Baseball for the Thinking Fan

Login | Register | Feedback

btf_logo
You are here > Home > Baseball Newsstand > Baseball Primer Newsblog > Discussion
Baseball Primer Newsblog
— The Best News Links from the Baseball Newsstand

Monday, October 08, 2018

OTP 2018 October 8: Hugh Jackman’s ‘The Front Runner’ Confronts The Political Conundrum Of Our Time

The picture drops us into the thick of the 1984 campaign right as Hart (a low-key and weather Jackman, oddly eschewing his usual star charisma) conceding the nomination to Walter Mondale. As you know, Mondale got slaughtered by Ronald Regan in the general election, losing all but one state in the electoral college (despite still winning 41% of the popular vote). So, four years later, Hart is back in the thick of it, as he argues that the failed 1984 campaign was partially to introduce himself on the national stage for the next time anyway. And this time out, he quickly becomes the front runner for the nomination and the general election.

 

Want to know what happened next? I guess you have to watch the movie.


(As always, views expressed in the article lede and comments are the views of the individual commenters and the submitter of the article and do not represent the views of Baseball Think Factory or its owner.)

Mellow Mouse, Benevolent Space Tyrant Posted: October 08, 2018 at 09:00 AM | 1559 comment(s) Login to Bookmark
  Tags: movie, off topic, politics

Reader Comments and Retorts

Go to end of page

Statements posted here are those of our readers and do not represent the BaseballThinkFactory. Names are provided by the poster and are not verified. We ask that posters follow our submission policy. Please report any inappropriate comments.

Page 12 of 16 pages ‹ First  < 10 11 12 13 14 >  Last ›
   1101. You Know Nothing JT Snow (YR) Posted: October 12, 2018 at 12:01 PM (#5765107)
Ray's too embarrassed to admit his role model is Hannity.


No, he's too embarassed to admit his actual role model is Charles Emmerson Winchester III. At least that guy did something valuable.
   1102. perros Posted: October 12, 2018 at 12:01 PM (#5765108)
Howie's comment was way out of character, huh?
   1103. perros Posted: October 12, 2018 at 12:03 PM (#5765109)
his actual role model is Charles Emmerson Winchester III.


Heh.
   1104. ERROR---Jolly Old St. Nick Posted: October 12, 2018 at 12:05 PM (#5765112)
MAGA Twitter update:

@Ray6971
So Google is celebrating a #Communist country today...


I can't tell if that guy is being serious or not. If so, WOW!

I think he actually is serious. Here's his full handle:
Ray
@Ray6971

   1105. DJS vs. The White Knights Posted: October 12, 2018 at 12:06 PM (#5765113)
I don't know if I was glad about it, but yeah, couldn't have been more wrong about that one.

Yeah, it seemed quite unlikely. But still, I'd avoid poop-eating bets for even mildly plausible things.

Now, I did wager on eating a hat this winter and not a nacho hat, but the Royals have to sign Machado and Harper. So I feel pretty good about that one.
   1106. The Yankee Clapper Posted: October 12, 2018 at 12:09 PM (#5765115)
Not a poll ... Beto O'Rourke raised $38.1 million in last three months, smashing all records in bid to oust Ted Cruz

What's O'Rourke getting for that money? After being listed as a Toss-Up in September, the race is back in the GOP column, and Cruz now has a 7% lead in the RCP poll average. It's nice that the vendors and consultants will apparently get paid, but that's not a lot of return on investment. But perhaps I'm wrong - send him more $$.
   1107. David Nieporent (now, with children) Posted: October 12, 2018 at 12:10 PM (#5765117)
I don't know if I was glad about it, but yeah, couldn't have been more wrong about that one.

Yeah, it seemed quite unlikely. But still, I'd avoid poop-eating bets for even mildly plausible things.
Also, bug-eating ones. (Ask Sam Wang.)
   1108. perros Posted: October 12, 2018 at 12:12 PM (#5765121)
We get too distracted by Trump's nonsense while ol' Jefferson Beauregard III is trying to stop the consent decree with the Chicago PD. You know, the same police department that tortured some folks:

Police officers under former Chicago police commander Jon Burge used electrical shock, burning and mock executions to elicit confessions from suspects, mostly African-American, from the early 1970s through the early 1990s.

The statute of limitations ran out on his alleged crimes, but Burge was convicted in 2010 of perjury in civil proceedings for lying about torture he oversaw.

Burge was released from prison to a halfway house in October after serving less than four years in prison. He was released from the halfway house earlier this year.

Burge still receives a pension for his years on the force.
   1109. perros Posted: October 12, 2018 at 12:13 PM (#5765123)
Of course Trump hates Sessions and wants to get rid of him.

   1110. PreservedFish Posted: October 12, 2018 at 12:19 PM (#5765126)
   1111. Zonk just has affection for alumni Posted: October 12, 2018 at 12:21 PM (#5765128)
FWIW - Burge doesn’t get his pension anymore. He died last month.... just in case anyone keeps lists of graves that need watering.
   1112. ERROR---Jolly Old St. Nick Posted: October 12, 2018 at 12:21 PM (#5765129)
What's O'Rourke getting for that money? After being listed as a Toss-Up in September, the race is back in the GOP column, and Cruz now has a 7% lead in the RCP poll average. It's nice that the vendors and consultants will apparently get paid, but that's not a lot of return on investment. But perhaps I'm wrong - send him more $$.

The Times has been doing live polling updates on Senate races, including Texas, and here's their footnote about "Turnout":
Assumptions about who is going to vote may be particularly important in this race.
Our poll under different turnout scenarios
Who will vote? Est. turnout Our poll result


The types of people who voted in 2014 4.4m Cruz +15

People whose voting history suggests they will vote, regardless of what they say 6.3m Cruz +9
Our estimate 6.3m Cruz +8

People who say they will vote, adjusted for past levels of truthfulness 6.8m Cruz +8

People who say they are almost certain to vote, and no one else 7.3m O’Rourke +3

The types of people who voted in 2016 7.9m Cruz +5

Every active registered voter 13.2m Cruz +4
   1113. Misirlou doesn't live in the restaurant Posted: October 12, 2018 at 12:25 PM (#5765132)
I think he actually is serious. Here's his full handle:
Ray
@Ray6971


I don't get it.
   1114. perros Posted: October 12, 2018 at 12:32 PM (#5765136)
just in case anyone keeps lists of graves that need watering.


Noted.
   1115. Davo and his Moose Tacos Posted: October 12, 2018 at 12:34 PM (#5765137)
Turkish court frees US pastor Andrew Brunson
The US evangelical preacher was accused of links with Kurdish rebels and supporters of Fethullah Gulen, whom Turkey blamed for the failed coup attempt. Gulen has denied any involvement.

Brunson, who has lived in Turkey for more than 20 years, has denied the charges and maintained his innocence.

(...He) was convicted of terror-related charges and sentenced to three years, one month and 15 days in jail on Friday.

But he was freed, taking into account the time already served and good conduct during the trial. The court also lifted his house arrest and overseas travel ban, paving the way for his return to the US.

(...)"This is the day our family has been praying for -- I am delighted to be on my way home to the United States," Brunson said in a statement.
   1116. perros Posted: October 12, 2018 at 12:37 PM (#5765138)
   1117. Mellow Mouse, Benevolent Space Tyrant Posted: October 12, 2018 at 12:41 PM (#5765140)
The Times has been doing live polling updates on Senate races, including Texas, and here's their footnote about "Turnout":


Turnout s always important and really hard to correctly estimate. It is extra important in midterm elections. It also happens to be the entire reason Texas is so very red - the blue demographic cohorts don't turnout and the red one's do.

Until election day no one has any idea what exactly Beto (and those down ballot) are getting for their money. It is still pretty unlikely Beto wins, but it would be even more unlikely if he didn't have a hoard of cash at his disposal.
   1118. ERROR---Jolly Old St. Nick Posted: October 12, 2018 at 12:41 PM (#5765141)
I think he actually is serious. Here's his full handle:
Ray
@Ray6971


I don't get it.

Understandable, but what I don't get is why the rest of his handle keeps getting zapped. Beats me why, but I'll try again:


EDIT: It happened again. But here's a link to his Twitter page.

Must be something about Twitter that I don't get, sort of like how Ray6971 doesn't get the difference between Puerto Rico and Cuba.
   1119. PreservedFish Posted: October 12, 2018 at 12:46 PM (#5765144)
Life regret: I was at a restaurant event in Los Angeles and I was hanging out with this man, an Italian "blacksmith to the stars", who I had met once or twice before. He told me that he was about to go to Werner Herzog's house for a dinner party. We talked about it for a while and I tried to get myself invited to Werner Herzog's house. I had no success. I look back on it and think that I should have been more shameless about asking to join him, or, contrariwise, should have simply happily followed him as if I were too stupid to understand that I wasn't wanted. Werner Herzog believes that cooking is the world's only other worthwhile profession, so I believe we could have found a few things to talk about.
   1120. zenbitz Posted: October 12, 2018 at 12:49 PM (#5765147)
I don't know if HIPAA is the worst law on the books, but it's the worst kind of law
   1121. Zonk just has affection for alumni Posted: October 12, 2018 at 12:52 PM (#5765151)
Turnout s always important and really hard to correctly estimate. It is extra important in midterm elections. It also happens to be the entire reason Texas is so very red - the blue demographic cohorts don't turnout and the red one's do.

Until election day no one has any idea what exactly Beto (and those down ballot) are getting for their money. It is still pretty unlikely Beto wins, but it would be even more unlikely if he didn't have a hoard of cash at his disposal.


FWIW, Beto does have a "plan" -- a long-touted plan and what his campaign has often said he's spending the money on, rather than blanketing the airwaves.

It's basically the Obama sort of stuff - vast data gathering, targeting of new voters and seldom voters, dumping resources into juicing turnout in seldom D areas AND wringing every last vote from D strongholds.

Everybody says that, of course... but sometimes, candidates actually mean it and it works.
   1122. PreservedFish Posted: October 12, 2018 at 12:55 PM (#5765155)
His plan definitely involves plastering this Maine resident's Facebook page with ads, I can tell you that much.
   1123. Lassus Posted: October 12, 2018 at 01:00 PM (#5765157)
I don't know if HIPAA is the worst law on the books, but it's the worst kind of law.

Care to elaborate? I may have missed something upthread.
   1124. Count Vorror Rairol Mencoon (CoB) Posted: October 12, 2018 at 01:01 PM (#5765158)
And so it continues:


Donald Trump has frequently and falsely crowed about the idea of so-called paid protesters, including most recently the sexual assault survivors who confronted senators in the lead-up to the Brett Kavanaugh confirmation. Now his administration may be trying to turn that concept on its head, by requiring citizens to pay to be able to protest, among other affronts to the first amendment.

Under the proposal introduced by the interior secretary, Ryan Zinke, in August, the administration is looking to close 80% of the sidewalks surrounding the White House, and has suggested that it could charge “event management” costs, for demonstrations.


Graun
   1125. Count Vorror Rairol Mencoon (CoB) Posted: October 12, 2018 at 01:06 PM (#5765160)
Well, Beto's got at least one new vote:


Beto O’Rourke, the charismatic Democrat who is stirring things up in Texas with his attempt to turf Ted Cruz out of the US Senate, is famous for leaving no stone unturned.

He has visited all of Texas’s 254 counties this past year. That’s a lot of shoe leather in a state larger than France.

But just how far O’Rourke is prepared to go in trying to wrest the state from being Republican stranglehold (the last time Texas sent a Democrat to the US Senate was in 1993) only became clear after the Guardian published an article on his campaign last week.

The end of the piece described a 21-year-old Latino man called Sebastian Esquivel who works in his family’s restaurant, Milupita Taco House, in Gonzales, a small cowboy town south of Austin. Some level of detachment from Esquivel was perhaps to be expected, given the low level of political engagement and turnout among Hispanic Texans.

But Esquivel revealed that only one member of his 20-strong extended Hispanic family had ever voted. Having been ignored for so long by politicians of all colors, what was the point?

“To be honest, to me it doesn’t really matter,” Esquivel said.

Sebastian Esquivel, 21, previously told the Guardian he chooses not to vote in elections because he doesn’t feel it will make a difference in his daily life.

“We all agreed that this November is too important for Sebastian and any Texan not to know what we’re fighting for,” Chris Evans, the campaign’s communications manager, posted on Medium. “But we also agreed that it’s not on Sebastian to reach out to us. It’s on us to reach out to him.”

So what did they do? From his car as he was travelling to Houston for a campaign event, O’Rourke tracked down the phone number of the restaurant and cold called Esquivel.

With the Hispanic cook on the line, the candidate outlined his progressive policies, which include a commitment to immigration reform, universal healthcare and gun control, and asked Esquivel in return about his concerns. Esquivel said that not only had he never voted, he wasn’t even registered to vote.

O’Rourke used his network of fired-up young volunteers to send a field worker to Gonzales, some 70 miles away, to persuade Esquivel to register. The cook duly did so, and the O’Rourke volunteer then drove him to the post box to mail it – this on the final day of registration allowed in Texas.


Graun
   1126. dlf Posted: October 12, 2018 at 01:09 PM (#5765161)
Which, again, makes me feel that TD, dlf, etc owe us some answers... Are two big sell-offs and volatility spikes in a 9 month span sans some intervening event usual?


Here is the really simplistic take: if you are prudently invested (you don't have all your money in bitcoin and pot stock; you aren't overly concentrated in a particular small set or type of investment; you have anything needed in the next 2-5 years in something as conservative as CDs ...) you should see big sell-offs as a GOOD THING.

As Buffett says, "if we weren't all crazy we would go insane." Oops, that was Jimmy, not Warren.

Warren would tell you that pull backs like this are an opportunity to buy. The stock market is the only market where the buyers are happy to pay more to buy the same thing and get depressed when the cost to buy goes down. Don't fall into that trap.

And more importantly for simple passive investors like me, remember that the market in the short term is a very poor reflection of the larger economy.


And in 600 more points I'll be drinking dlf's whiskey.....:)


TD knows this but for those following along from the wings:

S&P as of 1:00 pm EDT on 10/12 2,738. That is all the way down to a level we haven't seen since those long ago days of July 5, 2018.

Even after the pull back, we are up 2% on the year and roughly 30% since Traderdave and I made this bet. While no one knows what will happen in the short run, one of the safest bets ever is on the mid to long term results of the U.S. market.
   1127. PepTech, the Legendary Posted: October 12, 2018 at 01:15 PM (#5765163)
With all the unrest in the stock market, good thing we can point to the VIX for confidence in stabil... oh.
   1128. dlf Posted: October 12, 2018 at 01:35 PM (#5765175)
With all the unrest in the stock market, good thing we can point to the VIX ...


I suspect that I'm missing a meme somewhere, but why would anyone who isn't a day trader be worried about VIX? If you insist, look at the end of day figures for the S&P. Better yet, look at the trailing 200 day average of the S&P. But the daily degree of volatility? To anyone but speculators - and more so, very short term speculators - that has no more meaning than the size of the ice spikes on Europa does to the height of the surf in La Jolla.
   1129. tshipman Posted: October 12, 2018 at 01:42 PM (#5765180)
I suspect that I'm missing a meme somewhere, but why would anyone who isn't a day trader be worried about VIX?


SBB was citing the VIX as some kind of evidence of Trump being great.
   1130. David Nieporent (now, with children) Posted: October 12, 2018 at 01:42 PM (#5765181)
I suspect that I'm missing a meme somewhere,
FLTB spent the first year+ of Trump's presidency touting VIX as demonstrating there was nothing wrong with Trump's policies or conduct. Volatility was low, so therefore the markets weren't worried about Trump so therefore he must be making America great again.


EDIT: Coke to Shipman for typing infinitesimally faster.
   1131. Zonk just has affection for alumni Posted: October 12, 2018 at 01:44 PM (#5765184)
Here is the really simplistic take: if you are prudently invested (you don't have all your money in bitcoin and pot stock; you aren't overly concentrated in a particular small set or type of investment; you have anything needed in the next 2-5 years in something as conservative as CDs ...) you should see big sell-offs as a GOOD THING.

As Buffett says, "if we weren't all crazy we would go insane." Oops, that was Jimmy, not Warren.

Warren would tell you that pull backs like this are an opportunity to buy. The stock market is the only market where the buyers are happy to pay more to buy the same thing and get depressed when the cost to buy goes down. Don't fall into that trap.


Heh, I know... I generally resist the temptation to even bother checking the 401(k) and Roth during such sell-offs, but it's a hard temptation to resist... so inevitably, I do... and then get the freakout when realizing OMG, that number just dropped by more than I make in a month... in a couple freakin days!

To be perfectly honest, I suppose I'm sorta, kinda looking for a bottom in more of anticipatory fashion... not that I'm rich beyond my dreams or anything, but I find myself fairly debt free and with a money market emergency fund that's well past what it needs to be at this point. Thought about maybe buying a place, but still on the fence on that... to be honest, I think I prefer just calling someone to fix whatever and not having to deal with it. Anyway, while the rate hikes continue to make that look prettier and prettier, I know it could be doing better elsewhere. I suppose attempting to time the market in such a fashion is a fool's errand - but I guess that, at minimum, it's probably time to just dump some dollars into something with a little more growth potential than compounding interest.
   1132. Zonk just has affection for alumni Posted: October 12, 2018 at 01:56 PM (#5765191)
   1133. dlf Posted: October 12, 2018 at 01:56 PM (#5765193)
Thought about maybe buying a place, but still on the fence on that... to be honest, I think I prefer just calling someone to fix whatever and not having to deal with it. Anyway, while the rate hikes continue to make that look prettier and prettier, I know it could be doing better elsewhere.


FWIW, while rates are closing in on 5%, a significant upswing from the ~3% of a few years ago, this is still waaaay below historical norms. The world of refinancing has cratered, but new purchases are and should be, doing well.

Edit: note that this isn't a specific recommendation since I don't know your area at all. A a lot of folks in the housing industry think that some localities are in bubbles with prices having escalated faster than the fundamentals warrant. But unlike 2007, this is very localized. (Of course even 2007 was somewhat localized with NV, AZ, CA, and FL wacked and other places less so. (And the Rust Belt issues were different and longer standing than the 2000s inflated run-up.))

I suppose attempting to time the market in such a fashion is a fool's errand - but I guess that, at minimum, it's probably time to just dump some dollars into something with a little more growth potential than compounding interest.


Dollar cost averaging. I'm an idiot who knows nothing. But knowing I know nothing, I keep it simple. 80% of what I have are in very low cost index funds. (Fidelity just announced a zero cost one.) And when I have extra to add to my investments, I split it into 6 (or more) equal bits and add to my portfolio 1/6 at a time every 30 days.

FLTB spent the first year+ of Trump's presidency touting VIX as demonstrating there was nothing wrong with Trump's policies or conduct. Volatility was low, so therefore the markets weren't worried about Trump so therefore he must be making America great again.


Good grief, that is stupid. If the market went down a couple of basis points a day, volatility would be nil, but it would be a sign of a horrible economy; if the market went up a couple of dozen basis points on the even days and down N-1 the odd days, that would be a good economy, but an incredibly volatile market. The two just aren't correlated.
   1134. BrianBrianson Posted: October 12, 2018 at 01:58 PM (#5765194)
So, uhm, if I'm not retiring for ~30 years, I should just leave my (American) pension in mostly "Equities"? And put in my 5% to collect my employer's 9% (8%?)

My British pension is fixed payout, so I don't have to worry about such things. I can't collect it until I reach the UK's retirement age, which I assume will increase with a slope greater than or equal to one by the time I'm approaching it.
   1135. David Nieporent (now, with children) Posted: October 12, 2018 at 01:59 PM (#5765196)

My British pension is fixed payout, so I don't have to worry about such things. I can't collect it until I reach the UK's retirement age, which I assume will increase with a slope greater than or equal to one by the time I'm approaching it.
If you believe Davo/Perros, the UK should actually be underwater by then, so I don't think you'll be able to collect.
   1136. dlf Posted: October 12, 2018 at 02:09 PM (#5765201)
Heh, I know... I generally resist the temptation to even bother checking the 401(k) and Roth during such sell-offs, but it's a hard temptation to resist... so inevitably, I do... and then get the freakout when realizing OMG, that number just dropped by more than I make in a month... in a couple freakin days!


Jeff Bezos has "earned" something north of $225,000,000.00 today and has "lost" over $1,000,000,000.00 in the last week. Think of those kind of swings. :)

I used to report directly to the CEO / Chair of a publicly traded entity. The value of his ownership interest went from a bit more than two billion to just under five hundred million. In percentage terms, that is horrible. But the dude still has a financial net worth of half a billion dollars.

Edit: Yes, David, those scare quotes are intended.
   1137. Mellow Mouse, Benevolent Space Tyrant Posted: October 12, 2018 at 02:13 PM (#5765202)
China trade surplus with US widens to record $34.1B

I am shocked Trump's plan is not working. Shocked I say.
   1138. dlf Posted: October 12, 2018 at 02:15 PM (#5765203)
S&P as of 1:00 pm EDT on 10/12 2,738.


Between the time I wrote that and now, 2:10 pm, the S&P went down to 2720 and up to 2747, or down half a percent and back up to more than a quarter point over what it was. All in one hour. (Those are just the points I've seen; it may have swung more.) And that swing means NOTHING.
   1139. David Nieporent (now, with children) Posted: October 12, 2018 at 02:19 PM (#5765204)

And that swing means NOTHING.
You're wrong. It proves exactly what I already knew to be true.
   1140. Mellow Mouse, Benevolent Space Tyrant Posted: October 12, 2018 at 02:23 PM (#5765207)
New Forecast Sees Democrats Winning 229 House Seats

I want more, but if it happens it is a good outcome.
   1141. dlf Posted: October 12, 2018 at 02:27 PM (#5765211)
So, uhm, if I'm not retiring for ~30 years, I should just leave my (American) pension in mostly "Equities"? And put in my 5% to collect my employer's 9% (8%?)


I don't know you or your situation, but um, yes(*). Take the free(**) money from your employer and put it where you'll average something like 7% per year above inflation. Yes, the market can go down - it does so, on average, one year in three - but since the Civil War, there has never been a thirty year period where the U.S. market was down.

(*)Not to be taken as specific advice. No representation is made that the stupid comments by the stupid poster are any less stupid than the norm. No proof that no animals have been harmed in the making of this post. Mouse and mongoose beware.

(**)Yeah, it's not really free. You have to work for it. But assuming you keep the job and aren't able to negotiate for more in the paycheck instead (something vanishingly few employers would do) its the closest to a free lunch that you'll find.
   1142. Zonk just has affection for alumni Posted: October 12, 2018 at 02:39 PM (#5765221)
Thought about maybe buying a place, but still on the fence on that... to be honest, I think I prefer just calling someone to fix whatever and not having to deal with it. Anyway, while the rate hikes continue to make that look prettier and prettier, I know it could be doing better elsewhere.


FWIW, while rates are closing in on 5%, a significant upswing from the ~3% of a few years ago, this is still waaaay below historical norms. The world of refinancing has cratered, but new purchases are and should be, doing well.


Yeah, I know... what keeps holding me back is actually - I've found that since returning, I still love living in Chicago as much as I ever did. While I could afford a nice enough place in the city, for the same cost? If I were to look even to the suburbs, much less the freedom (lessened since returning, but still substantial enough) my job allows? To be honest, I've sort of shifted towards looking well below even the modest levels in my price range with the idea of #### it, I could buy, pay it off in relative short order... but then, I figure, why step down living arrangements for what amounts to even quasi investment.

So, uhm, if I'm not retiring for ~30 years, I should just leave my (American) pension in mostly "Equities"? And put in my 5% to collect my employer's 9% (8%?)

I don't know you or your situation, but um, yes(*). Take the free(**) money from your employer and put it where you'll average something like 7% per year above inflation. Yes, the market can go down - it does so, on average, one year in three - but since the Civil War, there has never been a thirty year period where the U.S. market was down.


*sigh*

I wish my employer did a match - they use a profit-sharing formula based on salary instead... which was great when I started, because I could only afford a point or so and it vests almost immediately (I think it was a year + 1 day). Nowadays, even a match at the lower end of the scales would be more beneficial to me. Oh well... I guess, being the good liberal I am, I should be applauding them because the plan's terms are probably better for entry level folks just starting out who can't really afford much in the manner of deductions for contributions. So I will.
   1143. JL72 Posted: October 12, 2018 at 02:48 PM (#5765224)
I wish my employer did a match - they use a profit-sharing formula based on salary instead... which was great when I started, because I could only afford a point or so and it vests almost immediately (I think it was a year + 1 day). Nowadays, even a match at the lower end of the scales would be more beneficial to me. Oh well... I guess, being the good liberal I am, I should be applauding them because the plan's terms are probably better for entry level folks just starting out who can't really afford much in the manner of deductions for contributions. So I will.


I am old enough to remember when this was exactly the scenario that was predicted when 401(k)'s came into vogue. The corporate world argued strenuously that they would never stop matching, because they needed to be competitive with other companies. Funny how that turned out wrong (but in their favor).
   1144. BrianBrianson Posted: October 12, 2018 at 02:53 PM (#5765230)
Well, I'm on soft money, so in all likelihood I'll be out on my ass by the end of August. If I can find a new job, then when I actually leave will have to be discussed, of course. The only thing I can actually do is re-balance how much of my pension is in "GUARANTEED", "EQUITIES", "REAL ESTATE", "FIXED INCOME", and "MULTI-ASSET" - I think I can also make specific fund picks within those, but I suspect that'd be tossing darts down a well, i.e., random and I've had no idea what was going on. I have a vague understanding that as you move closer to retirement, you might consider moving more of it into guaranteed/fixed income type deals.
   1145. This is going to be state of the art wall Posted: October 12, 2018 at 03:00 PM (#5765234)
And so it continues:


TGIF, it's a depressing news cycle otherwise.
   1146. Srul Itza Posted: October 12, 2018 at 03:07 PM (#5765237)
New Forecast Sees Democrats Winning 229 House Seats


From that article-let:

However, the margin of error is wide enough that they could end up with 205 seats, short of a majority, or 262 seats, well more than needed for a majority.


I have complete faith in the Democrats to find a way to beat the projection -- on the low side.

Party of Losers.
   1147. Zonk just has affection for alumni Posted: October 12, 2018 at 03:16 PM (#5765245)
I have complete faith in the Democrats to find a way to beat the projection -- on the low side.

Party of Losers.


Man, Srul.... your campaign to update the donkey branding image continues unabated :-)
   1148. dlf Posted: October 12, 2018 at 03:24 PM (#5765251)
Between the time I wrote that and now, 2:10 pm, the S&P went down to 2720 and up to 2747, or down half a percent and back up to more than a quarter point over what it was. All in one hour.


Friendly neighborhood stock update ... S&P up another 5 to 2751. VIX going crazy; economy still sound.

Edit: woops, up another point to 2752 between starting to type and finishing the post.

...

Partisans always give too much credit to their guy when the economy is good and too much blame when it is not. Presidents and Congresscritters have remarkably little ability to make a major change in the economy. And thus does dlf end his pontificating for the day. Off to have a glass of wine with mrs. dlf.
   1149. homerwannabee Posted: October 12, 2018 at 03:37 PM (#5765257)
The one thing the Hillary Clinton loss has taught me is this. It's basically meaningless until about 2 weeks until election day. October 17th 2016 Hillary Clinton was up by the Real Clear polling average with a 7.1% lead. She had many polls having her over 10 points. So pump the breaks on ANY prognostication until about two weeks until the election.
   1150. This is going to be state of the art wall Posted: October 12, 2018 at 03:41 PM (#5765258)
The one thing the Hillary Clinton loss has taught me is this. It's basically meaningless until about 2 weeks until election day. October 17th 2016 Hillary Clinton was up by the Real Clear polling average with a 7.1% lead. She had many polls having her over 10 points. So pump the breaks on ANY prognostication until about two weeks until the election.


Well there was a little October surprise that James "Cover my ass how do you think I rose this high in the FBI" Comey decided to spring.

In general I agree tho. Polls are useful for internal strategy but up until a couple weeks before the election they aren't that predictive, at least for competitive districts without huge registration differences.
   1151. Zonk just has affection for alumni Posted: October 12, 2018 at 03:46 PM (#5765265)
In general I agree tho. Polls are useful for internal strategy but up until a couple weeks before the election they aren't that predictive, at least for competitive districts without huge registration differences.


Yup.

They're valuable for making the necessarily final leg decisions on where to place your resources.

For the GOP, this has obviously meant who to cut loose. For the Dems, they're in the enviable position - helped simply by being at a nadir - of looking where else they can poach... and yes, for purposes of Clapper's favorite Florida seat - given the fact that something like 60+ challengers managed to raise 1m+ on their own - it also means that they can dump money and resources on those few races where a poor candidate is under-performing what should have been among the better opportunities.

The Dems real advantage right now is that a lot of their surprising possibilities - Beto being the prime example - don't actually need parties and PACs to shore them up; they've managed to become self-funding surprises.
   1152. tshipman Posted: October 12, 2018 at 03:50 PM (#5765268)
Partisans always give too much credit to their guy when the economy is good and too much blame when it is not. Presidents and Congresscritters have remarkably little ability to make a major change in the economy. And thus does dlf end his pontificating for the day. Off to have a glass of wine with mrs. dlf.


This is one of those statements that is true, but probably oversimplified.

The biggest economic policy decision of my lifetime was granting China permanent WTO status, which technically happened under W, but was really a bipartisan initiative. It's had the largest impact on the American economy.

The second biggest economic policy decision wasn't really a decision, but rather a non-decision, which was to allow mortgage securities to be essentially unregulated. This is much more of a W decision, and while it didn't CAUSE the 2008 recession, it probably exacerbated it into a much more severe problem.

The third biggest economic policy decision was probably QE + TARP, but I think we still don't know exactly how well it worked and what the counterfactual would have been.

I think two out of the three would have clearly been done in any administration and represented consensus American economic policy. The decision to not regulate mortgage securities is more difficult to evaluate, and approximately 3/4 administrations (100% of R and 50% of D) would have done it in my estimation.

We are most likely going to have a recession in the next year or so that will be heavily blamed on Trump's trade policies, but those will have only contributed approximately 5-10% to the overall causation, which is really just a standard business cycle recession.
   1153. BrianBrianson Posted: October 12, 2018 at 03:50 PM (#5765269)
Picking specific days is probably a mistake. Clinton's lead bounced between +2 and +7 in polls from June - November (on 538's average, anyhow). She his +2 on election day, so at the low end of what you should've been expecting. 538's odds of her winning bounced between 50% and 80% over that period, ending around 70% - so, if you were looking at the data in aggregate, you knew what was going on months out, and the actual outcome was a ~1 sigma outlier - i.e., normal and to be expected.

I mean, there are elections where things have really changed during the campaign. The canonical example in my lifetime may well be the '93 federal election in Canada (and HOW!). But the polling data for the 2016 presidential election was telling you what was going on ~half a year in advance.
   1154. Mellow Mouse, Benevolent Space Tyrant Posted: October 12, 2018 at 03:51 PM (#5765270)
Polls are useful for internal strategy but up until a couple weeks before the election they aren't that predictive, at least for competitive districts without huge registration differences.


I don't think this is actually true. Yes, the polls were wrong regarding Hillary. It happens. Over time polls have shown to be fairly predictive. I mean you guys now history exists and we have data, right?
   1155. This is going to be state of the art wall Posted: October 12, 2018 at 03:57 PM (#5765274)
I don't think this is actually true. Yes, the polls were wrong regarding Hillary. It happens. Over time polls have shown to be fairly predictive. I mean you guys now history exists and we have data, right?


Most people's exposure to political forecasting using data is 538 calling the Obama wins and then having Hillary at what like 70/30? The EC lends itself much better to this style of forecasting and poll aggregating and yet Trump did win and 538 was obviously not using old polls for their election eve projection.

Polls can't forecast turnout. In more granular races contested at state and district levels turnout is by far the biggest variable.
   1156. David Nieporent (now, with children) Posted: October 12, 2018 at 04:03 PM (#5765278)
I don't think this is actually true. Yes, the polls were wrong regarding Hillary.
No, they weren’t.
   1157. PepTech, the Legendary Posted: October 12, 2018 at 04:11 PM (#5765284)
Yes, the polls were wrong regarding Hillary.

No, they weren’t.
What were the final polls like in Wisconsin, Michigan, Ohio, Pennsylvania, North Carolina and Florida? Those are the only ones that really mattered. Were they wrong?

I don't know, and I'm too lazy to look, but I suspect the northern four were at least leaning blue, and NC/FL were toss-up or leaning red. Am I even close?
   1158. BrianBrianson Posted: October 12, 2018 at 04:16 PM (#5765285)
The outcome was consistent with the polls. The smaller an area you're asking about, the more uncertain you are.

Or - if you flip two coins, the most likely outcome is one heads, one tails. It happens 50% of the time. So, my coin flipping model tells you that's the most likely outcome. If you flip two coins and get two heads, is my model wrong? It says that should only happen 25% of the time, and a heads an a tails was more likely.
   1159. PreservedFish Posted: October 12, 2018 at 04:26 PM (#5765291)
I think what Nieps means to say is that a poll cannot be wrong (unless perhaps the data has been massaged or bungled). A poll is a snapshot. That snapshot can be misleading, but that's probably our fault for spinning it this way or that, not the poll's fault.
   1160. JL72 Posted: October 12, 2018 at 04:30 PM (#5765296)
Picking specific days is probably a mistake. Clinton's lead bounced between +2 and +7 in polls from June - November (on 538's average, anyhow). She his +2 on election day, so at the low end of what you should've been expecting. 538's odds of her winning bounced between 50% and 80% over that period, ending around 70% - so, if you were looking at the data in aggregate, you knew what was going on months out, and the actual outcome was a ~1 sigma outlier - i.e., normal and to be expected.


What I found interesting in the 2018 polling was how elastic the results were (for lack of a better term). Different events ("basket of deplorables" or "grab them by the p*ssy") would send the polls one way or the other, but after a couple of weeks they would settle back to about their steady state.

One question I would love to have answered is how the results would have changed if there had been a couple more weeks between the last Comey missive and the election, thus allowing the polls to move back to that steady state. Would people have still voted as they did (telling us that the polls were always off because people were answering like they felt they were supposed to), or would they have voted more like the poll results (telling us that a sizable group was holding their nose and voting for the person they felt was least objectionable at the time).
   1161. BrianBrianson Posted: October 12, 2018 at 04:37 PM (#5765303)
Well, sort of. A lot of people seemed to take "Clinton is a 70-30 favourite to win" to mean "It's unthinkable Clinton will lose" rather than "Clinton losing is about as probably as flipping two coins and having them both come up heads". But the polls were clear that they meant the latter, while people claiming the polls were wrong or misleading where somehow thinking they meant the former.
   1162. David Nieporent (now, with children) Posted: October 12, 2018 at 04:37 PM (#5765304)
I agree with both 1158 and 1159, and adopt both sets of arguments, but my point was actually simpler: the polls showed Hillary was several points ahead of him... and she was.
   1163. This is going to be state of the art wall Posted: October 12, 2018 at 04:39 PM (#5765306)
Well, sort of. A lot of people seemed to take "Clinton is a 70-30 favourite to win" to mean "It's unthinkable Clinton will lose" rather than "Clinton losing is about as probably as flipping two coins and having them both come up heads". But the polls were clear that they meant the latter, while people claiming the polls were wrong or misleading where somehow thinking they meant the former.


that would be ray.
   1164. Srul Itza Posted: October 12, 2018 at 04:40 PM (#5765308)
What were the final polls like in Wisconsin, Michigan, Ohio, Pennsylvania, North Carolina and Florida? Those are the only ones that really mattered. Were they wrong?


I think that may have been the problem: A lack of regular and updated polls in those states, and in particular, polls that concentrated on likely voters.

Given the fact that she did, indeed, win the popular vote, due to her huge lead in states like California, the overall polling showing her in the lead was not as inaccurate as all that (which may be DMN's point [EDIT --Did not see his confirmation before I posted]). But National polls do not model the Electoral College.
   1165. homerwannabee Posted: October 12, 2018 at 04:42 PM (#5765309)
Hillary was up by 1.9 points in Pennsylvania, +3.4 points in Michigan, and the real big screw up Wisconsin she was 6.5 points. She only lost three states where she had the lead in the polling. Election day she was supposed to win 272 to 266. Of course the final result before the faithless voters was 232 to 306.
   1166. ERROR---Jolly Old St. Nick Posted: October 12, 2018 at 04:43 PM (#5765310)
What were the final polls like in Wisconsin, Michigan, Ohio, Pennsylvania, North Carolina and Florida? Those are the only ones that really mattered. Were they wrong?

I don't know, and I'm too lazy to look, but I suspect the northern four were at least leaning blue, and NC/FL were toss-up or leaning red. Am I even close?


You're right about all of them but Ohio, where the final RCP average had Trump ahead by 2.2%. NC and Florida were leaning red (barely), while Michigan, Wisconsin and Pennsylvania had Clinton ahead. OTOH the final national poll average was almost spot-on.

EDIT: coke to homerwannabee
   1167. This is going to be state of the art wall Posted: October 12, 2018 at 04:47 PM (#5765314)
Hillary was up by 1.9 points in Pennsylvania, +3.4 points in Michigan, and the real big screw up Wisconsin she was 6.5 points. She only lost three states where she had the lead in the polling. Election day she was supposed to win 272 to 266. Of course the final result before the faithless voters was 232 to 306.


I think the unions told her the union men were going to vote for her and their wives would too. that didn't ####### happen. I'm still slogging through "Shattered" (haven't made it to E-Day yet just finished the primaries) so I'll be curious as to if that was the case. Why the hell she paid so little attention to those states is a great mystery.
   1168. BrianBrianson Posted: October 12, 2018 at 04:50 PM (#5765315)
and the real big screw up Wisconsin she was 6.5 points.


538 had her up 5.3 plus or minus (roughly) 5.3 in Wisconsin. She lost by ~0.8%. So, the polls were pretty much right. Like flipping a coin and getting three heads in a row. Not that likely, but you don't start ranting about how the coin is fixed.
   1169. This is going to be state of the art wall Posted: October 12, 2018 at 04:51 PM (#5765316)
about how the coin is fixed


rigged.
   1170. BrianBrianson Posted: October 12, 2018 at 04:52 PM (#5765317)
Okay, let's try a simpler tack.

POLLS ARE A STATISTICAL SAMPLING

Who's having trouble following that?
   1171. Ray (CTL) Posted: October 12, 2018 at 05:03 PM (#5765324)
I think one major thing people failed to understand about the election -- in a sea of them -- was the basic math of it. Even if Hillary had a 70% chance to win, the chance of her losing was, by definition, still equivalent to the chance that Rafael Devers will get on base. There are plenty of things to say around the edges of this narrow issue, and in particular about the limits of using the popular vote as a proxy in an Electoral College contest, but that was the core point, and the fundamental misunderstanding.

Is anyone shocked to his or her fainting couch when Rafael Devers reaches base? Of course not. Yet the O'Science Party was shocked here.
   1172. Ray (CTL) Posted: October 12, 2018 at 05:05 PM (#5765327)
Well, sort of. A lot of people seemed to take "Clinton is a 70-30 favourite to win" to mean "It's unthinkable Clinton will lose" rather than "Clinton losing is about as probably as flipping two coins and having them both come up heads". But the polls were clear that they meant the latter, while people claiming the polls were wrong or misleading where somehow thinking they meant the former.

that would be ray.


It's so interesting to me that the virtue signalers like Art Wall are the ones who feel the need to tell lies about the opinions of people they disagree with.
   1173. BrianBrianson Posted: October 12, 2018 at 05:05 PM (#5765328)
Yet the O'Science Party was shocked here.


Damn, Ray, everything you said was factual, eloquent, and well put .... right up until this last sentence.
   1174. perros Posted: October 12, 2018 at 05:06 PM (#5765329)
I think one major thing people failed to understand about the election -- in a sea of them -- was the basic math of it.


Leave it to Ray to turn Trump's chances of a hit from Rich Hill to Rafael Devers.

####### dumbass.
   1175. perros Posted: October 12, 2018 at 05:10 PM (#5765333)
For Art Wall:

Why you should listen to Trump's Rallies



So why I am writing about this? Why spend nearly seven bleary-eyed hours over six rallies listening to the President? That’s six full renditions of Lee Greenwood’s “I’m Proud to Be an American,” six times hearing Trump rip off Churchill’s “never surrender” speech, six times listening to him insult “low I.Q.” Maxine Waters and “Crooked Hillary” and “Crazy Bernie” and, his new favorite, “Da Nang Dick” Blumenthal.

Watching hours of Trump at his rallies, it’s easy to sympathize with the desire to ignore them. John Dean tweeted a picture of the crowd waiting in line for the Erie rally and derided it as a “meaningless show.” For supporters, it’s hyperbole, just rhetoric, entertainment, part of the unvarnished appeal; for opponents, it’s old news painful to watch, maybe, but inconsequential, narrow-casting to his base. One of the reasons we tune out is because views of Trump are so fixed. Look at the Presidential approval ratings, and “you would think it’s been a pretty boring couple years,” as Amy Walter, the Cook Political Report editor, likes to put it. Trump’s ratings have barely budged, no matter the day’s outrage or the nutty things he tells his followers: the same range of thirty-eight to forty-three per cent of Americans approve of him, according to the Real Clear Politics polling average, and the same majority of fifty to fifty-three per cent disapprove of him, as has been the case since the early weeks of his Administration.

Much of the coverage of these events tends to be theatre criticism, or news stories about a single inflammatory line or two, rating Trump’s performance or puzzling over the appeal to his followers. But what the President of the United States is actually saying is extraordinary, regardless of whether the television cameras are carrying it live. It’s not just the whoppers or the particular outrage riffs that do get covered, either. It’s the hate, and the sense of actual menace that the President is trying to convey to his supporters. Democrats aren’t just wrong in the manner of traditional partisan differences; they are scary, bad, evil, radical, dangerous. Trump and Trump alone stands between his audiences and disaster.

I listen because I think we are making a mistake by dismissing him, by pretending the words of the most powerful man in the world are meaningless. They do have consequences. They are many, and they are worrisome. In what he says to the world, the President is, as Ed Luce wrote in the Financial Times this week, “creating the space to do things which were recently unthinkable.” It’s not a reality show; it’s real.
   1176. perros Posted: October 12, 2018 at 05:24 PM (#5765339)
Glaaser's insight comes in listening, getting past the spectacle of it all. And while I agree Trump's rhetoric is dangerous -- one is reminded of Hitler's rallies though she never mentions them -- it's tv's dismissal of them that's the key point.

At the heart of my long opposition to temevision was stumbling upon the realization that it's the combination of sound + inage that gives gelevision ita propagandistic power. Turn down the sound and the crass manipulation of commercials becomes apparent. In a way they distract the mind to let the words sink in unguarded.

It's more complex than I can explain. Be interesting to read Jack Gladney's take on how the rise of the talkies coincided with thr rise of Adolf. Maybe good to go back and watch Lang's Dr. Marbuse trilogy.
   1177. perros Posted: October 12, 2018 at 05:25 PM (#5765340)
Love the pic of the back of Trump's head for the article. Demystifies the world's worst haircut.
   1178. perros Posted: October 12, 2018 at 05:31 PM (#5765343)
Why the hell she paid so little attention to those states is a great mystery.


Beyond the fact she's godawful at politics, it's hard to fathom how her smart kids got it ao wrong. I think they assumed she'd get all of Obama's votes plus WOMEN!

She's the anti-Beto. Could you see her bothering with any working class folks, let alone a w1-year-old Mexican line cook?
   1179. perros Posted: October 12, 2018 at 05:36 PM (#5765345)
A poll is a snapshot. That snapshot can be misleading, but that's probably our fault for spinning it this way or that, not the poll's fault.


So I can't blame Nate Silver for mistaking Hillary for Obama, too? I'll let Art fill me in, but I believe Bill was incredibly frustrated at the campaign for it's lack of touch on the ground. Politics isn't mere polling.
   1180. This is going to be state of the art wall Posted: October 12, 2018 at 05:36 PM (#5765346)
Glaaser's insight comes in listening, getting past the spectacle of it all. And while I agree Trump's rhetoric is dangerous -- one is reminded of Hitler's rallies though she never mentions them -- it's tv's dismissal of them that's the key point.


The overwhelming sense I get watching the audience at his rallies is that Trump makes them feel important and powerful while simultaneously playing on their worst fears about savages at the gate. There is a reason that he finishes every rally with the bit about "our heritage, forged on the frontier by strong pioneer men and women..." that hearkens back to the time when white families, of humble means, turned the wilderness into homesteads and pacified the Indians.

It's the worst of American myth making, the kind that Cormac so beautifully illustrates and then distills down to its essence in the Judge's monologues.

So I can't blame Nate Silver for mistaking Hillary for Obama, too? I'll let Art fill me in, but I believe Bill was incredibly frustrated at the campaign for it's lack of touch on the ground. Politics isn't mere polling.


One of the main takeaways from the chapters dealing with the primary in "Shattered" is that Bill was not properly utilized. He was ready to pound the pavement and for the most part Hillary and her team wanted him in the background. They felt he was at best a neutral influence.
   1181. perros Posted: October 12, 2018 at 05:37 PM (#5765347)
Five o'clock Friday. I should probably have spent the past half hour drinking.
   1182. perros Posted: October 12, 2018 at 05:38 PM (#5765348)

It's the worst of American myth making, the kind that Cormac so beautifully illustrates and then distills down to its essence in the Judge's monologues.


Gonna read it again this fall.

Honestly, I can't bear to watch those rallies.
   1183. Never Give an Inge (Dave) Posted: October 12, 2018 at 05:39 PM (#5765349)
I think what Nieps means to say is that a poll cannot be wrong (unless perhaps the data has been massaged or bungled). A poll is a snapshot. That snapshot can be misleading, but that's probably our fault for spinning it this way or that, not the poll's fault.

To use a baseball analogy, if you have a "true talent" .300 hitter, he may hit .290 one year and .310 the next year, and those are both within the expected range of outcomes given the sample size. We use a player's historical stats to estimate their true talent level and to try to forecast their future performance.

I think some people see the poll results (or the resulting 538 forecasts) as the election's "true talent" and the actual election results as its "single-season batting average". But the opposite is much closer to the truth. The election result is basically the true talent (or a *very* long season that we can basically take to be the true talent, within a very small margin of error). And the polls are basically short seasons that we're trying to use to estimate the true talent and forecast that very long season.*

With enough polls, we can get a larger sample size and a more accurate estimate of the election results. But this isn't like a baseball game, where if you replayed it 100 times you might get a different result 40% of the time (maybe that was true of the 2000 election given how close FL was, but not 2016).

The "true talent" of the electorate was a Trump victory (under our Electoral College system; he obviously didn't win a plurality of the vote) . Replay it 100 times and maybe you get a Hillary victory a few times, but it's not going to be 71. When 538 said Hillary had a 71% chance of winning the election, they were saying they forecast Hillary to win, but there's a 29% chance they're wrong due to some sort of systematic error in the polling or in their adjustments.

The polls are just collections of data -- they can't be "wrong" -- but they can be unrepresentative samples. When they are published, the results are weighted based on R/D/I splits to try to give an accurate portrayal of the electorate. That is one point where they can go "wrong". And then 538 takes that data and aggregates it and adjusts it, and that's another place where folks can go "wrong".

* in both cases, this is complicated by the fact that "true talent" (a batter's ability or people's voting preferences) can change, sometimes very quickly. But I don't think that's all that happened in 2016.

I agree with both 1158 and 1159, and adopt both sets of arguments, but my point was actually simpler: the polls showed Hillary was several points ahead of him... and she was.

538 had a couple of things wrong. They were very close on Clinton's overall level of support (48.5% forecast vs. 48.2% actual) but were low on Trump (44.9% vs. 46.1%), largely due to overestimating the third-party candidates (6.6% vs. 5.7%).

More importantly, they also had the overall distribution of that support wrong. Hillary won by bigger margins than expected in CA and NY; she lost by a much bigger margin than expected in OH; and she obviously lost some states that she was expected to win (and still "should have" won if the polls were only off by 1.2% on Trump). Among other "errors".

I'm not sure why that is, but they would be remiss not to update their models based on what they've learned this time around.

EDITed for clarity.
   1184. BrianBrianson Posted: October 12, 2018 at 05:54 PM (#5765354)
538 had a couple of things wrong. They were very close on Clinton's overall level of support (48.5% forecast vs. 48.2% actual) but were low on Trump (44.9% vs. 46.1%), largely due to overestimating the third-party candidates (6.6% vs. 5.7%).


Error bars! Jeez, how are we having a conversation where only Ray and David are making any kind of sense!

I'm not sure why that is, but they would be remiss not to update their models based on what they've learned this time around.


This would be an incredibly stupid thing to do. Overfitting your data like that is a kindergarten math mistake. A third of your predictions should be off by 1 sigma. A twentieth (so 2-3 states) should be off by two. Ohio was off by about 1 sigma on 538 - that's perfectly normal.
   1185. Howie Menckel Posted: October 12, 2018 at 05:57 PM (#5765355)
One question I would love to have answered is how the results would have changed if there had been a couple more weeks between the last Comey missive and the election, thus allowing the polls to move back to that steady state.

I'm trying to conjure up an image of a single voter who was set to vote for Hillary, then flipped to Trump or stayed home because of Comey's statement, yet would have flipflopped back again to showing up and voting for Hillary if only a couple of more weeks had passed.

geesh, I can be lampooned as the waffliest of the waffly, so if even I can't imagine such a person...
:)
   1186. Lassus Posted: October 12, 2018 at 06:01 PM (#5765357)
@amandwallwin - The real #HimToo is that men are more likely to be sexually assaulted than falsely accused.
   1187. This is going to be state of the art wall Posted: October 12, 2018 at 06:04 PM (#5765358)
I'm trying to conjure up an image of a single voter who was set to vote for Hillary, then flipped to Trump or stayed home because of Comey's statement, yet would have flipflopped back again to showing up and voting for Hillary if only a couple of more weeks had passed.


You have to be convinced of human fickleness?
   1188. Never Give an Inge (Dave) Posted: October 12, 2018 at 06:35 PM (#5765367)

Error bars! Jeez, how are we having a conversation where only Ray and David are making any kind of sense!

This would be an incredibly stupid thing to do. Overfitting your data like that is a kindergarten math mistake. A third of your predictions should be off by 1 sigma. A twentieth (so 2-3 states) should be off by two. Ohio was off by about 1 sigma on 538 - that's perfectly normal.


"Update their models" was probably not a good way to put it. But the amount/types of data available and the tools used for collecting it are constantly changing. We should certainly look back at whether those tools can be recalibrated or refined based on what we know about the actual results.
   1189. Chicago Joe Posted: October 12, 2018 at 06:45 PM (#5765369)
Zonk: Buy a 2-flat or even a three if you can find a good one. No mortgage, really, and depreciation helps with the income tax consequences.
   1190. Gonfalon Bubble Posted: October 12, 2018 at 06:46 PM (#5765370)
Preserved Fish:
I find the endless poll brandishing by YC and GB to be very strange behavior. "My team is winning." "No, my team is winning!"

David Nieporent:
What’s really strange about it is that neither is correct. Polls are not actual votes; the score remains tied at 0-0 until November 6. Touting poll results is like touting the betting line for a sporting event.

Davo:
The craziest thing to me is that, you know...the guy behind 538.com is a baseball nerd just like us! And he explains exactly how we are supposed to interpret the polls (“Ignore outliers, don’t obsess over day-to-day fluctuations, just examine broad trends”) in a way that makes perfect sense.....but then we ignore it all in favor of “Holy ####, Blue Team is up 11 now, last week they were down 2!”


I won't speak to Clapper's motivations, but it's interesting that reactions like the above are coming after my posts, which were specifically written to call out "some folks here" for ignoring larger context and eschewing a broad brush. If you eagerly swallow a single result at full face value, then you gotta swallow the next result that you don't like. But with people taking that as "my poll can beat up your poll" or "never wrestle with a pig," to my surprise I wasn't being painstakingly pedantic enough. I guess I should have added, "So don't swallow the first one!"

I mean, combining the facts that ten of the eleven polls in a House race are in near-alignment, and that a partisan has seized upon the lone poll that contains good GOP news (while, amazingly, dismissing "outlier polls")... was this unclear?

Polls provide information. Fundraising totals provide information. Electoral history provides information. The Dow Jones provides information. Even Trump rallies provide information. It's entirely a matter of how far you take that information, and how you apply it to hunches and assumptions and speculations and jaw movements, that's either more or less kosher.
   1191. Gonfalon Bubble Posted: October 12, 2018 at 06:47 PM (#5765371)
Clapper, #1106:
Not a poll ... Beto O'Rourke raised $38.1 million in last three months, smashing all records in bid to oust Ted Cruz
What's O'Rourke getting for that money? After being listed as a Toss-Up in September, the race is back in the GOP column, and Cruz now has a 7% lead in the RCP poll average. It's nice that the vendors and consultants will apparently get paid, but that's not a lot of return on investment. But perhaps I'm wrong - send him more $$.


O'Rourke? A national profile but not a Senate seat, presumably.

But for the twentieth time, other GOP candidates across the country aren't getting the infusions of Texas money they were counting on, because it's desperately needed for Ted Cruz and other unexpectedly at-risk Texas Representatives. O'Rourke forcing even a successful Republican triage effort of this magnitude is having direct effect on elections in Virginia and Michigan and Pennsylvania and so forth. Context!
   1192. Gonfalon Bubble Posted: October 12, 2018 at 06:47 PM (#5765372)
PepTech, #1127:
With all the unrest in the stock market, good thing we can point to the VIX for confidence in stabil... oh.


#Failed Barack Obama never got the Dow Jones to hit 25,000. But now, Donald Trump's singlehandedly done it TWICE!
   1193. BrianBrianson Posted: October 12, 2018 at 07:02 PM (#5765374)
"Update their models" was probably not a good way to put it. But the amount/types of data available and the tools used for collecting it are constantly changing. We should certainly look back at whether those tools can be recalibrated or refined based on what we know about the actual results.


Nominally, of course this is true. Another election is more data, so you can train your model a little more. But the model performed quite well - I see no reason to think it's possible to make a more accurate model given the measurement uncertainties. Indeed, that 538 has three models for the midterms shows he's interested in how to weight data - but they all give pretty much the same answer.

Really, the problem is people are freaking out than events with 20% or 30% chances of occurring occurred. Well, those are perfectly normal things to happen.
   1194. David Nieporent (now, with children) Posted: October 12, 2018 at 07:07 PM (#5765376)

This would be an incredibly stupid thing to do. Overfitting your data like that is a kindergarten math mistake. A third of your predictions should be off by 1 sigma. A twentieth (so 2-3 states) should be off by two. Ohio was off by about 1 sigma on 538 - that's perfectly normal.
But another issue: because states are winner take all (yeah, except Maine and Nebraska), we only "care" about errors around the 50% mark. It doesn't matter how much Ohio was off by, because it didn't change the outcome there. Whereas the error in WI/MI/PA flipped those states.
   1195. This is going to be state of the art wall Posted: October 12, 2018 at 07:10 PM (#5765378)
   1196. The Yankee Clapper Posted: October 12, 2018 at 07:22 PM (#5765379)
Beyond the fact she's godawful at politics, it's hard to fathom how her smart kids got it ao wrong. I think they assumed she'd get all of Obama's votes plus WOMEN!

That was pretty much the attitude of a lot of people here, too.
   1197. Count Vorror Rairol Mencoon (CoB) Posted: October 12, 2018 at 07:22 PM (#5765380)
Gonna have to provide some details for those of us who aren't going to waste our time or give YouTube a hit.
   1198. This is going to be state of the art wall Posted: October 12, 2018 at 07:35 PM (#5765382)
Gonna have to provide some details for those of us who aren't going to waste our time or give YouTube a hit.


Go down to your local mental health institution and ask to talk to one of the residents, offer a cig perhaps. Then just listen to them talk for about 30 minutes.
   1199. perros Posted: October 12, 2018 at 07:43 PM (#5765384)
I never watch the weather forecast, either. If it's a big thing like a hurricane -- or two inches of snow -- people freak out publicly well in advance of the storm.
   1200. Davo and his Moose Tacos Posted: October 12, 2018 at 07:49 PM (#5765385)
Pass
Page 12 of 16 pages ‹ First  < 10 11 12 13 14 >  Last ›

You must be Registered and Logged In to post comments.

 

 

<< Back to main

News

All News | Prime News

Old-School Newsstand


BBTF Partner

Dynasty League Baseball

Support BBTF

donate

Thanks to
Francis
for his generous support.

Bookmarks

You must be logged in to view your Bookmarks.

Hot Topics

NewsblogOTP 2018 October 15: The shift in focus from sport to politics
(1455 - 12:45pm, Oct 21)
Last: perros

NewsblogMookie Betts to second base? Alex Cora isn’t ruling it out | Boston.com
(11 - 12:44pm, Oct 21)
Last: villageidiom

NewsblogMLB must fix glaring problem that ruined an all-time classic
(99 - 12:39pm, Oct 21)
Last: ERROR---Jolly Old St. Nick

NewsblogOT - NBA Thread (2018-19 season kickoff edition)
(804 - 12:38pm, Oct 21)
Last: NJ in NY (Now with Toddler!)

NewsblogMLB -- Manny Machado, Yasiel Puig embrace their villain roles all the way to the World Series
(3 - 12:28pm, Oct 21)
Last: Jose is an Absurd Force of Nature

Gonfalon CubsNow what?
(115 - 12:21pm, Oct 21)
Last: Moses Taylor, aka Hambone Fakenameington

NewsblogLEAGUE CHAMPION SERIES OMNICHATTER! for the 2018 Playoffs!
(2616 - 10:43am, Oct 21)
Last: perros

NewsblogWhat It Took to Write About Baseball as a Woman
(26 - 10:27am, Oct 21)
Last: AndrewJ

NewsblogThe Brewers are becoming more and more positionless on defense | SI.com
(23 - 10:26am, Oct 21)
Last: Greg Pope

NewsblogFor Dave Dombrowski, Another World Series on the Path to the Hall of Fame - The New York Times
(3 - 7:43am, Oct 21)
Last: Der-K: at 10% emotional investment

NewsblogOT - October 2018 College Football thread
(156 - 11:42pm, Oct 20)
Last: Howie Menckel

NewsblogWhy The Dodgers' WS Odds are So High
(27 - 11:28pm, Oct 20)
Last: Baldrick

NewsblogCatch-All Pop Culture Extravaganza (October 2018)
(544 - 10:57pm, Oct 20)
Last: Lassus

NewsblogBest 2018-19 Hot Stove value may be in trade
(2 - 6:44pm, Oct 20)
Last: cardsfanboy

NewsblogAstros' Jose Altuve underwent surgery to repair right knee injury
(1 - 6:01pm, Oct 20)
Last: Count Vorror Rairol Mencoon (CoB)

Page rendered in 0.7692 seconds
46 querie(s) executed