Comments on FiveThirtyEight: Politics Done Right: Minnesota Lunchtime Update: It's Gonna Be Close

酒店經紀人, 菲梵酒店經紀, 酒店經紀, 禮服酒店上班, 酒店小姐兼職, 便服酒店經紀, ...

2009-10-02T04:15:08.145-04:00

酒店經紀人,
菲梵酒店經紀,
酒店經紀,
禮服酒店上班,
酒店小姐兼職,
便服酒店經紀,
酒店打工經紀,
制服酒店工作,
專業酒店經紀,
合法酒店經紀,
酒店暑假打工,
酒店寒假打工,
酒店經紀人,
菲梵酒店經紀,
酒店經紀,
禮服酒店上班,
酒店經紀人,
菲梵酒店經紀,
酒店經紀,
禮服酒店上班,
酒店小姐兼職,
便服酒店工作,
酒店打工經紀,
制服酒店經紀,
專業酒店經紀,
合法酒店經紀,
酒店暑假打工,
酒店寒假打工,
酒店經紀人,
菲梵酒店經紀,
酒店經紀,
禮服酒店上班,
酒店小姐兼職,
便服酒店工作,
酒店打工經紀,
制服酒店經紀,
酒店經紀,
菲
梵,

自慰,女生自慰,美女,高潮,失戀,性高潮,女性高潮,G點,自慰,性,性感,性感內衣,內衣,性感內衣,...

2009-08-02T01:20:54.323-04:00

自慰,女生自慰,美女,高潮,失戀,性高潮,女性高潮,G點,自慰,性,性感,性感內衣,內衣,性感內衣,性感睡衣,情趣內衣,比基尼內衣,人妻少婦性感睡衣,
失戀,性愛,性感,情慾,調情,挑逗,慾望,情人節禮物,自慰器,

外遇外遇外遇外遇外遇外遇外遇外遇外遇外遇外遇外遇外遇外遇外遇外遇外遇外遇 ...

2009-04-16T02:15:00.000-04:00

情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,...

2009-01-17T01:08:00.000-05:00

Opus,Not sure if you're still checking this thread...

2008-12-19T14:32:00.000-05:00

Opus,

Not sure if you're still checking this thread, but you can find the simplest example (a 3 candidate situation) at the wikipedia page for monotonicity. Again, this is the simplest example, not the only one. Nor are these the only numbers for which this example works. You can play around with the numbers and see that you get the same violation with 34/35/31 instead of 39/35/26. You can also get another perverse result by switching the final and initial vote tallies. Starting with 49/35/36 and letting 10 voters abandon Andrea for their second choice (say she called them a bunch of racists, or something) causes Andrea to win. In other words, despite these voters consciously trying to make Andrea lose, they have instead given her a victory.

Indeed, in the SoS spreadsheet of Franken's challe...

2008-12-19T12:43:00.000-05:00

Indeed, in the SoS spreadsheet of Franken's challenges (which may include about 20 Coleman interlopers or doubly-challenged ballots?):

18 are resolved as "NO ONE/IDENTIFYING MARK, and
27 are resolved as "NO ONE/OVERVOTE".

These two categories strike me as independent of "voter vulnerability", and (hence?) as a Dem/Rep wash. Insofar as they may comprise a lot of Franken's said-to-be-57 upheld challenges of Coleman votes, this seems to be evidence for (b).

In that case, I should have projected 29 Coleman wins from non-vote challenges, + 2 not 3 "flips, + the same 57 vote-challenge wins as Franken got, for 88. Which is 33 short of the 121 that would have produced a tie.

One other factor: I haven't checked whether the "Coleman interlopers" on Franken's list also appeared on Coleman's list. If so, then some portion of the Coleman +7 and Other +6 resulting from challenges the STrib listed as in Coleman's 20 should be counted as part of my prediction. (The spreadsheet formats are different enough so I can't simply do a merge and unique-lines sort.)

Absenting that last factor, it looks like Coleman is coming in 3-5 short of my 29, and from the mid-morning report on TheUptake's live blog that he had knocked out 44 initially-Franken votes, maybe 3-5 short of Franken's 57 too. That would bring this up exactly to Nate's 40-vote margin, which I saw while in the middle of gathering the above figures.

If "Other" stays closer to 250 than 280, while "+C...

2008-12-19T09:16:00.000-05:00

If "Other" stays closer to 250 than 280, while "+Coleman" hits the mid-260s, this will mean that Coleman won no more vote-challenges than Franken did.

My above-linked analysis implicitly presupposes that (a) having more "vulnerable Dems" will translate into more spoiled Franken votes.

I wonder if experience with recounts shows instead that (b) the total populations of winnable vote-challenges are roughly equal.

If (b) is the rule---as "Other" near 250 will support---then only the difference in the # (and quality) of winnable non-vote challenges matters, and that would be a "prior explanation" for Franken's "real lead" marching from +4 to somewhere around 30.

+135 I meant---hit wrong key. WV "entyping" (no k...

2008-12-18T23:28:00.000-05:00

+135 I meant---hit wrong key.

WV "entyping" (no kidding!): Misteaks happen entyping.

If Minnesota had used Instant runoff, there might ...

2008-12-18T23:25:00.001-05:00

If Minnesota had used Instant runoff, there might still have been a recount, and it would have been much harder to discern voter intent with voters being asked to make multiple votes for each contest.

Worse, there would be argument and debate as to whether the votes were "transferred" accurately, since IRV requires a complex formula to tabulate the votes.

North Carolina tested IRV in a small city, and the best election officials in the state (supposedly) couldn't even count 3,000 votes accurately, and the votes had to be recounted - all of them, and all 3 rounds.

In Australia, where they have a form of IRV, it can take anywhere from 1 week to a month to get the results of a single contest election.

With IRV, if you had to do a recount, it would be worse than recounting multiple elections, it would be a nightmare.

If Minnesota voters can't mark a simple ballot correctly, imagine what they would do when asked to make multiple choices for each contest.

Crazy.

And they may be 5% further along in Coleman's list...

2008-12-18T23:25:00.000-05:00

And they may be 5% further along in Coleman's list than I thought: The STrib here shows

Olmstead County | ROCHESTER W2 P2 | Challenge 27

as "Ballot 1 of 470", but the next ballot is same-county, Challenge 26. Hence I infer they are in last-in/first-out order, i.e. that the above was the last ballot reviewed today. This means they finished on ballot 634 of Coleman's Excel list of 1,016. Assuming the withdrawals and "blue" are evenly distributed, the progress is 634/1016 = 62.4%. That projects Coleman's 15 today into 24, putting him on +262 from +238, and projects "Other" from +84 today into +134, leaving it at 249 which is definitely below my error bars. Basically it represents Coleman's "excess" challenges netting zilcho, i.e. he projects to knock out no more initial-Franken votes than Franken did vice-versa with far fewer such challenges. Or it may represent my simply being off by 30+ there. I guess I can't say more than "we'll see".

I munged part of my post by a last-minute mouse-ed...

2008-12-18T22:50:00.000-05:00

I munged part of my post by a last-minute mouse-edit: before "Franken will at least try" was "() I also predict that ", and "Minn." was supposed to be short for Minneapolis :-).

Thanks again, Latte---and great rundown of the var...

2008-12-18T22:46:00.000-05:00

Thanks again, Latte---and great rundown of the variable factors. Rather than "parse today's results", I instead modified my pre-today prediction in earlier thread---with the knowledge that Franken had won 57 vote-challenges on Tue. & Wed., rather than 33 as I thought based on that thread. The +24 improvement looks significant, but actually it made little difference, because I was basing more on concrete reports and well-founded prior assumptions about the vote-fertility of the ground and the campaigns' behavior, not just extrapolating projections. And algebraically somethings canceled out. My concrete predictions for tomorrow:

() When Yellow Medicine County is reached---assuming the "Coleman dupe challenges" are summarily ruled out and before any "200 extra challenges" from either candidate are considered---the running tally at TheUptake (and the STrib if in-sync) will show "Coleman +270". That's unchanged from what I indicated 24 hours ago.

() At the same time, "Other" will hit 282. And by my figuring, that represents overall Franken +8, having gained a staggering +4 in the challenge phase! (Not the upwards-of-20 gain I thought, because vote-challenges were notably more frequently granted than I expected, a factor favoring Coleman despite Franken's initial success.)

Today, "Coleman" went from +238 to +253, and "Other" went from 114 to 198. Based on the % of Coleman's list gone thru, the numbers instead blandly project to hitting about +265 and 270, respectively. So that indeed would be a Franken +21 gain and +25 overall margin from this phase (ceteris paribus), hence my crack there about Nate's "+27". However, today's afternoon session was dominated by Minn, so I will hold to my original numbers. (Well, I could say the lower totals are still within my error bars, but as a chessplayer I'm temperamentally avid to hit my evaluations exactly.)
Franken will at least try to re-introduce 100-200 challenges. Indeed if my "8%-er threshold" is right, then he could expect to net 6--10 votes from that, and with MN SC definitely causing a delay in when he'd expect the improperly-rejected absentees to be counted, the present value of that potential handful of votes might be considerably more than he reckoned on at the start of the week.

Of course, as Nate is even saying, everyone is working from guesstimates. The one difference I claim is that as much as possible, I'm basing on:
() reports---by M. Elias, "mediapost", and "UWBHS";
() prior assumptions: more "vulnerable" ballots are from Dems, but given equal populations of Dem and Rep vulnerables, the results will average out the same; and
() reasonable inferences based on the campaigns' behaviors to date.

@just_looking,that's a great question, and I do no...

2008-12-18T22:43:00.000-05:00

@just_looking,
that's a great question, and I do not know their fate. I strongly suspect that they were left in the folders of copied challenges that all 5 board members got just because there was not time to withdraw them physically. And when those ballots came up in the folder, Trimble would announce "withdrawn", and the Board would instantly vote to assign the vote to Franken (or 3rd party as the case may be). However, there is no way for me (or anyone but a campaign insider) to know if those spot-withdrawals are all part of the 400 or some were additional newer withdrawals yet or if a part of the 400 was removed prior to the folder (which as I mentioned, I strongly suspect was the case).

In Franken's challenges case the situation was much the same, and it caught a lot of ppl by surprise when they actually got within a handful of ballots yesterday and then finished. At least that was my read.

@ William Land,
I have no idea as to the fate of the provisionals, and have been trying to figure/find that out. I was going to ask Noah Kunin (of theuptake.org) when I saw him in person during the hearings, but he was incredibly busy, and I was just a random dude off the street to him.

During the recount, none of the observers or our site leader knew the fate of the provisional ballots, and I never got a chance to ask the county election head.

So, they remain a great mystery to me. I want to hope they were already assigned, but am unsure as to when that would've taken place.

~ Latte (off to bed)

wv ingwp: the great northern vowel drought.

MN Latte Liberal:Thanks!!!This is incredibly compl...

2008-12-18T21:41:00.000-05:00

MN Latte Liberal:

Thanks!!!
This is incredibly complicated.
I knew nothing about the provisional ballots...any idea how many there are?

Latte,Do you know if the 477 Coleman challenges re...

2008-12-18T21:05:00.000-05:00

Latte,

Do you know if the 477 Coleman challenges reported by the STrib as reviewed today included those which Coleman withdrew at the last minute (400 of them)? I thought some were brought up and withdrawn during the count? Or were all of those taken out before the counting started?

@ William Land.I am going to try :)1. How many mor...

2008-12-18T20:38:00.000-05:00

@ William Land.
I am going to try :)

1. How many more Coleman challenges will be decided tomorrow?

afair we were about half way done today. if we subtract the number (~150 or so duplicates in the blue folder, your ?5) to be ruled on first thing tomorrow morning at 9AM, we get a fairly reasonable number under 400. Very roughly.

2. Have all of the Franken challenges been decided?

Franken's got another 200 or so they could introduce pending the Board's ruling on #5 per above. They are unlikely to add more imho because a) things are breaking their way anyway, b) that will only delay the proceedings and time works against them and c)the Board, based on what I saw today after 5PM in the live discussion with both Trimble and Elias will NOT approve Coleman's motion on this. They will not uphold challenges on duplicates as double votes if only because of the equal protection issue (Franken was not allowed to challenge in Wright County on this basis per local judge who happened to be SoS Office Rep).

3. How many Coleman and Franken challenges were withdrawn and when will they be added to the totals?

Coleman started with a number north of 3200 (look at the old 538 topics and Strib archives on the matter). Now he is down to 400 left. Franken was down to 390 as per Nate's estimate yesterday out of 115 less challenges than Coleman.

Afair, the w/d's will be added to the totals by the SoS when they tabulate all the results over the weekend. I honestly do not expect anything quasi-official from Ritchie before Monday.

4. Do we really have reason to believe that Franken will get more of the absentee ballots when they are counted?

People in the know here and on pollster.com keep referring to the early voting (which would be absentee only in MN) polls that seemed to indicate that as the election date drew close there were 1-2 polls showing that folks who had already voted (absentee) broke +8 for Franken. Again, no link, going from my memory alone.

5. What with the so-called "duplicate" votes that Coleman keeps claiming exist?

This is fascinating and has been discussed ad nauseam in earlier threads. Basically, some absentee ballots are printed off wrong format oversees/filed electronically, etc. and a duplicate has to be made for each. Then each original will bear a mark "original #N" and the corresponding duplicate that was fed through the counter ought to have the same "duplicate N" marking. Often [enough] times, for whatever reason, the dupes cannot be found during the recount. Hence [false] concern by Coleman/Trimble about double voting, now handily debunked by Elias. And that, folks, was a thing of beauty. A beauty, I tells ya!

6. Is there another category of votes I should know about?

Provisional ballots. Those cast on election night where the credentials of the voter had yet to be confirmed. Sorry about this can of worms :)

~Latte

wv crock

I tend to believe that the Star Tribune's internet...

2008-12-18T20:20:00.000-05:00

I tend to believe that the Star Tribune's internet vote on the ballots will be 99.9% accurate for withdrawn ballots[they were withdrawn because they were frivolous so anyone voting should be able to figure them out. Looking at the resolved challenges, ballot box stuffing doesn't appear to be a problem]. From that it seems that Franken +89 is the most accurate current state of the race that we have with the biggest question mark in my mind being how the star tribune is allocating the green/blue folder ballots.

Can't wait for KWR to parse today's results. I mis...

2008-12-18T20:19:00.000-05:00

Can't wait for KWR to parse today's results. I missed some of the proceedings as I was driving to the CB meeting, where my laptop battery eventually died.

@shma :)
We did, and I would only restate that I am still a fan of IRV with a proviso that the IRV kicks in under specially defined set of rules, not just during lack of majority.

@green libertarian,
just got home, have not read the ruling yet, but from the MPR and Air America and telephone analysis :)) I get the interpretation that what all counties and campaigns have to agree on is a uniform standard under which to count the wrongly rejected votes, not agree on every vote. Everyone I heard from might be wrong, and I do not discount that, but I would not lose hope quite yet.

I had Allan Page's dissenting opinion read to me as I drove, and I tend to concur with him up to a point regarding chaos. Mob rule should not be, the SOS ought to take helm at this point and proclaim: "these are my standards according to the MN law. Deal with it, peons!" Be firm, lay down the lay for the counties now that he's got the umbrella SC ruling.

~ Latte

wv: tingla. I've got this happy tingla feeling after watching so many Coleman challenges get rejected today live.

So, Coleman appears to up by 2 votes.A few questio...

2008-12-18T20:10:00.000-05:00

So, Coleman appears to up by 2 votes.
A few questions:

1. How many more Coleman challenges will be decided tomorrow?
2. Have all of the Franken challenges been decided?
3. How many Coleman and Franken challenges were withdrawn and when will they be added to the totals?
4. Do we really have reason to believe that Franken will get more of the absentee ballots when they are counted?
5. What with the so-called "duplicate" votes that Coleman keeps claiming exist?
6. Is there another category of votes I should know about?

Thanks!

The MN Supremes did NOT order the counties to coun...

2008-12-18T20:06:00.000-05:00

The MN Supremes did NOT order the counties to count the erroneously rejected absentee ballots. They said count ONLY the ones that BOTH campaigns agree were erroneously rejected. That'll never happen. They WIMPED out, just read the great Justice Alan Page's scathing opinion in dissent.

Looks like Franken may have to take this to federal court, unless, highly unlikely, Coleman concedes after the Canvassing Board certifies the election in favor of Franken (likely).

I like IRV. Always have. But can you imagine the k...

2008-12-18T19:52:00.000-05:00

I like IRV. Always have.

But can you imagine the kinds of trouble some nitwit like Mr/Ms Lizard People would have with an IRV ballot? He/she would probably mark EVERY candidate as a first preference and then put "Lizard People" as 0.5.

There are times when I think some people just shouldn't vote at all.

Two things: #1 I am a fan of Instant Runoff #2 I a...

2008-12-18T19:19:00.001-05:00

Two things:

#1 I am a fan of Instant Runoff

#2 I am less of a fan than I use to be.

Recently I learned of other possible reforms besides IRV. A relatively few number of people have evaluated these options and opined. They have pretty polarized views and it seems to me each side has some valid arguments and each side has some, more ahhh, seemingly manure- type arguments.

With the greatest of respect to the folks at Fairvote.org and RangeVoting.org, something as important as our democracy ought not be debated by a few theorists and 501(c)3 activists. Every community thinking about reform should be aware of the options, have its *own* conversation about them, and contribute its thinking to the national conversation as a whole.

All that said, I'm still an IRV guy, for now. However, I do think communities should take a hard look at at the various alternatives and then consciously decide which they think is the best one to try in their own town. Maybe its IRV, but maybe its one of the others.

Steve E

The idea of "instant run-off" voting sounds good, ...

2008-12-18T19:19:00.000-05:00

The idea of "instant run-off" voting sounds good, but I am worried about the fact that it is very different from what voters are used to and would force them to consider things they haven't had to. How many voters will (or won't) understand it and utilize it?

I know that the city of Cambridge, MA had a ssytem of sequential preference voting for city council at one point. (I don't know if they still do.) There were 9 at large seats on the city council and voters could rank order up to 9 candidates. A certain percentage of votes guarranteed a seat. The lowest candidate was eliminated and the votes were then awarded on the second count to those voters' second choice. Those candiates who achieved the threshhold vote were elected on the first count, and some proportional method of assigning "excess votes" to the second choice candidates on the seond oount would be used. The process get repeated for each round of counting until 9 candidates were elected. As I recall, the one year I was living there when this happened the final count wasn't arrived at until Friday froma Tueday election.

I know that not all voters used all nine votes. I would imagine that there was some scholarly research on those voting patterns, probably by researchers from MIT as Hasrdard's government department was strongly anti-behavioralist at that time.

I recall immediately after a senior honors oral exam that the head of the Government department asked me what I was going to be doing the next year, and when I told him that I was going to go to Michigan for graduate work a sneer crept into his expression and he said something like "Oh, so you're gong to play with comoputers."

(Barney Frank, at that time a grad student and a teaching fellow for the chair, was a member of the examining grooup.)

I don't see how Coleman wins with only a 10% win r...

2008-12-18T19:04:00.000-05:00

I don't see how Coleman wins with only a 10% win rate. I thought he reduced his challenges to 800 (used to be 1000, subtracted 400, and added 200).

That will leave him about 30 to 40 votes short (assuming Franken's campaign was correct about the 4 vote margin if all ballots were rejected).

It would be painful if Franken lost only because the absentee ballots were counted.

Thanks joseph,you made my night.

2008-12-18T19:01:00.000-05:00

Thanks joseph,you made my night.

Comments on FiveThirtyEight: Politics Done Right: Minnesota Lunchtime Update: It's Gonna Be Close

酒店經紀人, 菲梵酒店經紀, 酒店經紀, 禮服酒店上班, 酒店小姐兼職, 便服酒店經紀, ...

自慰,女生自慰,美女,高潮,失戀,性高潮,女性高潮,G點,自慰,性,性感,性感內衣,內衣,性感內衣,...

外遇外遇外遇外遇外遇外遇外遇外遇外遇 外遇 外遇 外遇外遇 外遇 外遇 外遇 外遇 外遇 ...

情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,...

Opus,Not sure if you're still checking this thread...

Indeed, in the SoS spreadsheet of Franken's challe...

If "Other" stays closer to 250 than 280, while "+C...

+135 I meant---hit wrong key. WV "entyping" (no k...

If Minnesota had used Instant runoff, there might ...

And they may be 5% further along in Coleman's list...

I munged part of my post by a last-minute mouse-ed...

Thanks again, Latte---and great rundown of the var...

@just_looking,that's a great question, and I do no...

MN Latte Liberal:Thanks!!!This is incredibly compl...

Latte,Do you know if the 477 Coleman challenges re...

@ William Land.I am going to try :)1. How many mor...

I tend to believe that the Star Tribune's internet...

Can't wait for KWR to parse today's results. I mis...

So, Coleman appears to up by 2 votes.A few questio...

The MN Supremes did NOT order the counties to coun...

I like IRV. Always have. But can you imagine the k...

Two things: #1 I am a fan of Instant Runoff #2 I a...

The idea of "instant run-off" voting sounds good, ...

I don't see how Coleman wins with only a 10% win r...

Thanks joseph,you made my night.

外遇外遇外遇外遇外遇外遇外遇外遇外遇外遇外遇外遇外遇外遇外遇外遇外遇外遇 ...