So Michael D IS running again!

realdanbreen

GarIT wrote: »

No they aren't. Probabilities must always equal to 1 when all mutually exclusive outcomes are added together. Odds are an analysis of a situation by the bookies to decide what rate would be favourable enough to make people play but unfavourable enough to make the bookies money + a loading factor to hedge against a bets winning or losing so the bookie isn't taking a large risk. If odds were probabilities every bookies would have the same odds.

FFS! All I had said was that the polls showed Hillary winning and were wrong and the polls showed Brexit being defeated and was wrong:rolleyes:

jmcc

realdanbreen wrote: »

FFS! All I had said was that the polls showed Hillary winning and were wrong and the polls showed Brexit being defeated and was wrong:rolleyes:

Wonderful seeing all this expertise about odds and Probability. The polls with Brexit had serious flaws in that the samples were not quite representative of the electorate. The polls in the US 2016 election had issues with response rates and led to complete f*ckwittery like Huffington Post's claim that HRC had a 98.2% chance of winning. With the Irish GE in 2016, the clueless weighting of voters on their likelihood to vote (only voters with an 80% or greater likeihood to vote were considered "voters") caused serious problems for one polling company.

The polls for this election are going to run into a serious problem with sampling the younger voter demographic if they rely on telephone polling rather than face to face polling. Fixed line phones are more likely to be owned by older voters. Younger voters are more likely to have only mobile phones.

Regards...jmcc

pixelburp

Squatter wrote: »

I suppose it all depends on whether or not one is happy to accept deceit from any Irish Constitutional office holder.

Because no matter how much you may equivocate about how nice he and his missus are is to small children, neighbouring monarchs and cats, the fact remains that many of us believe that he lied to us.

You are happy to forgive him, and that's your right, however I (and, it appears, quite a few others) aren't quite as tolerant of such an act.

And there I'll leave it because there's nothing else to be said.

It's not a question of being "happy to accept deceit", it's that I don't classify Higgins' decision AS deceit, a lie or any kind of disgraceful duplicity. Nor is it about whether he's a nice guy or not - I never made that argument - simply that it's a bit hyperbolic to hold politicians to some emotional standard that you wouldn't hold anyone else by.

It's not like politicians don't break promises on a regular basis, but drawing moral equivalence between a change-of-mind over a personal decision, and a demonstrative or deliberate reversal of a manifesto policy, is a little disingenuous.

The only reason the 'one term' thing came up at all was because of Higgins' age - that's literally it. Were Higgins 10 years younger this wouldn't even be a talking point, yet being the only 'non gimmick' candidate (bar Gay Mitchell, but who honestly cared about him) there was probably some intent to come up with a prickly question, all in the interest of balance. Seemed like a reasonable approach for any ageing individual to hedge ones bets.

Full disclosure: I'm not even that enthused about Higgins running again, I actually DO think he's a little to old for the role - but clearly Higgins feels his age isn't as large a barrier as perhaps he estimated back in 2011. He's probably feeling healthier and heartier than he feared. If a decent candidate runs against him, Higgins mightn't get my vote - but I don't draw parallels between a change of mind and rank duplicity.

oscarBravo

jmcc wrote: »

The polls in the US 2016 election had issues with response rates and led to complete f*ckwittery like Huffington Post's claim that HRC had a 98.2% chance of winning.

Let's take even that outlier of a prediction. Was it wrong?

Before I roll a pair of dice, I can confidently predict that there's a 97.2% chance that the outcome won't be "snake eyes". If I subsequently roll two ones, does that make my prediction wrong? Note that I didn't claim that I won't roll a two, and Huffpo didn't claim that Trump couldn't possibly win.

Like I said, people who say "the polls were wrong" are generally using it as a shorthand way of saying "I don't understand how statistics works", and that way lies "we have had enough of experts".

Hotblack Desiato

pixelburp wrote: »

Were Higgins 10 years younger this wouldn't even be a talking point, yet being the only 'non gimmick' candidate (bar Gay Mitchell, but who honestly cared about him)

Ah come on, David Norris was a far more credible candidate than Gay "Dublin Olympics" "Irish Sea Tunnel" Mitchell.

blanch152

Hotblack Desiato wrote: »

Ah come on, David Norris was a far more credible candidate than Gay "Dublin Olympics" "Irish Sea Tunnel" Mitchell.

Is that the same Norris who was unable to work as a lecturer on disability grounds and was claiming a disability pension while at the same time was fully fit and able to lecture us all from his privileged position as a Senator?

jmcc

oscarBravo wrote: »

Let's take even that outlier of a prediction. Was it wrong?

Yes. It was based on poor analysis of polls. It even drove Nate Silver to call it out on Twitter using profanity to explain the abject cluelessness. I'm sure you've heard of Nate Silver and 538.

Like I said, people who say "the polls were wrong" are generally using it as a shorthand way of saying "I don't understand how statistics works", and that way lies "we have had enough of experts".

I do understand how polls work and I also understand their limitations. You obviously don't. One of the biggest problems with opinion polls is response rate. This means that rather than listing a non-response, the pollsters have to poll more than the number in the sample to get their sample. Thus a "sample" of 1,000 voters may not be a sample of 1,000 voters but rather a subset of a sample of more than 1,000 voters where some voters did not wish to respond to the pollster. Some pollsters record that percentage but ir rarely makes into to the polls that are published in the newspapers and TV. And that's just the response rate issue.

The telephone polling method is a transplant from the US and despite modifications, it is highly problematic in an Irish situation due to the differences in how area codes work in the US model and how they work in the Irish model. Irish telephone area codes can span county borders.

Now if you want to read the comments of real experts then read this:
https://www.aapor.org/Education-Resources/Reports/An-Evaluation-of-2016-Election-Polls-in-the-U-S.aspx

Right there in section 1.2.3 it specifically deals with nonresponse bias and the problems with underestimating Trump support.

The last thing that any pollster wants is to end up like the Literary Digest poll.
https://en.wikipedia.org/wiki/The_Literary_Digest#Presidential_poll

And here's the HuffPo's apology:
https://www.huffingtonpost.com/entry/pollster-forecast-donald-trump-wrong_us_5823e1e5e4b0e80b02ceca15

Regards...jmcc

oscarBravo

jmcc wrote: »

Yes. It was based on poor analysis of polls.

Right, but my point, which you seem determined to miss, is that being a flawed poll, doesn't make it wrong, at least not in the sense that many people are so fond of accusing polls of being wrong.

I'll reiterate the important part of my post, which you omitted when you quoted me: if I predict that there's a 97.2% chance that I won't roll a two, and then I roll a two, was my prediction wrong?

jmcc

oscarBravo wrote: »

Right, but my point, which you seem determined to miss, is that being a flawed poll, doesn't make it wrong, at least not in the sense that many people are so fond of accusing polls of being wrong.

It wasn't a poll. It was a model based on the data from a number of polls. It is all there in the HuffPo apology.

Regards...jmcc

oscarBravo

jmcc wrote: »

It wasn't a poll. It was a model based on the data from a number of polls. It is all there in the HuffPo apology.

Oy vey.

Are you aware of what my point is, and are disagreeing with it?

Are you aware of what my point is, but are determined to make this a conversation about the HuffPo not-a-poll instead?

If I start the whole conversation again but this time talk about 538's prediction that Hillary Clinton had a 71.4% chance of winning, will you discuss the actual point I'm trying to make?

Let's try that. Were 538 wrong?

jmcc

oscarBravo wrote: »

Are you aware of what my point is, and are disagreeing with it?

You think you know about polling but you don't. It is that simple. You didn't understand what the HuffPo prediction was based upon and thought it was a single poll. You brought up a spurious dice example but this is a problem with polling, its methodologies, interpretation and limitations.

Just to explain in simple terms that you may understand. Polls are snapshots of opinions at a specific time. They are not perfect and often incorporate flaws and, even though pollsters try hard to exclude them, biases. The HuffPo prediction was deeply flawed and its apology blamed, among other things, the data it used.

Regards...jmcc

oscarBravo

jmcc wrote: »

You think you know about polling but you don't. It is that simple.

I clearly forgot option 3: "I don't care what your point is, I just have a compulsion to tell you you're wrong about something."

Hotblack Desiato

blanch152 wrote: »

Is that the same Norris who was unable to work as a lecturer on disability grounds and was claiming a disability pension while at the same time was fully fit and able to lecture us all from his privileged position as a Senator?

Yep being a senator is very much a part-time job. It was private health insurance he was claiming off, not disability benefit btw.

jmcc

oscarBravo wrote: »

I clearly forgot option 3: "I don't care what your point is, I just have a compulsion to tell you you're wrong about something."

You repeated the clueless media cliches about "people who say "the polls were wrong" are generally using it as a shorthand way of saying "I don't understand how statistics works" "and " "we have had enough of experts" ". The problem is that the media didn't understand polling and its limitations. As in the cases of Brexit and the 2016 US election, many were quite partisan and biased. Thus when people who really did understand polling and Statistics (experts) were pointing out that there were problems with the reliability of the polls and their results, they were largely ignored by journalists who where oh so sure that Brexit would be rejected and that HRC would be POTUS. Both of these outcomes took them by surprise.

Your introduction of your dice example without understanding what was behind the HuffPo "prediction" is typical of the cluelessness that surrounds the issue. You didn't understand what the HuffPo prediction was based upon. You thought it was a poll. And to show that you were paying attention in your Probability class, you introduced a totally spurious and irrelevant dice example. HuffPo tried to blame poor data and excluded data in its apology. Real experts such as those in the AAPOR as linked above tried to analyse what happened with polling in the US presidential election.

There were also flaws in the polling for the recent Repeal referndum in that some pollsters using phone polling were predicting a close result while those using face to face polling were predicting a strong yes result. Perhaps you can explain why.

Regards...jmcc

Zubeneschamali

Hotblack Desiato wrote: »

It was private health insurance he was claiming off, not disability benefit btw.

Trinity put him on permanent sick leave - an infamously gay man who has contracted a strange virus, they were not touching that with a bargepole back in the 90s.

Beechwoodspark

Norris was a disastrous candidate for president.

Is there a chance that 2 dragons den investors (Duffy and Gallagher) could run for the park? If so, bizarre.

Zubeneschamali

jmcc wrote: »

There were also flaws in the polling for the recent Repeal referndum in that some pollsters using phone polling were predicting a close result while those using face to face polling were predicting a strong yes result.

Can you link to one of the polls that predicted a close result?

I thought I was following that one closely, and never saw any such poll.

https://twitter.com/KeithMillsD7/status/1000311764490358785

jmcc

Zubeneschamali wrote: »

Can you link to one of the polls that predicted a close result?

I thought I was following that one closely, and never saw any such poll.

https://twitter.com/KeithMillsD7/status/1000311764490358785

https://www.redcresearch.ie/crowd-predicts-narrow-yes-victory/

https://redcresearch.ie/wp-content/uploads/2018/05/SBP-May-2018-Poll-Report-8th-Referendum-GE-Vote-Intention-1.pdf

RedC used telephone polling and, I think, split their samples into fixed line and mobile.

Regards...jmcc

Zubeneschamali

jmcc wrote: »

https://www.redcresearch.ie/crowd-predicts-narrow-yes-victory/

https://redcresearch.ie/wp-content/uploads/2018/05/SBP-May-2018-Poll-Report-8th-Referendum-GE-Vote-Intention-1.pdf

OK, so their actual polling was not close at all, varying from 3:1 Yes to 2:1 Yes, with the poll closer to polling day being exactly right.

They mangled it in analysis of how to allocate the undecideds, got it utterly wrong and were able to print an "exciting" prediction with a closer result.

jmcc

Zubeneschamali wrote: »

OK, so their actual polling was not close at all, varying from 3:1 Yes to 2:1 Yes, with the poll closer to polling day being exactly right.

They mangled it in analysis of how to allocate the undecideds, got it utterly wrong and were able to print an "exciting" prediction with a closer result.

I think it had more to do with the sampling. Many of those voting in the referendum were first time voters and as such they were unlikely to own fixed line phones. Instead they probably owned mobile phones. I think that the number of new voters was over 100K. With the fixed line sample, that's 500 responses from older demographics. Most of those new voters would have been almost completely excluded from the fixed line sample unless they answered the fixed line phone by accident. Most of those in the fixed line sample would have had mobile phone anyway (mobile phone ownership is high in Ireland). It wasn't a normal election style situation with multiple choices. By not accounting for the influx of new voters who would have been mobile phone based, the effect of these new voters was effectively missed.

Regards...jmcc

Zubeneschamali

jmcc wrote: »

I think it had more to do with the sampling. Many of those voting in the referendum were first time voters and as such they were unlikely to own fixed line phones. Instead they probably owned mobile phones. I think that the number of new voters was over 100K. With the fixed line sample, that's 500 responses from older demographics. Most of those new voters would have been almost completely excluded from the fixed line sample unless they answered the fixed line phone by accident. Most of those in the fixed line sample would have had mobile phone anyway (mobile phone ownership is high in Ireland). It wasn't a normal election style situation with multiple choices. By not accounting for the influx of new voters who would have been mobile phone based, the effect of these new voters was effectively missed.

Regards...jmcc

No, here are the actual percentages from your links, raw and then excluding Don't Know/Won't Vote:

60/20/20 - 75% Yes
56/26/18 - 68% Yes
53/26/21 - 67% Yes
56/27/17 - 67% Yes

Result on the day - 66.4% Yes.

Absolutely nothing to explain, their polling was dead on.

jmcc

Zubeneschamali wrote: »

No, here are the actual percentages from your links, raw and then excluding Don't Know/Won't Vote:

60/20/20 - 75% Yes
56/26/18 - 68% Yes
53/26/21 - 67% Yes
56/27/17 - 67% Yes

Result on the day - 66.4% Yes.

Absolutely nothing to explain, their polling was dead on.

After the vote.

Regards...jmcc

spurious

Hotblack Desiato wrote: »

Yep being a senator is very much a part-time job. It was private health insurance he was claiming off, not disability benefit btw.

Oh away with you and your facts...

Hotblack Desiato

jmcc wrote: »

https://www.redcresearch.ie/crowd-predicts-narrow-yes-victory/

https://redcresearch.ie/wp-content/uploads/2018/05/SBP-May-2018-Poll-Report-8th-Referendum-GE-Vote-Intention-1.pdf

RedC used telephone polling and, I think, split their samples into fixed line and mobile.

Regards...jmcc

Poll excluding DK was spot on. Where they went wrong was crystal-ball-gazing and coming up with a very large "shy No" vote which just didn't happen.

This was not a sampling issue, the sample turned out to be exactly representative. They just assumed that the don't knows would overwhelmingly vote a particular way.

Zubeneschamali

jmcc wrote: »

After the vote.

No - before the vote, on 4 different dates.

You said There were also flaws in the polling for the recent Repeal referndum in that some pollsters using phone polling were predicting a close result

This redc data shows the opposite - their polling was exactly correct (excluding don't knows). Perhaps you meant some other poll.

Drifter50

bubblypop wrote: »

Has a serving president ever died?

Yes, Cearbhall O` Dalaigh died 2 years into his term in the late 70`s I think

eagle eye

Beechwoodspark wrote:

Is there a chance that 2 dragons den investors (Duffy and Gallagher) could run for the park? If so, bizarre.

Yeah and they are on the wrong show. Bill Cullen would be the man to go for it. The Apprentice is the ticket to becoming a President!!!

alaimacerc

Drifter50 wrote: »

Yes, Cearbhall O` Dalaigh died 2 years into his term in the late 70`s I think

He died five years after the start of his term, but two years after he'd resigned!

Erskine Childers it was who died in post, the chap before him.

Zubeneschamali

alaimacerc wrote: »

He died five years after the start of his term, but two years after he'd resigned!

A thundering disgrace.

Beechwoodspark

Craughwell has pulled out. Just announced.

Zubeneschamali

Beechwoodspark wrote: »

Craughwell has pulled out. Just announced.

I had to look him up in Wikipedia to remind myself who he was. Utter waste of time him running.

Beechwoodspark

Zubeneschamali wrote: »

I had to look him up in Wikipedia to remind myself who he was. Utter waste of time him running.

Anecdotally a number of ppl I talked to particularly outside Dublin (midlands especially) were interested in hearing more from him and potential vote for him.

He has a lot of connections through his unions work, teaching, etc.

Zubeneschamali

Beechwoodspark wrote: »

Anecdotally a number of ppl I talked to particularly outside Dublin (midlands especially) were interested in hearing more from him and potential vote for him.

I've lived in Athlone for 25+ years and literally never heard of him.

eastwest

Zubeneschamali wrote: »

I've lived in Athlone for 25+ years and literally never heard of him.

A lot more people know his name now, so he got his result.
As for the presidency, he hadn't a hope, but he probably knew that.

dulpit

Zubeneschamali wrote: »

I've lived in Athlone for 25+ years and literally never heard of him.

Where does this sense of pride in not knowing something come from? I wouldn't be too familiar with Gerard Craughwell, but he has been all over the news related to the presidency for the last few months, if you've been anyway keeping an eye/ear on the news you'd have heard it.

In one sense - he's been successful (he said he wanted there to be an election and it looks like there will be). On the other hand, sounds like he wasn't able to get to 20 votes in the Oireachtas. Should free up enough votes for 1 independent though...

Beechwoodspark

Zubeneschamali wrote: »

I've lived in Athlone for 25+ years and literally never heard of him.

He’d be very well known in teaching circles -was the president of the TUI for two years.

Also would have loads of connections among county councilors the length and breadth of the country. He got in a bit of controversy for sending them advice on how to maximise expenses claims.

blanch152

Hotblack Desiato wrote: »

Yep being a senator is very much a part-time job. It was private health insurance he was claiming off, not disability benefit btw.

He was getting a disability payment from TCD.

Zubeneschamali

dulpit wrote: »

Where does this sense of pride in not knowing something come from?

I'm not saying I'm proud of not knowing him - Beechwoodspark was saying he is known in the midlands especially. Well, I live bang in the middle and he is not so well known that I ever heard of him.

So - his chances at a national level were always between zero and nothing.

Hotblack Desiato

blanch152 wrote: »

He was getting a disability payment from TCD.

Nope. He was receiving a payment from an insurance company, from a policy he paid for with his own money.

dulpit wrote: »

In one sense - he's been successful (he said he wanted there to be an election and it looks like there will be).

Yes, a pointless election. Craughwell appears to be quite the fan of wasting public money.

Presidential hopeful defends advising local councillors about how to 'maximise' mileage expenses

realdanbreen

Beechwoodspark wrote: »

He got in a bit of controversy for sending them advice on how to maximise expenses claims.

Straight away that makes him well qualified for the Presidency!

Hotblack Desiato

From the Galway City forum during the 2016 count:

Borders no.2 wrote: »

Poor Patrick Feeney with only 22 votes. You have to give credit to some of the independent candidates who have no real financial backing behind them for putting themselves forward. It seems foolish to risk such a public humiliation looking on from a distance but fair play to them for giving it a go.

Not content with failing abysmally to run a campaign in a single constituency, he now wants to run a nationwide one:

https://www.irishtimes.com/news/politics/presidential-hopefuls-set-out-their-stalls-in-waterford-1.3582376

Retired Aer Lingus employee Patrick Feeney, from Galway, who secured 22 votes when he ran in the 2016 general election as an Independent, said he wants to be “the innovation president of Ireland” and wants a nomination for the presidential election “as a challenge for democracy against what’s going on at the moment around politics”.

Also, I doubt 1.4 million Yes voters will be too impressed with this guff:

Ms Freeman was asked by Fianna Fail councillor Adam Wyse if, as a “No” voter in the recent abortion referendum, she would sign the resultant legislation into law, and said that while she did vote No, she would “absolutely respect the vote” of the people and sign the act into law. “I voted No, but I didn’t vote No for religious reasons, I voted No because I spend all my adult life trying to conserve life.”

eastwest

Hotblack Desiato wrote: »

From the Galway City forum during the 2016 count:



Not content with failing abysmally to run a campaign in a single constituency, he now wants to run a nationwide one:

https://www.irishtimes.com/news/politics/presidential-hopefuls-set-out-their-stalls-in-waterford-1.3582376
:

You need one hopeless case who can't put two words together in every campaign.
Remember the guy, TJ (?) From cavan who ran last time? About as far removed from presidential material as it's possible to be, but he really believed he could do it and around a hundred voters thought likewise. He and his supporters decided for some reason to picket the count in castlebar but they picketed the town hall instead for a couple of hours before they realised that the count was being held in the TF hotel!
Don't you just love democracy!

Hotblack Desiato

Wikipedia wrote:

The seven candidates were Mary Davis, Sean Gallagher, Michael D. Higgins, Martin McGuinness, Gay Mitchell, David Norris and Dana Rosemary Scallon.

TJ from Cavan how would the likes of him get nominated?

blanch152

Hotblack Desiato wrote: »

TJ from Cavan how would the likes of him get nominated?

We already have the space for the conspiracy theory nut taken by the anti-vaxxer from SF:

https://www.independent.ie/irish-news/politics/sf-mep-in-line-for-run-at-ras-ruled-out-hpv-vaccination-for-her-daughter-37166082.html

blazard

realdanbreen wrote: »

So when Michael D was campaigning for president and said it would be a one off term he was lying?
Many voted for him because he was only going for one term and wasn't trying to hog the job for 14 years.

he wasnt lying, he changed his mind at the time


i do think he will regret the upcoming debate,however

eastwest

Hotblack Desiato wrote: »

Wikipedia wrote:

The seven candidates were Mary Davis, Sean Gallagher, Michael D. Higgins, Martin McGuinness, Gay Mitchell, David Norris and Dana Rosemary Scallon.

TJ from Cavan how would the likes of him get nominated?

My mistake, I was thinking of the last European elections and the TV debate for that. There were a couple of candidates there who didn't get a hundred votes between them.

AGC

blazard wrote: »

he wasnt lying, he changed his mind at the time


i do think he will regret the upcoming debate,however

He will regret it how?

realdanbreen

blazard wrote: »

he wasnt lying, he changed his mind at the time


i do think he will regret the upcoming debate,however

, Gotya, It's a bit like when a bishop was caught lying one time he said he wasn't lying instead he 'had a mental reservation'

realdanbreen

AGC wrote: »

He will regret it how?

Well if he is quizzed properly we will see that despite all the 'Ochóns and grá mo chrois and reciting poetry and talking in riddles, he has not done or attempted to do anything constructive for this country. And before it's pointed out that his role is mainly ceremonial there is nothing to prevent him from speaking out in plain English about the inequality that is there for all to see.

AGC

realdanbreen wrote: »

Well if he is quizzed properly we will see that despite all the 'Ochóns and grá mo chrois and reciting poetry and talking in riddles, he has not done or attempted to do anything constructive for this country. And before it's pointed out that his role is mainly ceremonial there is nothing to prevent him from speaking out in plain English about the inequality that is there for all to see.

Have you listened to many of his speeches or remarks where he has gone beyond his ‘remit’ and made comments on various matters?

Don’t think you could ever question the president on equality issues.