Because here it's treated more like training for a vocation and there it's more education for its own sake (i.e., they take a lot of courses outside their major). I prefer their system personally.Quote:
Originally Posted by Keith
Printable View
Because here it's treated more like training for a vocation and there it's more education for its own sake (i.e., they take a lot of courses outside their major). I prefer their system personally.Quote:
Originally Posted by Keith
This sounds like a groundbreaking argument for a dissertation/study, and something that wouldn't be all that difficult to verify with research. Maybe your armchair brilliance is unmatched by anyone in the field ... or maybe you're just coming up with slapshod theories for why the polls with your favored results are better than other polls. I don't know, just spitballing here.Quote:
Originally Posted by OngBonga
Not sure why you're so obsessed with Brexit upset. The polling average was well within the margin of error, especially considering they were consistently reporting double digits for Undecided voters. Seems a little over-reactionary to completely blow up your idea of the value of polls over that one result.
I'm probably talking shit, but I'd like to think there's some basis to that. idk, it's just in my very small sample of facebook friends, most people who were talking about Brexit were remainers, yet most people I spoke to in private about the issue said they were voting leave. It's not an obsession by any means, merely an observation that could be heavily flawed and misinterpreted by my stupid stonedness.
Certainly, I know of people who withdrew from public debate because of accusations of racism. That was a card that remainers were using a lot, and I sense the same is happening in the States with Trump. That might or might not have an effect on public debate.
Of course, withdrawing from public debate might or might not even have any impact of people's tendancy to be honest with pollsters.
So it's probably bollocks. But... considering remain was big favourite with the bookies, it was something of an upset as far as I'm concerned. I thought we'd vote to stay.
Yeah but we understand how odds work right? We also understand how bookies work?
Makes me laugh when you hear omg bookies got it wrong, they're not there to try and predict the result (being able to would be very helpful obvs) they are there to try and make money from an event.
Ya, I don't understand why anyone would be shy about telling the truth to someone in a poll.
btw, Hillary by the end of the campaign
https://www.youtube.com/watch?v=uAfEyR4_VZw
Well I keep saying bookies, but I'm actually referring to betfair prices, a betting exchange, which doesn't reflect the odds the bookies set, instead it reflects where the money is going (a reflection of the betting public's opinion). The bookies will usually have the prices a little less generous than betfair.
Betfair provides us with a measure for what a reasonably large sample size of people think will be the outcome of an event. The bigger the market in terms of people, the more accurate (on average) the odds are relative to probability.
In the case of the referendum, it was a very large market. Remain was odds on favourite because the vast majority of money was being staked on remain.
The bookies are irrelevant really, I should stop saying bookies and be more clear about what I mean.
I wouldn't, but that's probably got nothing to do with my political lean. I just don't like being stopped by people in the street. I'd tell them to get a proper job, before continuing on my way to the jobcentre.Quote:
Ya, I don't understand why anyone would be shy about telling the truth to someone in a poll.
That's how bookies set their odds too though.Quote:
Originally Posted by OngBonga
Same here (except for the jobcentre part), but it's anonymous and it's not like they're asking them if they're a pedo or summat.Quote:
Originally Posted by OngBonga
Of course.Quote:
Originally Posted by Poopadoop
But betfair is a more accurate measure.
How so? I'm not arguing it isn't I just don't understand.Quote:
Originally Posted by OngBonga
I'm stoned and not being very accurate with my language. I do understand how bookies work... their odds are nearly always tighter than betfair because their odds are what they deem to be probability + a margin of error in their favour for profit. Betfair is not a profit-driven company, it's a mass market of punters. Of course the average punter is trying to make money, but he is also by nature not very good at it because he is gambling. Betfair odds get close to probability because it's essentially a mass poll, with the added honesty factor of people putting money on their opinions.
It's a useful tool, and shouldn't be ignored because it's a bunch of gamblers. Their opinions are not insiginificant.
Because there's no profit factor, it's pure betting. Betfair cuts out the middle man that is the bookmaker, which means it's a direct reflection of where the money is going. The bookies will change their odds as the bets come in, but the change in odds might not be an accurate measure of the amount of money that is being staked.Quote:
Originally Posted by Poopadoop
In a betting exchange, especially a large market where there are bets waiting to be matched at various odds, the current odds are a very accurate measure of how much money is being staked relative to the other option(s).
The bookies will be able to tell you where most of the money is going, just not as accurately, not unless you know how much the bookmaker in question is adjusting the price to account for profit.
I don't think most bookies pad the odds I think they charge a fee for their services. Could be wrong about that, I don't know a lot about bookies i admit.
Well the bookmakers are nearly always less generous than betfair. On the occasion they are more generous, there's money to be made.Quote:
Originally Posted by Poopadoop
Eh, I don't understand what you mean by 'generous'. Assuming it's an either/or outcome (i.e., either A happens or B happens), then if you're adjusting the odds to be more generous to one side (say the people who bet on A), you're automatically being less generous to the other side (the B bettors).
Maybe I'm missing something tho.
As an example, let's look at Murray vs Djokovic... it's 2-2 going into deciding set, nothing between them.
Odds on betfair will very likely be 2.0 for both, let's assume they are... if you bet on both of them, you'd break even... not taking into account that you owe betfair a small % for their service.
Now go to the bookies. You're not going to get 2.0 for both, you'd see perhaps 1.9 for both... that 0.1 is profit for the bookmaker. That isn't there on betfair, or maybe it is but to a much smaller degree.
Generous in terms of odds.Quote:
Originally Posted by Poopadoop
2:1 is less generous than 3:1.
Ok ya that makes sense. So in essence the bookies' fee is set into the odds. I was thinking more along the lines of sports betting where there's often an over/under kind of thing going on.
Still don't understand how that applies to the Brexit betting though. If the betfair odds for Brexit and Brextay were both higher than, ah never mind doesn't matter.
Yeah what started out as an observation relating to our referendum has ended up with us talking shit about bookmakers.
I'm smoking nice weed, what's your excuse?
I'm talking to you.
I'd like to see evidence of this.Quote:
Originally Posted by OngBonga
Are their customers smarter than other customers?
I think what he means is they don't pad their odds, so it's a more accurate reflection of how people are actually betting.
The more interesting question to me is who bets on a referendum? It seems infinitely less fun than betting on a sporting event or other game.
Yeah I mean betfair don't have "customers" in the same sense bookmakers do.
Betfair's customers are betting against each other.
A bookmaker's customers are betting against the bookmaker.
That means the bookmaker must set correct odds to profit, while betfair merely needs to exist.
My friends dad makes a nice bit of money putting up stupid odds on popular tv shows where people bet on those they really like/dislike so it's a very exploitable market.Quote:
Originally Posted by Poopadoop
Or it is until betfair start taking much bigger cuts of your money.
People who might want to mitigate losses as a result of us staying/leaving, or people who think they know something that others don't? idk, I'm guessing.Quote:
Originally Posted by Poopadoop
I had no interest in betting on it. I was only interested in the odds from a polling pov.
Ya betting on who lasts longer on Big Brother or whatever. Jesus Christ.Quote:
Originally Posted by ImSavy
Sounds like your friend's dad found a good racket though. Those people must be fucking morons.
Those were my guesses too, along with pathological gamblers who bet on everything under the sun.Quote:
Originally Posted by OngBonga
It's one of the few gambling stories I've been told where I've actually thought yeah that sounds like a legit clever way to make money. We're probably talking a couple hundred quid over the course of a series though, nothing special but a nice bit of spending money.Quote:
Originally Posted by Poopadoop
In the summer of 2014 I bet at 20:1 that Justin Bieber would be admitted to rehab before the end of 2015.
Lost that bet. Dude probably should have gone though imo.
I once tested a funny parlay to see what the odds were.
Bodog offered a bet on the royal baby specs a few years ago when the princess or whatever was preggerz.
A black, ginger, fat baby named Obama paid like 50,000:1
Does seem like a long shot...I might have taken 50k:1 tho just for the hell of it.Quote:
Originally Posted by BankItDrew
The parallels between Brexit and 2016 US Presidential election are important. There were very high undecideds in Brexit, and they broke for Leave. Remain was "incumbency" yet didn't poll as well as the milieu of establishment does if it's favored. The same is true in this election. There are very high undecideds (and third party, which are also mostly undecideds). The "incumbency" is intensely unpopular, much moreso than Remain was. In the last month of Trump entering 3rd Act mode (acting presidential), undecideds have broken so remarkably for him that he gained ~10 points on Clinton.
If Trump doesn't blow past her and landslide this thing, we'll have rigging to thank.
Which poll are you looking at now Wuf? The USC numbers are about the same as a month ago.Quote:
Originally Posted by wufwugy
I could be wrong in that it's closer to 2 months. It was shortly after the Khan fiasco that Trump entered full presidential mode and he gradually began closing in on Clinton. It was well documented in real time, which is where my memory comes from. If you want data evidence though, you can maybe see the trend on 538. Shortly after his convention bounce was the Khan thing, where he plummeted. His rise has been steady since. Granted 538 shows their win percentage instead of the polls, perhaps unless you look around
http://projects.fivethirtyeight.com/...ex_cid=rrpromo
iirc the usc numbers always were different than the other polls. really im just referring to aggregation of the polls over the last ~6 weeks. he was way behind but then he closed. this caused clinton to finally give a faux presser and other things like the "why am i not 50 points ahead" comment. she thought she had it in the bag 6 weeks ago. now she's scrambling. it's becuase he closed ~10 poionts
i dont think im gonna try to find the data to show it, but i remember quite well how over the last ~6 weeks he went from polling terribly in ohio and florida to leading there now. same w/ nevada. the electoral map gives clinton only the slimmest of victories right now, according to the polls
Ya, that's what I thought you meant. ;)Quote:
Originally Posted by wufwugy
wait wtf i never said 2 points. i said 2 months. you tricked me!
You're getting smarter all the time. ;)Quote:
Originally Posted by wufwugy
It is an interesting race in a lot of places though. Ohio and Florida for example. Could go either way.
Debate should be fun. I wonder how the hologram of Hilllary will do.
Hillogram? Cartogram of Crooked? Rapin' Bill's Light Show?
Ya not sure her actual self will be able to stand there that long without a malfunction. Kind of hoping I can get some video in time to show to my class on Tues. Always nice to go in with fresh material.
p.s. Poor old crook. I do feel sorry for her.
I lose all my "feel sorry fors" when it comes to somebody who covers up her husband's violent rape, somebody who deliberately undermines western values and people, and somebody whose incompetence is so severe that if I had done the same thing I would have been fired and prosecuted yet she is being offered a promotion.
Perhaps, but we don't know how much of that is true and how much has been fabricated or at least exaggerated to make her look bad.
I don't doubt she's a crook because that seems like a given considering she's in politics.
Either way I wouldn't wish brain damage on anyone.
In fact, one of the fascinating things about US elections to me is the amount of blatant poo-flinging that goes on. It seems like a standard tactic to give out disinformation on the other guy.
The third thing is well documented and true. The second thing has been verified by a variety of well-respected sources, but not to the exhaustive nature of the claims of a reasonable observer. The first thing is very likely true to a reasonable observer but not confirmed.
Early mail ballots in Florida strongly suggesting landslide for the not crooked candidate.
Cue Alex Jones in being 100% serial about HRC and her lizard family tree.
If this is true it probably increases my respect for Trump the most:
https://i.sli.mg/yEw03Z.jpg
The Florida absentee ballot delivery numbers suggest the result may be around 53/46/1 Trump/Clinton/Other. My gut says closer to 54/45/1. This would suggest Trump will win many blue wall states.
So the debate tomorrow: 90 minutes, no breaks. Hillary requested a foot stool but it was refused. When was the last time she stood up for 90 minutes straight I wonder?
When she was threatening Juanita Broaddrick at the fundraiser.Quote:
Originally Posted by Poopadoop
She will dominate.
I'd love an explanation for why/how.
Drivel.Quote:
Originally Posted by wufwugy
I don't see how Trump doesn't crush HRC in the debates (as far as overall national perception is concerned).
Facts don't matter against Trump. He has a way of dismantaling his opponents on stage.
Have you seen the cool shit holograms can do?Quote:
Originally Posted by wufwugy
That makes so much sense. Thank you.Quote:
Originally Posted by boost
Wuf, can you explain how you reached that conclusion? All you said was something about absentee ballot delivery numbers and how they led you to forecast some result. It did all seem a bit vague. Maybe that's because I don't know wtf an absentee ballot delivery number is or how it's supposed to be a forecasting tool.
I have no clue what an absentee ballot is either and I just skimmed over that comment of wuf's, while nodding at boost's response.
An absentee ballot is what they give to people who can't go to a voting booth on election day. Soldiers overseas, for example. Still don't know where the delivery number thing comes in or how it leads to the lolTrumpwinsFlorida conclusion.
So there will be a disproportionate amount of military workers in the ballot? Hmm I wonder if that will affect the prediction.Quote:
Soldiers overseas, for example.
That's my guess. A lot of soldiers asked for ballots maybe and they tend to vote Republican? Even if that's true, I'm not sure how that leads to the precise predictions he's giving though.Quote:
Originally Posted by OngBonga
But, I'm sure Wuf will clarify the situation for us soon and give rock-hard evidence for his calculations based on something he read on one of those Trump for Prez websites.
Florida 2012 absentee ballot returns (I think it was returns, not delivery) were 43% Democrat, 40% Republican, 17% Independent/Other. The election results were ~50/49/1.
Florida 2016 absentee deliveries (not returns) are 43% Republican, 37% Democrat, 20% Independent/Other.
If we assume that this breakdown matches booth voting, that deliveries match returns (they probably don't, but they should be expected to be close), and if we assume that each candidate receives 100% of their party support (they don't but it is probably an irrelevant distinction), then we can say that Obama got ~43% on a 43/55/2 split of Independent/Other (it was slightly higher), and Romney got the 55. These numbers don't add up exactly but they're very close. This made Obama's vote total 43+.43*17=50.31 (he got 50.01).
Assuming this breakdown for 2016, it gives Trump 43+.55*20=54. Though, I think he will get closer to 60% of Independent/Other, which would give him 55% of the vote, but previously when I stated my thoughts I used a different method that dropped everything by 1 percentage point (which is where I got the "53% but I think 54%).
Ya, perfect. And if we extrapolate these results from people in one state who have asked for a ballot but haven't voted yet to the entire country, Trump wins every state 55% to 45% give or take a percentage point or two margin of error. Might be three points in some states where the election takes place on a cloudy day which tends to do something funny, but I'm being conservative here.
https://www.youtube.com/watch?v=8FBtwBAuyE8
You did the same type of thing with Crooked's health that I do here. It's about using what you have to make the best judgments you can.
On a side note, there is a lot that suggests landslide that the Ostrich Brigade denies. Undecideds being so high and the convergence for Trump we've seen suggests that we could be seeing a 60-70% break of undecideds to Trump. The high poll numbers Trump shows in blue states. The below the charts enthusiasm regarding Crooked rallies/signage compared to the huge enthusiasm for Trump. These each individually suggest landslide, but obviously they're rough.
Oh btw I didn't mean that the florida numbers will transpose to the rest of the country. There is very strong correlation, however. Where the country moves, states move, and vice versa. Given how Florida is pretty much the most evenly divided state in the country, a significant win there would happen along with a landslide in the rest of the country some crazy high percentage of the time.
There's a difference between saying "The accumulation of converging lines of evidence points to a definite motor impairment that could be PD" versus "This one bit of evidence tells us exactly what the outcome of the election in this state will be within a percentage point."Quote:
Originally Posted by wufwugy
Still don't know half of what you're talking about but at least you're not making giant leaps of reason here afaik.Quote:
Originally Posted by wufwugy
I was joking. Satire, sarcasm, mocking, whatever you want to call it. Not serious there.Quote:
Originally Posted by wufwugy
Good news. I didn't say that.Quote:
Originally Posted by Poopadoop
Speaking of Hillary, is the debate still going ahead? I'm half expecting her to cancel on some euphemism for "unable to keep her balance for 90 minutes" grounds.
I'm terribly sorry if I misunderstood you. But no-one else seemed to understand you either.Quote:
Originally Posted by wufwugy
I used to (and probably still do) speak in emphatic tones and words, so my use of "this suggests that" is probably often viewed as "this therefore absolutely that."Quote:
Originally Posted by Poopadoop
*British sarcasm. Politely taking the piss.Quote:
Originally Posted by Poopadoop
Hehe don't tell him!Quote:
Originally Posted by OngBonga
The reason I think people took umbrage with this was that it was difficult to see how absentee ballot numbers could predict the outcome so precisely or why your gut should add another point for Trump. From that you then concluded that Trump will win many more states.Quote:
Originally Posted by wufwugy
I read it pre-edit.
And that is sarcasm very rare in the states.
Ah I see. I meant if the assessment was right then it would mean the blue wall coming down.Quote:
Originally Posted by Poopadoop
...and here you go from one assumption to some math to another assumption to some more math and so on without providing any evidence that your assumptions are accurate. That doesn't mean they're wrong, it just means no one can judge whether or not they're a solid basis for making predictions.Quote:
Originally Posted by wufwugy
For example, what's to say that the relationship between the absentee ballots by party in 2012 and the overall outcome in Florida will be the same this year? Is there some consistency in this relationship over the past X number of elections? If so, that would strengthen this assumption.
What's to say deliveries will match returns this time, or that each candidate receives 100% of their party vote, and that the split of the independent vote this year will be the same as in 2012?
Any one of these things could be different than in 2012, and you even acknowledge that yourself. So doing math based on assuming they're all going to be true will obviously give you some pretty unreliable numbers. You could be way low, way high or somewhere near the truth. It's impossible to judge though because none of your assumptions appear to be solid.
That's why i was joking saying let's just assume that these numbers apply across the country. Ya dig?
I've lived in America too. Great place, but ya no-one would get that joke. Sorry Wuffy.Quote:
Originally Posted by wufwugy
Why would Trump winning Florida mean a landslide? I'm not being funny I really just am that ignorant of US politics (you're lucky I even understood what a 'blue wall' is).Quote:
Originally Posted by wufwugy
Wow I didn't realise so many people did this absentee voting thing.
http://www.scpr.org/news/2016/09/23/...sentee-voting/
This is interesting:
"Early voters are people who have already made up their minds," McDonald said. "Clinton and Trump supporters will vote right now, and it won't matter what happens until Election Day. They're well-educated and dedicated."
Hmmm:
"It's not clear how the feelings toward these candidates will affect early voting. Partisan divisions are at their highest levels in decades, and although both candidates hold record-low favorability ratings, Clinton and Trump supporters tend to be pretty adamant about their candidate of choice."
No kidding...
And apparently the results of these early votes are made public as they come in. Didn't know that either.
The assumptions have a degree of accuracy, just nobody knows how much. This is what happens in predicting elections.Quote:
Originally Posted by Poopadoop
Nobody, except that it will be close because it is always close. In a way, I'm doing what the pollsters (the ones you think I may be dismissing) are doing: extrapolating from past trends.Quote:
For example, what's to say that the relationship between the absentee ballots by party in 2012 and the overall outcome in Florida will be the same this year?
My real thoughts are that Trump will win a larger proportion of his party and the other party than Romney did, and that Clinton will win a lower proportion of her party and the other party than Obama did, which would mean that my numbers are overly conservative. I haven't added this because it is much harder to quantify, so I just picked a model that would explain 2012 and have placed it on 2016.
I haven't seen data on this in the past. Without data on it, a good assumption is that it'll be close because returns should correlate well with deliveries.Quote:
What's to say deliveries will match returns this time,
They won't, but using the more accurate numbers won't change things much. IIRC candidates tend to get high 80s-mid 90s of their party support. Getting more accurate on this parameter would help Trump in my opinion.Quote:
or that each candidate receives 100% of their party vote,
It likely will be much more divergent. Trump's polls with independents is quite a bit better than I remember of Romney's.Quote:
and that the split of the independent vote this year will be the same as in 2012?
Welcome to how we do it in politics. I beat the polls in 2012 by doing this sort of stuff. I hope that's not a coincidence.Quote:
Any one of these things could be different than in 2012, and you even acknowledge that yourself. So doing math based on assuming they're all going to be true will obviously give you some pretty unreliable numbers. You could be way low, way high or somewhere near the truth. It's impossible to judge though because none of your assumptions appear to be solid.
Florida is the state closest to 50/50. While it is very likely to not be the tipping point state this cycle, the tipping point state is likely to run ~3 points more blue than Florida. This suggests that if Trump were to win by 9 points in Florida, he would win by ~6 nationally, which qualifies as landslide.Quote:
Originally Posted by Poopadoop
Right now it's looking like Colorado will be the tipping point state. I think Trump will win many bluer states beyond that, which is a different way to signify landslide in the US.
Actually, a better way of describing what I meant is that Florida is likely to run about 3 points redder than the nation. So a 9 point Trump win in Florida would likely be associated with a 6 point win nationally.
I'm not saying you don't have reasons for making these assumptions. What I'm saying is that each assumption you add into your model adds a corresponding amount of noise, to the point where your prediction comes close to being meaningless.
Add to this the fact that most professional modellers are predicting Florida to be a statistical tossup, and there's even less reason to accept your analysis of Florida being a Trump ass-kicking as being a solid one.
Moreover, and I don't mean to suggest there's anything deliberate about this on your part, but all of your various predictions seem to be closely in line with your professed desires.
You know how I've been talking about how the problem with the polls are that they're using poor parameters? Well, the Florida absentee ballot numbers suggest very strongly that to be the case. This type of thing puts some huge asterisks next to much of the professional analyses. The level of rigor of the model I presented is about as much as the one they're using. In fact, my model is probably better than theirs. They're assuming laughably bad demographic turnouts. I'm assuming a basic correlation between ballot types and votes given. Florida has a 9 point swing towards Republican in their 2 million early voters. It would be very hard for this to mean that Trump would not have a similar swing.Quote:
Originally Posted by Poopadoop
I might argue that this type of thing helps keep me more honest than I otherwise would be. I don't like being a fool.Quote:
Moreover, and I don't mean to suggest there's anything deliberate about this on your part, but all of your various predictions seem to be closely in line with your professed desires.
Many pollsters are using such awful models that they may be better explained by corruption than incompetence.
Wow that's hard to watch, to think people make their mind up over these debates is frightening. Absolute dross.
He is destroying her so bad it hurts to watch.
Love the "you've been fighting ISIS your entire adult life" line. Bait that the doofus journos gonna take tomorrow. They'll be arguing about how ISIS was actually created under Obama's term.
I agree though. It was very upsetting to me before I learned that when it comes to persuasion, facts don't matter.Quote:
Originally Posted by ImSavy
That doesn't make it less crap to watch.Quote:
Originally Posted by wufwugy
Hopefully someone else starts posting in here with you so I get a running commentary without having to watch.
Dunno. I think the Hillary double is holding her own.
Oooh, race relations. Time for Trump to show his great statesmanship.
Hillary for sure looks good-ish here because she doesn't look like death and seems normal.
I'm curious that Trump mentioned stop and frisk. I wonder what the intention is. He A/B tested that one a week or so ago. It must have tested well, but I thought it didn't.
I think Trump's strategy is to beat her up in this debate but then get much softer in the final debate. It'll give him the best of both worlds: airing her dirties and looking aggressive but still ending on the look of calm and collected.