increasing lack of response to the polls.
>
>And yet Silver's projections appear to have been extremely accurate.

Reading a reprint of an article by Ken Bensinger, Los Angeles Times,
my first thought was that if they get too good at this, we'll have one
guy picked in a lottery sitting down in front of Multivac answering all
the poll questions.

Mark Zenier mze...@eskimo.com
Googleproofaddress(account:mzenier provider:eskimo domain:com)

David Loewe, Jr.

unread,
Nov 10, 2012, 5:49:55 PM11/10/12
to
On Sat, 10 Nov 2012 15:00:31, gold...@ocf.berkeley.edu (David Goldfarb)
wrote:

>David V. Loewe, Jr <dave...@charter.net> wrote:
>>On Sat, 10 Nov 2012 07:28:16, gold...@ocf.berkeley.edu (David Goldfarb)
>>wrote:
>>>David Loewe, Jr. <dlo...@mindspring.com> wrote:
>>
>>>>>And yet Silver's projections appear to have been extremely accurate.
>>>>
>>>>Well, on the second point, he was correct. When there are two options
>>>>and you choose the right one, you look a lot better.
>>>
>>>He got a lot more right than just "who will win"
>>
>>No one said otherwise.
>
>It sounded like you did.

Only if you read things that are not there.

Frankly, until you told me, I did not know how many states Silver got
right or wrong in his 2012 analysis, so commenting on that aspect would
have been utter folly for me to do.
--
"I guess I wouldn't believe in anything anymore if it weren't
for my lucky astrology mood watch."
- Steve Martin

Philip Chee

unread,
Nov 10, 2012, 11:58:11 PM11/10/12
to
Didn't Isaac Asimov write a short story using this premise?

> Mark Zenier mze...@eskimo.com
> Googleproofaddress(account:mzenier provider:eskimo domain:com)

David Goldfarb

unread,
Nov 11, 2012, 1:30:16 AM11/11/12
to
In article <63k0q7....@news.alt.net>,
Philip Chee <phi...@aleytys.pc.my> wrote:
>On Fri, 9 Nov 2012 19:51:27 GMT, Mark Zenier wrote:
>> Reading a reprint of an article by Ken Bensinger, Los Angeles Times,
>> my first thought was that if they get too good at this, we'll have one
>> guy picked in a lottery sitting down in front of Multivac answering all
>> the poll questions.
>
>Didn't Isaac Asimov write a short story using this premise?

Yes. The title was "Franchise". I assume that's exactly what Mark was
referring to -- the name "Multivac" is pretty much specific to Asimov.

--
David Goldfarb |"Understanding is a three-edged sword."
goldf...@gmail.com |
gold...@ocf.berkeley.edu | -- Babylon 5, "Deathwalker"

Philip Chee

unread,
Nov 11, 2012, 8:43:32 AM11/11/12
to
On 11/11/2012 14:30, David Goldfarb wrote:
> In article <63k0q7....@news.alt.net>,
> Philip Chee <phi...@aleytys.pc.my> wrote:
>>On Fri, 9 Nov 2012 19:51:27 GMT, Mark Zenier wrote:
>>> Reading a reprint of an article by Ken Bensinger, Los Angeles Times,
>>> my first thought was that if they get too good at this, we'll have one
>>> guy picked in a lottery sitting down in front of Multivac answering all
>>> the poll questions.
>>
>>Didn't Isaac Asimov write a short story using this premise?
>
> Yes. The title was "Franchise". I assume that's exactly what Mark was
> referring to -- the name "Multivac" is pretty much specific to Asimov.

Just saw this:
<https://twitter.com/mdlevinson/status/266064439160229888>
"Nate Silver's great-great-great-great-great-great-grandson's name is
going to be Hari Seldon. #natesilverfacts"
0 new messages