Nate Silver 538
Comments
-
The Juggler said:
How I Acted Like A Pundit And Screwed Up On Donald Trump
Trump’s nomination shows the need for a more rigorous approach.
By Nate Silver
Filed under 2016 Election
Published May 18, 2016
....
Nate my friend, love your site, but, you did not find that more rigorous approach that dreary November and could be making the exact same mistake in 2020.
You are zero for two identifying trump voters. I give you credit for continually trying.
0 -
Great... there's better than a 23% chance. So what is it, 24%? 88%? The beauty of your argument i suppose is that you can't be wrong even if you're not using math, only old articles. If Trump wins, you say...I was right! Same thing if he loses. Neither would be empirically true.Lerxst1992 said:mrussel1 said:
Dude.. sorry, you don't have the background or analysis in this topic to declare that Silver's 75 or 77% is way off. Your assumption also discounts the massive move of suburbanites toward the D. That was clear in the analysis post 2018 and the numbers are bearing out the same this time. Trump won the suburbs in 16 and he is losing them today. They are not exactly chock full of NCWs. Women have also jettisoned the R party dramatically since 16 as well. So the statement that NCW's moving from 55 to 60% turnout wins the election for Trump ignores every other demographic move.Lerxst1992 said:mrussel1 said:
This is exactly correct. Somehow people (particularly Trump supporters) think something less than 100% is equal to 101%.dignin said:
The issue I have is you calling 538's math foolish. You seem to want to say that 538 should be calling this more 50/50, when the math actually calls for around 75/25. What you don't seem to understand is that in 4 simulated coin flips Biden would win 3 times while Trump wins once. Nobody here is excited about those odds, of course we want Biden winning 4 of those coin flips but the math doesn't show Trump winning 2 of those coin flips either. Biden is still the odds on favourite.Lerxst1992 said:Lerxst1992 said:Cook political report has an interesting tool where you can adjust turnout by demographic and see how that impacts electoral college.
It’s starts with trumps best demo, non college whites at 55% (Biden win w 307 electoral votes). Increasing that demo to 60% gets trump a win with about 280 EVs. 59% turnout, reduces it to 230 EVsI cant imagine a 1% turnout difference in trumps base nets a 50 EV difference, but it underscores how difficult a job pollsters have to get the polls accurate and account for even a tiny increase in trump base turnout. It ain’t helping my confidence with pollsters and if it’s this much of a difference 538 is really off their rockers w .77
HOW IT WORKS: Start with the results of the previous election, adjusted fordemographic change since 2016. Then, adjust the sliders below to see how shifts inturnout and support among five demographic groups could swing the Electoral CollegeSorry but this is damn interesting, as is the Supreme Court vs House analysis to decide contested elections. Clearly I’m in the wrong place.
This election isn't another 2016, way too many variables have changed.Not foolish, just underestimating the trump effect on elections.
Most significantly, elections are not about flipping a coin. According to Cook, Trumps 2016 performance at getting NCWs is about 4% from a huge statistical advantage, where one percent of this demo can net trump 50 EVs.He was at .55 NCW 2016 turnout. At .59 Biden still wins with @ 280 EVs. Thats great, we can all exhale. At .60 turnout, it goes to 230 and trump wins. THATS the tipping point here, this potential wave trump has access too. Being by far The largest demo (by far can not be emphasized enough), it is foolish to discount this phenomenon. That’s not even close to flipping a coin. It’s extraordinarily biased and a weighted coin.I’m hoping trump never gets there, but I am not shutting my eyes about it either. He proved in 2016 he knows how to get new voters from this demo to vote. No reason he can not find new voters there again. That’s why 77% is way off.Dude, the mistake Silver is making is this is not statistics, it’s political science. Trump has tapped into something pollsters did not identify last time. There is a better than 23% chance he can do it again.
lets fire up that time machine...again,
....PoliticsClinton Dominates in Key Philadelphia Suburbs, Bloomberg Poll Finds
Losing Pennsylvania’s 20 Electoral College votes would sharply curtail Trump’s paths to the White House.October 13, 2016, 5:00 AM EDT0 -
mrussel1 said:
Great... there's better than a 23% chance. So what is it, 24%? 88%? The beauty of your argument i suppose is that you can't be wrong even if you're not using math, only old articles. If Trump wins, you say...I was right! Same thing if he loses. Neither would be empirically true.Lerxst1992 said:mrussel1 said:
Dude.. sorry, you don't have the background or analysis in this topic to declare that Silver's 75 or 77% is way off. Your assumption also discounts the massive move of suburbanites toward the D. That was clear in the analysis post 2018 and the numbers are bearing out the same this time. Trump won the suburbs in 16 and he is losing them today. They are not exactly chock full of NCWs. Women have also jettisoned the R party dramatically since 16 as well. So the statement that NCW's moving from 55 to 60% turnout wins the election for Trump ignores every other demographic move.Lerxst1992 said:mrussel1 said:
This is exactly correct. Somehow people (particularly Trump supporters) think something less than 100% is equal to 101%.dignin said:
The issue I have is you calling 538's math foolish. You seem to want to say that 538 should be calling this more 50/50, when the math actually calls for around 75/25. What you don't seem to understand is that in 4 simulated coin flips Biden would win 3 times while Trump wins once. Nobody here is excited about those odds, of course we want Biden winning 4 of those coin flips but the math doesn't show Trump winning 2 of those coin flips either. Biden is still the odds on favourite.Lerxst1992 said:Lerxst1992 said:Cook political report has an interesting tool where you can adjust turnout by demographic and see how that impacts electoral college.
It’s starts with trumps best demo, non college whites at 55% (Biden win w 307 electoral votes). Increasing that demo to 60% gets trump a win with about 280 EVs. 59% turnout, reduces it to 230 EVsI cant imagine a 1% turnout difference in trumps base nets a 50 EV difference, but it underscores how difficult a job pollsters have to get the polls accurate and account for even a tiny increase in trump base turnout. It ain’t helping my confidence with pollsters and if it’s this much of a difference 538 is really off their rockers w .77
HOW IT WORKS: Start with the results of the previous election, adjusted fordemographic change since 2016. Then, adjust the sliders below to see how shifts inturnout and support among five demographic groups could swing the Electoral CollegeSorry but this is damn interesting, as is the Supreme Court vs House analysis to decide contested elections. Clearly I’m in the wrong place.
This election isn't another 2016, way too many variables have changed.Not foolish, just underestimating the trump effect on elections.
Most significantly, elections are not about flipping a coin. According to Cook, Trumps 2016 performance at getting NCWs is about 4% from a huge statistical advantage, where one percent of this demo can net trump 50 EVs.He was at .55 NCW 2016 turnout. At .59 Biden still wins with @ 280 EVs. Thats great, we can all exhale. At .60 turnout, it goes to 230 and trump wins. THATS the tipping point here, this potential wave trump has access too. Being by far The largest demo (by far can not be emphasized enough), it is foolish to discount this phenomenon. That’s not even close to flipping a coin. It’s extraordinarily biased and a weighted coin.I’m hoping trump never gets there, but I am not shutting my eyes about it either. He proved in 2016 he knows how to get new voters from this demo to vote. No reason he can not find new voters there again. That’s why 77% is way off.Dude, the mistake Silver is making is this is not statistics, it’s political science. Trump has tapped into something pollsters did not identify last time. There is a better than 23% chance he can do it again.
lets fire up that time machine...again,
....PoliticsClinton Dominates in Key Philadelphia Suburbs, Bloomberg Poll Finds
Losing Pennsylvania’s 20 Electoral College votes would sharply curtail Trump’s paths to the White House.October 13, 2016, 5:00 AM EDTThe funny thing is I’d estimate Biden at .63, maybe .14 lower than 538, based on tipping point states. I think Biden is the favorite, but there is a stronger chance pollsters are making the same mistake again than .23; my main concern is trump is looking much stronger lately in these states, NC FL AZ PA. With these states within or near the moe, if having trouble believing .77 is accurate. If there is an “unpolled white vote” as there was in 2016 trump could be tied or ahead in those states.
the 2 main tipping point states, 538 has Biden at .77 in PA and .64 in AZ. The two closest tipping point states have a .13 spread. I find that debatable.
Its interesting that gauging demographic accuracy in polling is part of 538s process, but if we look at PA, they actually give biden better rating on that category (Biden 53/ trump 46). That just seems wrong, but I can not prove it, except to point out this is where pollsters got trump wrong in the past.
“ Average with demographics-based vote share projection”0 -
lolLerxst1992 said:mrussel1 said:
Great... there's better than a 23% chance. So what is it, 24%? 88%? The beauty of your argument i suppose is that you can't be wrong even if you're not using math, only old articles. If Trump wins, you say...I was right! Same thing if he loses. Neither would be empirically true.Lerxst1992 said:mrussel1 said:
Dude.. sorry, you don't have the background or analysis in this topic to declare that Silver's 75 or 77% is way off. Your assumption also discounts the massive move of suburbanites toward the D. That was clear in the analysis post 2018 and the numbers are bearing out the same this time. Trump won the suburbs in 16 and he is losing them today. They are not exactly chock full of NCWs. Women have also jettisoned the R party dramatically since 16 as well. So the statement that NCW's moving from 55 to 60% turnout wins the election for Trump ignores every other demographic move.Lerxst1992 said:mrussel1 said:
This is exactly correct. Somehow people (particularly Trump supporters) think something less than 100% is equal to 101%.dignin said:
The issue I have is you calling 538's math foolish. You seem to want to say that 538 should be calling this more 50/50, when the math actually calls for around 75/25. What you don't seem to understand is that in 4 simulated coin flips Biden would win 3 times while Trump wins once. Nobody here is excited about those odds, of course we want Biden winning 4 of those coin flips but the math doesn't show Trump winning 2 of those coin flips either. Biden is still the odds on favourite.Lerxst1992 said:Lerxst1992 said:Cook political report has an interesting tool where you can adjust turnout by demographic and see how that impacts electoral college.
It’s starts with trumps best demo, non college whites at 55% (Biden win w 307 electoral votes). Increasing that demo to 60% gets trump a win with about 280 EVs. 59% turnout, reduces it to 230 EVsI cant imagine a 1% turnout difference in trumps base nets a 50 EV difference, but it underscores how difficult a job pollsters have to get the polls accurate and account for even a tiny increase in trump base turnout. It ain’t helping my confidence with pollsters and if it’s this much of a difference 538 is really off their rockers w .77
HOW IT WORKS: Start with the results of the previous election, adjusted fordemographic change since 2016. Then, adjust the sliders below to see how shifts inturnout and support among five demographic groups could swing the Electoral CollegeSorry but this is damn interesting, as is the Supreme Court vs House analysis to decide contested elections. Clearly I’m in the wrong place.
This election isn't another 2016, way too many variables have changed.Not foolish, just underestimating the trump effect on elections.
Most significantly, elections are not about flipping a coin. According to Cook, Trumps 2016 performance at getting NCWs is about 4% from a huge statistical advantage, where one percent of this demo can net trump 50 EVs.He was at .55 NCW 2016 turnout. At .59 Biden still wins with @ 280 EVs. Thats great, we can all exhale. At .60 turnout, it goes to 230 and trump wins. THATS the tipping point here, this potential wave trump has access too. Being by far The largest demo (by far can not be emphasized enough), it is foolish to discount this phenomenon. That’s not even close to flipping a coin. It’s extraordinarily biased and a weighted coin.I’m hoping trump never gets there, but I am not shutting my eyes about it either. He proved in 2016 he knows how to get new voters from this demo to vote. No reason he can not find new voters there again. That’s why 77% is way off.Dude, the mistake Silver is making is this is not statistics, it’s political science. Trump has tapped into something pollsters did not identify last time. There is a better than 23% chance he can do it again.
lets fire up that time machine...again,
....PoliticsClinton Dominates in Key Philadelphia Suburbs, Bloomberg Poll Finds
Losing Pennsylvania’s 20 Electoral College votes would sharply curtail Trump’s paths to the White House.October 13, 2016, 5:00 AM EDTThe funny thing is I’d estimate Biden at .63, maybe .14 lower than 538, based on tipping point states. I think Biden is the favorite, but there is a stronger chance pollsters are making the same mistake again than .23; my main concern is trump is looking much stronger lately in these states, NC FL AZ PA. With these states within or near the moe, if having trouble believing .77 is accurate. If there is an “unpolled white vote” as there was in 2016 trump could be tied or ahead in those states.
the 2 main tipping point states, 538 has Biden at .77 in PA and .64 in AZ. The two closest tipping point states have a .13 spread. I find that debatable.
Its interesting that gauging demographic accuracy in polling is part of 538s process, but if we look at PA, they actually give biden better rating on that category (Biden 53/ trump 46). That just seems wrong, but I can not prove it, except to point out this is where pollsters got trump wrong in the past.
“ Average with demographics-based vote share projection”
At least you've moved on to PA from WI after we told you that was the bigger concern. Remember how you were telling us WI was in danger a few weeks ago? You've barely mentioned that state lately...
Post edited by The Juggler onwww.myspace.com0 -
My real question is how do you poll the people that are trying desperately to hold on to the white privilege system and will say all day that they aren’t voting for trump,especially to pollsters, but still check his name in the privacy of the voting booth. I think Trump has a much better chance than anyone thinks based on passive white supremacy, but since it can’t be easily quantifiable and put down in numbers it’s something that could prove to be the ultimate November surprise.Lerxst1992 said:mrussel1 said:
Great... there's better than a 23% chance. So what is it, 24%? 88%? The beauty of your argument i suppose is that you can't be wrong even if you're not using math, only old articles. If Trump wins, you say...I was right! Same thing if he loses. Neither would be empirically true.Lerxst1992 said:mrussel1 said:
Dude.. sorry, you don't have the background or analysis in this topic to declare that Silver's 75 or 77% is way off. Your assumption also discounts the massive move of suburbanites toward the D. That was clear in the analysis post 2018 and the numbers are bearing out the same this time. Trump won the suburbs in 16 and he is losing them today. They are not exactly chock full of NCWs. Women have also jettisoned the R party dramatically since 16 as well. So the statement that NCW's moving from 55 to 60% turnout wins the election for Trump ignores every other demographic move.Lerxst1992 said:mrussel1 said:
This is exactly correct. Somehow people (particularly Trump supporters) think something less than 100% is equal to 101%.dignin said:
The issue I have is you calling 538's math foolish. You seem to want to say that 538 should be calling this more 50/50, when the math actually calls for around 75/25. What you don't seem to understand is that in 4 simulated coin flips Biden would win 3 times while Trump wins once. Nobody here is excited about those odds, of course we want Biden winning 4 of those coin flips but the math doesn't show Trump winning 2 of those coin flips either. Biden is still the odds on favourite.Lerxst1992 said:Lerxst1992 said:Cook political report has an interesting tool where you can adjust turnout by demographic and see how that impacts electoral college.
It’s starts with trumps best demo, non college whites at 55% (Biden win w 307 electoral votes). Increasing that demo to 60% gets trump a win with about 280 EVs. 59% turnout, reduces it to 230 EVsI cant imagine a 1% turnout difference in trumps base nets a 50 EV difference, but it underscores how difficult a job pollsters have to get the polls accurate and account for even a tiny increase in trump base turnout. It ain’t helping my confidence with pollsters and if it’s this much of a difference 538 is really off their rockers w .77
HOW IT WORKS: Start with the results of the previous election, adjusted fordemographic change since 2016. Then, adjust the sliders below to see how shifts inturnout and support among five demographic groups could swing the Electoral CollegeSorry but this is damn interesting, as is the Supreme Court vs House analysis to decide contested elections. Clearly I’m in the wrong place.
This election isn't another 2016, way too many variables have changed.Not foolish, just underestimating the trump effect on elections.
Most significantly, elections are not about flipping a coin. According to Cook, Trumps 2016 performance at getting NCWs is about 4% from a huge statistical advantage, where one percent of this demo can net trump 50 EVs.He was at .55 NCW 2016 turnout. At .59 Biden still wins with @ 280 EVs. Thats great, we can all exhale. At .60 turnout, it goes to 230 and trump wins. THATS the tipping point here, this potential wave trump has access too. Being by far The largest demo (by far can not be emphasized enough), it is foolish to discount this phenomenon. That’s not even close to flipping a coin. It’s extraordinarily biased and a weighted coin.I’m hoping trump never gets there, but I am not shutting my eyes about it either. He proved in 2016 he knows how to get new voters from this demo to vote. No reason he can not find new voters there again. That’s why 77% is way off.Dude, the mistake Silver is making is this is not statistics, it’s political science. Trump has tapped into something pollsters did not identify last time. There is a better than 23% chance he can do it again.
lets fire up that time machine...again,
....PoliticsClinton Dominates in Key Philadelphia Suburbs, Bloomberg Poll Finds
Losing Pennsylvania’s 20 Electoral College votes would sharply curtail Trump’s paths to the White House.October 13, 2016, 5:00 AM EDTThe funny thing is I’d estimate Biden at .63, maybe .14 lower than 538, based on tipping point states. I think Biden is the favorite, but there is a stronger chance pollsters are making the same mistake again than .23; my main concern is trump is looking much stronger lately in these states, NC FL AZ PA. With these states within or near the moe, if having trouble believing .77 is accurate. If there is an “unpolled white vote” as there was in 2016 trump could be tied or ahead in those states.
the 2 main tipping point states, 538 has Biden at .77 in PA and .64 in AZ. The two closest tipping point states have a .13 spread. I find that debatable.
Its interesting that gauging demographic accuracy in polling is part of 538s process, but if we look at PA, they actually give biden better rating on that category (Biden 53/ trump 46). That just seems wrong, but I can not prove it, except to point out this is where pollsters got trump wrong in the past.
“ Average with demographics-based vote share projection”Scio me nihil scire
There are no kings inside the gates of eden0 -
ohmygodstatic111 said:
My real question is how do you poll the people that are trying desperately to hold on to the white privilege system and will say all day that they aren’t voting for trump,especially to pollsters, but still check his name in the privacy of the voting booth. I think Trump has a much better chance than anyone thinks based on passive white supremacy, but since it can’t be easily quantifiable and put down in numbers it’s something that could prove to be the ultimate November surprise.Lerxst1992 said:mrussel1 said:
Great... there's better than a 23% chance. So what is it, 24%? 88%? The beauty of your argument i suppose is that you can't be wrong even if you're not using math, only old articles. If Trump wins, you say...I was right! Same thing if he loses. Neither would be empirically true.Lerxst1992 said:mrussel1 said:
Dude.. sorry, you don't have the background or analysis in this topic to declare that Silver's 75 or 77% is way off. Your assumption also discounts the massive move of suburbanites toward the D. That was clear in the analysis post 2018 and the numbers are bearing out the same this time. Trump won the suburbs in 16 and he is losing them today. They are not exactly chock full of NCWs. Women have also jettisoned the R party dramatically since 16 as well. So the statement that NCW's moving from 55 to 60% turnout wins the election for Trump ignores every other demographic move.Lerxst1992 said:mrussel1 said:
This is exactly correct. Somehow people (particularly Trump supporters) think something less than 100% is equal to 101%.dignin said:
The issue I have is you calling 538's math foolish. You seem to want to say that 538 should be calling this more 50/50, when the math actually calls for around 75/25. What you don't seem to understand is that in 4 simulated coin flips Biden would win 3 times while Trump wins once. Nobody here is excited about those odds, of course we want Biden winning 4 of those coin flips but the math doesn't show Trump winning 2 of those coin flips either. Biden is still the odds on favourite.Lerxst1992 said:Lerxst1992 said:Cook political report has an interesting tool where you can adjust turnout by demographic and see how that impacts electoral college.
It’s starts with trumps best demo, non college whites at 55% (Biden win w 307 electoral votes). Increasing that demo to 60% gets trump a win with about 280 EVs. 59% turnout, reduces it to 230 EVsI cant imagine a 1% turnout difference in trumps base nets a 50 EV difference, but it underscores how difficult a job pollsters have to get the polls accurate and account for even a tiny increase in trump base turnout. It ain’t helping my confidence with pollsters and if it’s this much of a difference 538 is really off their rockers w .77
HOW IT WORKS: Start with the results of the previous election, adjusted fordemographic change since 2016. Then, adjust the sliders below to see how shifts inturnout and support among five demographic groups could swing the Electoral CollegeSorry but this is damn interesting, as is the Supreme Court vs House analysis to decide contested elections. Clearly I’m in the wrong place.
This election isn't another 2016, way too many variables have changed.Not foolish, just underestimating the trump effect on elections.
Most significantly, elections are not about flipping a coin. According to Cook, Trumps 2016 performance at getting NCWs is about 4% from a huge statistical advantage, where one percent of this demo can net trump 50 EVs.He was at .55 NCW 2016 turnout. At .59 Biden still wins with @ 280 EVs. Thats great, we can all exhale. At .60 turnout, it goes to 230 and trump wins. THATS the tipping point here, this potential wave trump has access too. Being by far The largest demo (by far can not be emphasized enough), it is foolish to discount this phenomenon. That’s not even close to flipping a coin. It’s extraordinarily biased and a weighted coin.I’m hoping trump never gets there, but I am not shutting my eyes about it either. He proved in 2016 he knows how to get new voters from this demo to vote. No reason he can not find new voters there again. That’s why 77% is way off.Dude, the mistake Silver is making is this is not statistics, it’s political science. Trump has tapped into something pollsters did not identify last time. There is a better than 23% chance he can do it again.
lets fire up that time machine...again,
....PoliticsClinton Dominates in Key Philadelphia Suburbs, Bloomberg Poll Finds
Losing Pennsylvania’s 20 Electoral College votes would sharply curtail Trump’s paths to the White House.October 13, 2016, 5:00 AM EDTThe funny thing is I’d estimate Biden at .63, maybe .14 lower than 538, based on tipping point states. I think Biden is the favorite, but there is a stronger chance pollsters are making the same mistake again than .23; my main concern is trump is looking much stronger lately in these states, NC FL AZ PA. With these states within or near the moe, if having trouble believing .77 is accurate. If there is an “unpolled white vote” as there was in 2016 trump could be tied or ahead in those states.
the 2 main tipping point states, 538 has Biden at .77 in PA and .64 in AZ. The two closest tipping point states have a .13 spread. I find that debatable.
Its interesting that gauging demographic accuracy in polling is part of 538s process, but if we look at PA, they actually give biden better rating on that category (Biden 53/ trump 46). That just seems wrong, but I can not prove it, except to point out this is where pollsters got trump wrong in the past.
“ Average with demographics-based vote share projection”
Groundhog Day. All of these threads. lol
There is no evidence of this happening. There have been hardly any difference in the live phone polling and online polling this year. Both are anonymous, by the way. But for the people who seem convinced that Trump supporters are lying to pollsters over the phone for fear of being shamed or something (I don't know), how do you explain that Biden's doing just as well with the even more anonymous online polling?
This theory makes zero sense.
Folks---it's okay to be scared. But my god, it's not all doom and gloom. Biden's in very good shape with just over a month to go. That point cannot be disputed.Post edited by The Juggler onwww.myspace.com0 -
This is the only election I have ever felt this level of insecurity around since I turned 18 and could start participating. Hopefully I am wrong and this election can be measured using the same metrics as previous elections and there is nothing to worry about.The Juggler said:
ohmygodstatic111 said:
My real question is how do you poll the people that are trying desperately to hold on to the white privilege system and will say all day that they aren’t voting for trump,especially to pollsters, but still check his name in the privacy of the voting booth. I think Trump has a much better chance than anyone thinks based on passive white supremacy, but since it can’t be easily quantifiable and put down in numbers it’s something that could prove to be the ultimate November surprise.Lerxst1992 said:mrussel1 said:
Great... there's better than a 23% chance. So what is it, 24%? 88%? The beauty of your argument i suppose is that you can't be wrong even if you're not using math, only old articles. If Trump wins, you say...I was right! Same thing if he loses. Neither would be empirically true.Lerxst1992 said:mrussel1 said:
Dude.. sorry, you don't have the background or analysis in this topic to declare that Silver's 75 or 77% is way off. Your assumption also discounts the massive move of suburbanites toward the D. That was clear in the analysis post 2018 and the numbers are bearing out the same this time. Trump won the suburbs in 16 and he is losing them today. They are not exactly chock full of NCWs. Women have also jettisoned the R party dramatically since 16 as well. So the statement that NCW's moving from 55 to 60% turnout wins the election for Trump ignores every other demographic move.Lerxst1992 said:mrussel1 said:
This is exactly correct. Somehow people (particularly Trump supporters) think something less than 100% is equal to 101%.dignin said:
The issue I have is you calling 538's math foolish. You seem to want to say that 538 should be calling this more 50/50, when the math actually calls for around 75/25. What you don't seem to understand is that in 4 simulated coin flips Biden would win 3 times while Trump wins once. Nobody here is excited about those odds, of course we want Biden winning 4 of those coin flips but the math doesn't show Trump winning 2 of those coin flips either. Biden is still the odds on favourite.Lerxst1992 said:Lerxst1992 said:Cook political report has an interesting tool where you can adjust turnout by demographic and see how that impacts electoral college.
It’s starts with trumps best demo, non college whites at 55% (Biden win w 307 electoral votes). Increasing that demo to 60% gets trump a win with about 280 EVs. 59% turnout, reduces it to 230 EVsI cant imagine a 1% turnout difference in trumps base nets a 50 EV difference, but it underscores how difficult a job pollsters have to get the polls accurate and account for even a tiny increase in trump base turnout. It ain’t helping my confidence with pollsters and if it’s this much of a difference 538 is really off their rockers w .77
HOW IT WORKS: Start with the results of the previous election, adjusted fordemographic change since 2016. Then, adjust the sliders below to see how shifts inturnout and support among five demographic groups could swing the Electoral CollegeSorry but this is damn interesting, as is the Supreme Court vs House analysis to decide contested elections. Clearly I’m in the wrong place.
This election isn't another 2016, way too many variables have changed.Not foolish, just underestimating the trump effect on elections.
Most significantly, elections are not about flipping a coin. According to Cook, Trumps 2016 performance at getting NCWs is about 4% from a huge statistical advantage, where one percent of this demo can net trump 50 EVs.He was at .55 NCW 2016 turnout. At .59 Biden still wins with @ 280 EVs. Thats great, we can all exhale. At .60 turnout, it goes to 230 and trump wins. THATS the tipping point here, this potential wave trump has access too. Being by far The largest demo (by far can not be emphasized enough), it is foolish to discount this phenomenon. That’s not even close to flipping a coin. It’s extraordinarily biased and a weighted coin.I’m hoping trump never gets there, but I am not shutting my eyes about it either. He proved in 2016 he knows how to get new voters from this demo to vote. No reason he can not find new voters there again. That’s why 77% is way off.Dude, the mistake Silver is making is this is not statistics, it’s political science. Trump has tapped into something pollsters did not identify last time. There is a better than 23% chance he can do it again.
lets fire up that time machine...again,
....PoliticsClinton Dominates in Key Philadelphia Suburbs, Bloomberg Poll Finds
Losing Pennsylvania’s 20 Electoral College votes would sharply curtail Trump’s paths to the White House.October 13, 2016, 5:00 AM EDTThe funny thing is I’d estimate Biden at .63, maybe .14 lower than 538, based on tipping point states. I think Biden is the favorite, but there is a stronger chance pollsters are making the same mistake again than .23; my main concern is trump is looking much stronger lately in these states, NC FL AZ PA. With these states within or near the moe, if having trouble believing .77 is accurate. If there is an “unpolled white vote” as there was in 2016 trump could be tied or ahead in those states.
the 2 main tipping point states, 538 has Biden at .77 in PA and .64 in AZ. The two closest tipping point states have a .13 spread. I find that debatable.
Its interesting that gauging demographic accuracy in polling is part of 538s process, but if we look at PA, they actually give biden better rating on that category (Biden 53/ trump 46). That just seems wrong, but I can not prove it, except to point out this is where pollsters got trump wrong in the past.
“ Average with demographics-based vote share projection”
Groundhog Day. All of these threads. lol
There is no evidence of this happening. There have been hardly any difference in the live phone polling and online polling this year. Both are anonymous, by the way. But for the people who seem convinced that Trump supporters are lying to pollsters over the phone for fear of being shamed or something (I don't know), how do you explain that Biden's doing just as well with the even more anonymous online polling?
This theory makes zero sense.
Folks---it's okay to be scared. But my god, it's not all doom and gloom. Biden's in very good shape with just over a month to go. That point cannot be disputed.Scio me nihil scire
There are no kings inside the gates of eden0 -
I actually have Biden at 68.94% based on some online articles i read0
-
You were not worried four years ago??static111 said:
This is the only election I have ever felt this level of insecurity around since I turned 18 and could start participating. Hopefully I am wrong and this election can be measured using the same metrics as previous elections and there is nothing to worry about.The Juggler said:
ohmygodstatic111 said:
My real question is how do you poll the people that are trying desperately to hold on to the white privilege system and will say all day that they aren’t voting for trump,especially to pollsters, but still check his name in the privacy of the voting booth. I think Trump has a much better chance than anyone thinks based on passive white supremacy, but since it can’t be easily quantifiable and put down in numbers it’s something that could prove to be the ultimate November surprise.Lerxst1992 said:mrussel1 said:
Great... there's better than a 23% chance. So what is it, 24%? 88%? The beauty of your argument i suppose is that you can't be wrong even if you're not using math, only old articles. If Trump wins, you say...I was right! Same thing if he loses. Neither would be empirically true.Lerxst1992 said:mrussel1 said:
Dude.. sorry, you don't have the background or analysis in this topic to declare that Silver's 75 or 77% is way off. Your assumption also discounts the massive move of suburbanites toward the D. That was clear in the analysis post 2018 and the numbers are bearing out the same this time. Trump won the suburbs in 16 and he is losing them today. They are not exactly chock full of NCWs. Women have also jettisoned the R party dramatically since 16 as well. So the statement that NCW's moving from 55 to 60% turnout wins the election for Trump ignores every other demographic move.Lerxst1992 said:mrussel1 said:
This is exactly correct. Somehow people (particularly Trump supporters) think something less than 100% is equal to 101%.dignin said:
The issue I have is you calling 538's math foolish. You seem to want to say that 538 should be calling this more 50/50, when the math actually calls for around 75/25. What you don't seem to understand is that in 4 simulated coin flips Biden would win 3 times while Trump wins once. Nobody here is excited about those odds, of course we want Biden winning 4 of those coin flips but the math doesn't show Trump winning 2 of those coin flips either. Biden is still the odds on favourite.Lerxst1992 said:Lerxst1992 said:Cook political report has an interesting tool where you can adjust turnout by demographic and see how that impacts electoral college.
It’s starts with trumps best demo, non college whites at 55% (Biden win w 307 electoral votes). Increasing that demo to 60% gets trump a win with about 280 EVs. 59% turnout, reduces it to 230 EVsI cant imagine a 1% turnout difference in trumps base nets a 50 EV difference, but it underscores how difficult a job pollsters have to get the polls accurate and account for even a tiny increase in trump base turnout. It ain’t helping my confidence with pollsters and if it’s this much of a difference 538 is really off their rockers w .77
HOW IT WORKS: Start with the results of the previous election, adjusted fordemographic change since 2016. Then, adjust the sliders below to see how shifts inturnout and support among five demographic groups could swing the Electoral CollegeSorry but this is damn interesting, as is the Supreme Court vs House analysis to decide contested elections. Clearly I’m in the wrong place.
This election isn't another 2016, way too many variables have changed.Not foolish, just underestimating the trump effect on elections.
Most significantly, elections are not about flipping a coin. According to Cook, Trumps 2016 performance at getting NCWs is about 4% from a huge statistical advantage, where one percent of this demo can net trump 50 EVs.He was at .55 NCW 2016 turnout. At .59 Biden still wins with @ 280 EVs. Thats great, we can all exhale. At .60 turnout, it goes to 230 and trump wins. THATS the tipping point here, this potential wave trump has access too. Being by far The largest demo (by far can not be emphasized enough), it is foolish to discount this phenomenon. That’s not even close to flipping a coin. It’s extraordinarily biased and a weighted coin.I’m hoping trump never gets there, but I am not shutting my eyes about it either. He proved in 2016 he knows how to get new voters from this demo to vote. No reason he can not find new voters there again. That’s why 77% is way off.Dude, the mistake Silver is making is this is not statistics, it’s political science. Trump has tapped into something pollsters did not identify last time. There is a better than 23% chance he can do it again.
lets fire up that time machine...again,
....PoliticsClinton Dominates in Key Philadelphia Suburbs, Bloomberg Poll Finds
Losing Pennsylvania’s 20 Electoral College votes would sharply curtail Trump’s paths to the White House.October 13, 2016, 5:00 AM EDTThe funny thing is I’d estimate Biden at .63, maybe .14 lower than 538, based on tipping point states. I think Biden is the favorite, but there is a stronger chance pollsters are making the same mistake again than .23; my main concern is trump is looking much stronger lately in these states, NC FL AZ PA. With these states within or near the moe, if having trouble believing .77 is accurate. If there is an “unpolled white vote” as there was in 2016 trump could be tied or ahead in those states.
the 2 main tipping point states, 538 has Biden at .77 in PA and .64 in AZ. The two closest tipping point states have a .13 spread. I find that debatable.
Its interesting that gauging demographic accuracy in polling is part of 538s process, but if we look at PA, they actually give biden better rating on that category (Biden 53/ trump 46). That just seems wrong, but I can not prove it, except to point out this is where pollsters got trump wrong in the past.
“ Average with demographics-based vote share projection”
Groundhog Day. All of these threads. lol
There is no evidence of this happening. There have been hardly any difference in the live phone polling and online polling this year. Both are anonymous, by the way. But for the people who seem convinced that Trump supporters are lying to pollsters over the phone for fear of being shamed or something (I don't know), how do you explain that Biden's doing just as well with the even more anonymous online polling?
This theory makes zero sense.
Folks---it's okay to be scared. But my god, it's not all doom and gloom. Biden's in very good shape with just over a month to go. That point cannot be disputed.www.myspace.com0 -
mrussel1 said:
I'll agree with that, although we probably differ at the thousandths decimal, so let's not go there.MayDay10 said:I actually have Biden at 68.94% based on some online articles i read
Anything over 68.91% seems a little out there...
1995 Milwaukee 1998 Alpine, Alpine 2003 Albany, Boston, Boston, Boston 2004 Boston, Boston 2006 Hartford, St. Paul (Petty), St. Paul (Petty) 2011 Alpine, Alpine
2013 Wrigley 2014 St. Paul 2016 Fenway, Fenway, Wrigley, Wrigley 2018 Missoula, Wrigley, Wrigley 2021 Asbury Park 2022 St Louis 2023 Austin, Austin
2024 Napa, Wrigley, Wrigley0 -
i agree with juggler in that i don't know why anyone would lie to a pollster when it's anonymous, unless there's some unfounded conspiracy theory/paranoia out there on the right that it's not really anonymous?By The Time They Figure Out What Went Wrong, We'll Be Sitting On A Beach, Earning Twenty Percent.0
-
No I knew trump was gonna win because of the closet white supremacists in 2016 and because of Hillary’s poor campaign. I don’t think this is a repeat of 2016. I think it is a different animal altogether that we are viewing through the 2016 lens and saying wow look how much better we are doing compared to 2016, rather than focusing on what unique unknowns exist in 2020.The Juggler said:
You were not worried four years ago??static111 said:
This is the only election I have ever felt this level of insecurity around since I turned 18 and could start participating. Hopefully I am wrong and this election can be measured using the same metrics as previous elections and there is nothing to worry about.The Juggler said:
ohmygodstatic111 said:
My real question is how do you poll the people that are trying desperately to hold on to the white privilege system and will say all day that they aren’t voting for trump,especially to pollsters, but still check his name in the privacy of the voting booth. I think Trump has a much better chance than anyone thinks based on passive white supremacy, but since it can’t be easily quantifiable and put down in numbers it’s something that could prove to be the ultimate November surprise.Lerxst1992 said:mrussel1 said:
Great... there's better than a 23% chance. So what is it, 24%? 88%? The beauty of your argument i suppose is that you can't be wrong even if you're not using math, only old articles. If Trump wins, you say...I was right! Same thing if he loses. Neither would be empirically true.Lerxst1992 said:mrussel1 said:
Dude.. sorry, you don't have the background or analysis in this topic to declare that Silver's 75 or 77% is way off. Your assumption also discounts the massive move of suburbanites toward the D. That was clear in the analysis post 2018 and the numbers are bearing out the same this time. Trump won the suburbs in 16 and he is losing them today. They are not exactly chock full of NCWs. Women have also jettisoned the R party dramatically since 16 as well. So the statement that NCW's moving from 55 to 60% turnout wins the election for Trump ignores every other demographic move.Lerxst1992 said:mrussel1 said:
This is exactly correct. Somehow people (particularly Trump supporters) think something less than 100% is equal to 101%.dignin said:
The issue I have is you calling 538's math foolish. You seem to want to say that 538 should be calling this more 50/50, when the math actually calls for around 75/25. What you don't seem to understand is that in 4 simulated coin flips Biden would win 3 times while Trump wins once. Nobody here is excited about those odds, of course we want Biden winning 4 of those coin flips but the math doesn't show Trump winning 2 of those coin flips either. Biden is still the odds on favourite.Lerxst1992 said:Lerxst1992 said:Cook political report has an interesting tool where you can adjust turnout by demographic and see how that impacts electoral college.
It’s starts with trumps best demo, non college whites at 55% (Biden win w 307 electoral votes). Increasing that demo to 60% gets trump a win with about 280 EVs. 59% turnout, reduces it to 230 EVsI cant imagine a 1% turnout difference in trumps base nets a 50 EV difference, but it underscores how difficult a job pollsters have to get the polls accurate and account for even a tiny increase in trump base turnout. It ain’t helping my confidence with pollsters and if it’s this much of a difference 538 is really off their rockers w .77
HOW IT WORKS: Start with the results of the previous election, adjusted fordemographic change since 2016. Then, adjust the sliders below to see how shifts inturnout and support among five demographic groups could swing the Electoral CollegeSorry but this is damn interesting, as is the Supreme Court vs House analysis to decide contested elections. Clearly I’m in the wrong place.
This election isn't another 2016, way too many variables have changed.Not foolish, just underestimating the trump effect on elections.
Most significantly, elections are not about flipping a coin. According to Cook, Trumps 2016 performance at getting NCWs is about 4% from a huge statistical advantage, where one percent of this demo can net trump 50 EVs.He was at .55 NCW 2016 turnout. At .59 Biden still wins with @ 280 EVs. Thats great, we can all exhale. At .60 turnout, it goes to 230 and trump wins. THATS the tipping point here, this potential wave trump has access too. Being by far The largest demo (by far can not be emphasized enough), it is foolish to discount this phenomenon. That’s not even close to flipping a coin. It’s extraordinarily biased and a weighted coin.I’m hoping trump never gets there, but I am not shutting my eyes about it either. He proved in 2016 he knows how to get new voters from this demo to vote. No reason he can not find new voters there again. That’s why 77% is way off.Dude, the mistake Silver is making is this is not statistics, it’s political science. Trump has tapped into something pollsters did not identify last time. There is a better than 23% chance he can do it again.
lets fire up that time machine...again,
....PoliticsClinton Dominates in Key Philadelphia Suburbs, Bloomberg Poll Finds
Losing Pennsylvania’s 20 Electoral College votes would sharply curtail Trump’s paths to the White House.October 13, 2016, 5:00 AM EDTThe funny thing is I’d estimate Biden at .63, maybe .14 lower than 538, based on tipping point states. I think Biden is the favorite, but there is a stronger chance pollsters are making the same mistake again than .23; my main concern is trump is looking much stronger lately in these states, NC FL AZ PA. With these states within or near the moe, if having trouble believing .77 is accurate. If there is an “unpolled white vote” as there was in 2016 trump could be tied or ahead in those states.
the 2 main tipping point states, 538 has Biden at .77 in PA and .64 in AZ. The two closest tipping point states have a .13 spread. I find that debatable.
Its interesting that gauging demographic accuracy in polling is part of 538s process, but if we look at PA, they actually give biden better rating on that category (Biden 53/ trump 46). That just seems wrong, but I can not prove it, except to point out this is where pollsters got trump wrong in the past.
“ Average with demographics-based vote share projection”
Groundhog Day. All of these threads. lol
There is no evidence of this happening. There have been hardly any difference in the live phone polling and online polling this year. Both are anonymous, by the way. But for the people who seem convinced that Trump supporters are lying to pollsters over the phone for fear of being shamed or something (I don't know), how do you explain that Biden's doing just as well with the even more anonymous online polling?
This theory makes zero sense.
Folks---it's okay to be scared. But my god, it's not all doom and gloom. Biden's in very good shape with just over a month to go. That point cannot be disputed.
i mean I’ve already seen and heard coworkers talk about how the $750 tax is no big deal because rich people make their money in a different way and it gets taxed differently.There are a lot of things going on with framing the election in a way that Don can challenge the results in a close election. So the polling may well be right on and Don still “wins” I’m hoping for a high turnout to offset this.Post edited by static111 onScio me nihil scire
There are no kings inside the gates of eden0 -
I agree that there are other external factors to be worried about. The GOP are going to do everything in their power to "steal" this election. They have been openly talking about it for some time.static111 said:
No I knew trump was gonna win because of the closet white supremacists in 2016 and because of Hillary’s poor campaign. I don’t think this is a repeat of 2016. I think it is a different animal altogether that we are viewing through the 2016 lens and saying wow look how much better we are doing compared to 2016, rather than focusing on what unique unknowns exist in 2020.IThe Juggler said:
You were not worried four years ago??static111 said:
This is the only election I have ever felt this level of insecurity around since I turned 18 and could start participating. Hopefully I am wrong and this election can be measured using the same metrics as previous elections and there is nothing to worry about.The Juggler said:
ohmygodstatic111 said:
My real question is how do you poll the people that are trying desperately to hold on to the white privilege system and will say all day that they aren’t voting for trump,especially to pollsters, but still check his name in the privacy of the voting booth. I think Trump has a much better chance than anyone thinks based on passive white supremacy, but since it can’t be easily quantifiable and put down in numbers it’s something that could prove to be the ultimate November surprise.Lerxst1992 said:mrussel1 said:
Great... there's better than a 23% chance. So what is it, 24%? 88%? The beauty of your argument i suppose is that you can't be wrong even if you're not using math, only old articles. If Trump wins, you say...I was right! Same thing if he loses. Neither would be empirically true.Lerxst1992 said:mrussel1 said:
Dude.. sorry, you don't have the background or analysis in this topic to declare that Silver's 75 or 77% is way off. Your assumption also discounts the massive move of suburbanites toward the D. That was clear in the analysis post 2018 and the numbers are bearing out the same this time. Trump won the suburbs in 16 and he is losing them today. They are not exactly chock full of NCWs. Women have also jettisoned the R party dramatically since 16 as well. So the statement that NCW's moving from 55 to 60% turnout wins the election for Trump ignores every other demographic move.Lerxst1992 said:mrussel1 said:
This is exactly correct. Somehow people (particularly Trump supporters) think something less than 100% is equal to 101%.dignin said:
The issue I have is you calling 538's math foolish. You seem to want to say that 538 should be calling this more 50/50, when the math actually calls for around 75/25. What you don't seem to understand is that in 4 simulated coin flips Biden would win 3 times while Trump wins once. Nobody here is excited about those odds, of course we want Biden winning 4 of those coin flips but the math doesn't show Trump winning 2 of those coin flips either. Biden is still the odds on favourite.Lerxst1992 said:Lerxst1992 said:Cook political report has an interesting tool where you can adjust turnout by demographic and see how that impacts electoral college.
It’s starts with trumps best demo, non college whites at 55% (Biden win w 307 electoral votes). Increasing that demo to 60% gets trump a win with about 280 EVs. 59% turnout, reduces it to 230 EVsI cant imagine a 1% turnout difference in trumps base nets a 50 EV difference, but it underscores how difficult a job pollsters have to get the polls accurate and account for even a tiny increase in trump base turnout. It ain’t helping my confidence with pollsters and if it’s this much of a difference 538 is really off their rockers w .77
HOW IT WORKS: Start with the results of the previous election, adjusted fordemographic change since 2016. Then, adjust the sliders below to see how shifts inturnout and support among five demographic groups could swing the Electoral CollegeSorry but this is damn interesting, as is the Supreme Court vs House analysis to decide contested elections. Clearly I’m in the wrong place.
This election isn't another 2016, way too many variables have changed.Not foolish, just underestimating the trump effect on elections.
Most significantly, elections are not about flipping a coin. According to Cook, Trumps 2016 performance at getting NCWs is about 4% from a huge statistical advantage, where one percent of this demo can net trump 50 EVs.He was at .55 NCW 2016 turnout. At .59 Biden still wins with @ 280 EVs. Thats great, we can all exhale. At .60 turnout, it goes to 230 and trump wins. THATS the tipping point here, this potential wave trump has access too. Being by far The largest demo (by far can not be emphasized enough), it is foolish to discount this phenomenon. That’s not even close to flipping a coin. It’s extraordinarily biased and a weighted coin.I’m hoping trump never gets there, but I am not shutting my eyes about it either. He proved in 2016 he knows how to get new voters from this demo to vote. No reason he can not find new voters there again. That’s why 77% is way off.Dude, the mistake Silver is making is this is not statistics, it’s political science. Trump has tapped into something pollsters did not identify last time. There is a better than 23% chance he can do it again.
lets fire up that time machine...again,
....PoliticsClinton Dominates in Key Philadelphia Suburbs, Bloomberg Poll Finds
Losing Pennsylvania’s 20 Electoral College votes would sharply curtail Trump’s paths to the White House.October 13, 2016, 5:00 AM EDTThe funny thing is I’d estimate Biden at .63, maybe .14 lower than 538, based on tipping point states. I think Biden is the favorite, but there is a stronger chance pollsters are making the same mistake again than .23; my main concern is trump is looking much stronger lately in these states, NC FL AZ PA. With these states within or near the moe, if having trouble believing .77 is accurate. If there is an “unpolled white vote” as there was in 2016 trump could be tied or ahead in those states.
the 2 main tipping point states, 538 has Biden at .77 in PA and .64 in AZ. The two closest tipping point states have a .13 spread. I find that debatable.
Its interesting that gauging demographic accuracy in polling is part of 538s process, but if we look at PA, they actually give biden better rating on that category (Biden 53/ trump 46). That just seems wrong, but I can not prove it, except to point out this is where pollsters got trump wrong in the past.
“ Average with demographics-based vote share projection”
Groundhog Day. All of these threads. lol
There is no evidence of this happening. There have been hardly any difference in the live phone polling and online polling this year. Both are anonymous, by the way. But for the people who seem convinced that Trump supporters are lying to pollsters over the phone for fear of being shamed or something (I don't know), how do you explain that Biden's doing just as well with the even more anonymous online polling?
This theory makes zero sense.
Folks---it's okay to be scared. But my god, it's not all doom and gloom. Biden's in very good shape with just over a month to go. That point cannot be disputed.
i mean I’ve already seen and heard coworkers talk about how the $750 tax is no big deal because rich people make their money in a different way and it gets taxed differently.There are a lot of things going on with framing the election in a way that Don can challenge the results in a close election. So the polling may well be right on and Don’t still “wins” I’m hoping for a high turnout to offset this.
Polling, shy Trump voters is way, way down on this list of my concerns.
0 -
The popular vote differential needs to be at least 10 million. That would be hard to argue. I'd like to see Biden cross the 70 million mark because it will be near impossible for Trump to crack 60 million this time. We've lost the 3rd party vote splitters and Trump has not convinced any sane people to move their vote to him. What he has gained are the 18+ offspring of his base. Still not enough to account for those he has lost, both to death and misguided hope he would be presidential. An absolute beating will be the only way to make this orange stain slink off in to the shadows of his private life and hope he doesn't get indicted on one of his many illegal activities. The plea deal will probably be for him to shut the fuck up and go away or be prosecuted and his family name burned to the ground, not that we shouldn't be doing that anyway to avoid a rise to national power by one of his children in 20 years.static111 said:
No I knew trump was gonna win because of the closet white supremacists in 2016 and because of Hillary’s poor campaign. I don’t think this is a repeat of 2016. I think it is a different animal altogether that we are viewing through the 2016 lens and saying wow look how much better we are doing compared to 2016, rather than focusing on what unique unknowns exist in 2020.The Juggler said:
You were not worried four years ago??static111 said:
This is the only election I have ever felt this level of insecurity around since I turned 18 and could start participating. Hopefully I am wrong and this election can be measured using the same metrics as previous elections and there is nothing to worry about.The Juggler said:
ohmygodstatic111 said:
My real question is how do you poll the people that are trying desperately to hold on to the white privilege system and will say all day that they aren’t voting for trump,especially to pollsters, but still check his name in the privacy of the voting booth. I think Trump has a much better chance than anyone thinks based on passive white supremacy, but since it can’t be easily quantifiable and put down in numbers it’s something that could prove to be the ultimate November surprise.Lerxst1992 said:mrussel1 said:
Great... there's better than a 23% chance. So what is it, 24%? 88%? The beauty of your argument i suppose is that you can't be wrong even if you're not using math, only old articles. If Trump wins, you say...I was right! Same thing if he loses. Neither would be empirically true.Lerxst1992 said:mrussel1 said:
Dude.. sorry, you don't have the background or analysis in this topic to declare that Silver's 75 or 77% is way off. Your assumption also discounts the massive move of suburbanites toward the D. That was clear in the analysis post 2018 and the numbers are bearing out the same this time. Trump won the suburbs in 16 and he is losing them today. They are not exactly chock full of NCWs. Women have also jettisoned the R party dramatically since 16 as well. So the statement that NCW's moving from 55 to 60% turnout wins the election for Trump ignores every other demographic move.Lerxst1992 said:mrussel1 said:
This is exactly correct. Somehow people (particularly Trump supporters) think something less than 100% is equal to 101%.dignin said:
The issue I have is you calling 538's math foolish. You seem to want to say that 538 should be calling this more 50/50, when the math actually calls for around 75/25. What you don't seem to understand is that in 4 simulated coin flips Biden would win 3 times while Trump wins once. Nobody here is excited about those odds, of course we want Biden winning 4 of those coin flips but the math doesn't show Trump winning 2 of those coin flips either. Biden is still the odds on favourite.Lerxst1992 said:Lerxst1992 said:Cook political report has an interesting tool where you can adjust turnout by demographic and see how that impacts electoral college.
It’s starts with trumps best demo, non college whites at 55% (Biden win w 307 electoral votes). Increasing that demo to 60% gets trump a win with about 280 EVs. 59% turnout, reduces it to 230 EVsI cant imagine a 1% turnout difference in trumps base nets a 50 EV difference, but it underscores how difficult a job pollsters have to get the polls accurate and account for even a tiny increase in trump base turnout. It ain’t helping my confidence with pollsters and if it’s this much of a difference 538 is really off their rockers w .77
HOW IT WORKS: Start with the results of the previous election, adjusted fordemographic change since 2016. Then, adjust the sliders below to see how shifts inturnout and support among five demographic groups could swing the Electoral CollegeSorry but this is damn interesting, as is the Supreme Court vs House analysis to decide contested elections. Clearly I’m in the wrong place.
This election isn't another 2016, way too many variables have changed.Not foolish, just underestimating the trump effect on elections.
Most significantly, elections are not about flipping a coin. According to Cook, Trumps 2016 performance at getting NCWs is about 4% from a huge statistical advantage, where one percent of this demo can net trump 50 EVs.He was at .55 NCW 2016 turnout. At .59 Biden still wins with @ 280 EVs. Thats great, we can all exhale. At .60 turnout, it goes to 230 and trump wins. THATS the tipping point here, this potential wave trump has access too. Being by far The largest demo (by far can not be emphasized enough), it is foolish to discount this phenomenon. That’s not even close to flipping a coin. It’s extraordinarily biased and a weighted coin.I’m hoping trump never gets there, but I am not shutting my eyes about it either. He proved in 2016 he knows how to get new voters from this demo to vote. No reason he can not find new voters there again. That’s why 77% is way off.Dude, the mistake Silver is making is this is not statistics, it’s political science. Trump has tapped into something pollsters did not identify last time. There is a better than 23% chance he can do it again.
lets fire up that time machine...again,
....PoliticsClinton Dominates in Key Philadelphia Suburbs, Bloomberg Poll Finds
Losing Pennsylvania’s 20 Electoral College votes would sharply curtail Trump’s paths to the White House.October 13, 2016, 5:00 AM EDTThe funny thing is I’d estimate Biden at .63, maybe .14 lower than 538, based on tipping point states. I think Biden is the favorite, but there is a stronger chance pollsters are making the same mistake again than .23; my main concern is trump is looking much stronger lately in these states, NC FL AZ PA. With these states within or near the moe, if having trouble believing .77 is accurate. If there is an “unpolled white vote” as there was in 2016 trump could be tied or ahead in those states.
the 2 main tipping point states, 538 has Biden at .77 in PA and .64 in AZ. The two closest tipping point states have a .13 spread. I find that debatable.
Its interesting that gauging demographic accuracy in polling is part of 538s process, but if we look at PA, they actually give biden better rating on that category (Biden 53/ trump 46). That just seems wrong, but I can not prove it, except to point out this is where pollsters got trump wrong in the past.
“ Average with demographics-based vote share projection”
Groundhog Day. All of these threads. lol
There is no evidence of this happening. There have been hardly any difference in the live phone polling and online polling this year. Both are anonymous, by the way. But for the people who seem convinced that Trump supporters are lying to pollsters over the phone for fear of being shamed or something (I don't know), how do you explain that Biden's doing just as well with the even more anonymous online polling?
This theory makes zero sense.
Folks---it's okay to be scared. But my god, it's not all doom and gloom. Biden's in very good shape with just over a month to go. That point cannot be disputed.
i mean I’ve already seen and heard coworkers talk about how the $750 tax is no big deal because rich people make their money in a different way and it gets taxed differently.There are a lot of things going on with framing the election in a way that Don can challenge the results in a close election. So the polling may well be right on and Don’t still “wins” I’m hoping for a high turnout to offset this.It's a hopeless situation...0 -
My own personal observation is that due to a lack of Team Trump Treason yard signs and many Biden yard signs in NH and a very red district in another New England state, I'm sensing that the Team Trump Treason enthusiasm from 2016 is not there and that voters are tired of his schtick, the inept covid response and the slights against the military having worn people down, mainly dems and indies that took a chance on him in 2016. That said, his deplorable base will continue to support him rabidly, regardless of what he does or what comes out.
My uneducated guess and prediction: Biden wins with 344 electoral votes.09/15/1998 & 09/16/1998, Mansfield, MA; 08/29/00 08/30/00, Mansfield, MA; 07/02/03, 07/03/03, Mansfield, MA; 09/28/04, 09/29/04, Boston, MA; 09/22/05, Halifax, NS; 05/24/06, 05/25/06, Boston, MA; 07/22/06, 07/23/06, Gorge, WA; 06/27/2008, Hartford; 06/28/08, 06/30/08, Mansfield; 08/18/2009, O2, London, UK; 10/30/09, 10/31/09, Philadelphia, PA; 05/15/10, Hartford, CT; 05/17/10, Boston, MA; 05/20/10, 05/21/10, NY, NY; 06/22/10, Dublin, IRE; 06/23/10, Northern Ireland; 09/03/11, 09/04/11, Alpine Valley, WI; 09/11/11, 09/12/11, Toronto, Ont; 09/14/11, Ottawa, Ont; 09/15/11, Hamilton, Ont; 07/02/2012, Prague, Czech Republic; 07/04/2012 & 07/05/2012, Berlin, Germany; 07/07/2012, Stockholm, Sweden; 09/30/2012, Missoula, MT; 07/16/2013, London, Ont; 07/19/2013, Chicago, IL; 10/15/2013 & 10/16/2013, Worcester, MA; 10/21/2013 & 10/22/2013, Philadelphia, PA; 10/25/2013, Hartford, CT; 11/29/2013, Portland, OR; 11/30/2013, Spokane, WA; 12/04/2013, Vancouver, BC; 12/06/2013, Seattle, WA; 10/03/2014, St. Louis. MO; 10/22/2014, Denver, CO; 10/26/2015, New York, NY; 04/23/2016, New Orleans, LA; 04/28/2016 & 04/29/2016, Philadelphia, PA; 05/01/2016 & 05/02/2016, New York, NY; 05/08/2016, Ottawa, Ont.; 05/10/2016 & 05/12/2016, Toronto, Ont.; 08/05/2016 & 08/07/2016, Boston, MA; 08/20/2016 & 08/22/2016, Chicago, IL; 07/01/2018, Prague, Czech Republic; 07/03/2018, Krakow, Poland; 07/05/2018, Berlin, Germany; 09/02/2018 & 09/04/2018, Boston, MA; 09/08/2022, Toronto, Ont; 09/11/2022, New York, NY; 09/14/2022, Camden, NJ; 09/02/2023, St. Paul, MN; 05/04/2024 & 05/06/2024, Vancouver, BC; 05/10/2024, Portland, OR;
Libtardaplorable©. And proud of it.
Brilliantati©0 -
September 27, 2020 at 2:10 p.m. EDTAdd to list
While the conventional wisdom seems to be that the election will be very close, opening up all sorts of mischief from President Trump, his attorney general and Republicans, let’s look at the facts. They suggest this race is not close.
The most recent Washington Post-ABC News poll shows former vice president Joe Biden leading by 10 points — a huge margin — with voting already underway. The New York Times-Siena College poll shows Biden with an eight-point lead. By historic standards, a presidential election with such a lead in the popular vote would be somewhere between “commanding” and a “blowout.”
Contrary to conventional wisdom (in sync with my own suspicion), the Supreme Court opening is revving up Democrats. As The Post reports, “64 percent of Biden supporters say the court vacancy makes it ‘more important’ that he win the election, compared with 37 percent of Trump supporters who say the same about their candidate.” In the New York Times-Siena polling, the court fight looks like a loser for Trump and his Republican enablers. Per the Times: “56 percent said they preferred to have the election act as a sort of referendum on the vacancy. Only 41 percent said they wanted Mr. Trump to choose a justice before November.” Worse still:
More striking, the voters Mr. Trump and endangered Senate Republicans must reclaim to close the gap in the polls are even more opposed to a hasty pick: 62 percent of women, 63 percent of independents and 60 percent of college-educated white voters said they wanted the winner of the campaign to fill the seat.The warning signs for Republicans are also stark on the issue of abortion, on which Judge Barrett, a fiercely conservative jurist, could offer a pivotal vote should she be confirmed: 60 percent of those surveyed believe abortion should be legal all or some of the time.The Affordable Care Act, which Republicans are counting on the Supreme Court to gut, also remains very popular (57 to 38), according to the Times-Siena poll.
In the presidential race, Biden is doing better — much better in some cases — with groups Hillary Clinton lost big. The Post reports: “Biden leads among White women with college degrees by 41 points and is almost even among White women without degrees. In 2016, Clinton lost White women without college degrees by 23 points.” Biden’s improvement both with Whites without college degrees and college-educated Whites is remarkable. (“Clinton won college-graduate voters overall by 21 points in Pew’s survey of confirmed voters, and she lost those without a college education by seven points. Today, Biden is ahead among college-educated voters overall by 30 points and narrowly behind among those without college degrees, by six points.”) Biden also leads among older voters, a reverse of 2016.
The New York Times-Siena polls shows similar results. “Mr. Biden is winning 60 percent of white women with college degrees, compared with 34 percent for Mr. Trump, and he is beating the president among men with college degrees, 50 percent to 45 percent,” the pollsters found. “Four years ago, according to exit polls, Ms. Clinton won college-educated white women by only seven percentage points and lost college-educated white men to Mr. Trump by 14 points.”
These big margins are showing up in key state polls as well. Biden’s lead in the FiveThirtyEight averages is about seven points in Michigan and Wisconsin, about five in Pennsylvania and about 3.5 in Arizona. Biden is statistically tied in states Trump won easily in 2016 and should have wrapped up weeks ago. (e.g., North Carolina, Florida, Texas, Georgia).
Polls can be wrong, but many polls would have to be really wrong to provide comfort to Republicans. The candidates are set to debate Tuesday — although Trump, in meltdown mode, is hollering for Biden to take a drug test before or after. (Maybe Biden should agree, if Trump releases his tax returns.) Biden could, I guess, have an outing so terrible as to alarm voters. Barring that and noting that voting is well underway in many states, Biden is the heavy favorite at this point. Republicans down ticket should worry about Trump dragging them over the cliff into political oblivion.
https://www.washingtonpost.com/opinions/2020/09/27/chances-blowout-are-high/
09/15/1998 & 09/16/1998, Mansfield, MA; 08/29/00 08/30/00, Mansfield, MA; 07/02/03, 07/03/03, Mansfield, MA; 09/28/04, 09/29/04, Boston, MA; 09/22/05, Halifax, NS; 05/24/06, 05/25/06, Boston, MA; 07/22/06, 07/23/06, Gorge, WA; 06/27/2008, Hartford; 06/28/08, 06/30/08, Mansfield; 08/18/2009, O2, London, UK; 10/30/09, 10/31/09, Philadelphia, PA; 05/15/10, Hartford, CT; 05/17/10, Boston, MA; 05/20/10, 05/21/10, NY, NY; 06/22/10, Dublin, IRE; 06/23/10, Northern Ireland; 09/03/11, 09/04/11, Alpine Valley, WI; 09/11/11, 09/12/11, Toronto, Ont; 09/14/11, Ottawa, Ont; 09/15/11, Hamilton, Ont; 07/02/2012, Prague, Czech Republic; 07/04/2012 & 07/05/2012, Berlin, Germany; 07/07/2012, Stockholm, Sweden; 09/30/2012, Missoula, MT; 07/16/2013, London, Ont; 07/19/2013, Chicago, IL; 10/15/2013 & 10/16/2013, Worcester, MA; 10/21/2013 & 10/22/2013, Philadelphia, PA; 10/25/2013, Hartford, CT; 11/29/2013, Portland, OR; 11/30/2013, Spokane, WA; 12/04/2013, Vancouver, BC; 12/06/2013, Seattle, WA; 10/03/2014, St. Louis. MO; 10/22/2014, Denver, CO; 10/26/2015, New York, NY; 04/23/2016, New Orleans, LA; 04/28/2016 & 04/29/2016, Philadelphia, PA; 05/01/2016 & 05/02/2016, New York, NY; 05/08/2016, Ottawa, Ont.; 05/10/2016 & 05/12/2016, Toronto, Ont.; 08/05/2016 & 08/07/2016, Boston, MA; 08/20/2016 & 08/22/2016, Chicago, IL; 07/01/2018, Prague, Czech Republic; 07/03/2018, Krakow, Poland; 07/05/2018, Berlin, Germany; 09/02/2018 & 09/04/2018, Boston, MA; 09/08/2022, Toronto, Ont; 09/11/2022, New York, NY; 09/14/2022, Camden, NJ; 09/02/2023, St. Paul, MN; 05/04/2024 & 05/06/2024, Vancouver, BC; 05/10/2024, Portland, OR;
Libtardaplorable©. And proud of it.
Brilliantati©0 -
I dont think that it existsHughFreakingDillon said:i agree with juggler in that i don't know why anyone would lie to a pollster when it's anonymous, unless there's some unfounded conspiracy theory/paranoia out there on the right that it's not really anonymous?
Last time around, there was a weak turnout of democrat voters. Too much of the Democrat/Anti-Trump vote went to Stein or Johnson instead of Clinton. I also think there were undecideds (who dont exist now in significant numbers) who threw a 'protest' vote to Trump, not believing he would actually win.0 -
Undecided voters typically break to the challenger at a % similar to the approval/disapproval rate. Clinton was functionally the incumbent in 16.MayDay10 said:
I dont think that it existsHughFreakingDillon said:i agree with juggler in that i don't know why anyone would lie to a pollster when it's anonymous, unless there's some unfounded conspiracy theory/paranoia out there on the right that it's not really anonymous?
Last time around, there was a weak turnout of democrat voters. Too much of the Democrat/Anti-Trump vote went to Stein or Johnson instead of Clinton. I also think there were undecideds (who dont exist now in significant numbers) who threw a 'protest' vote to Trump, not believing he would actually win.0 -
This is the primary concern we should all have. Definitely agree with that.static111 said:
No I knew trump was gonna win because of the closet white supremacists in 2016 and because of Hillary’s poor campaign. I don’t think this is a repeat of 2016. I think it is a different animal altogether that we are viewing through the 2016 lens and saying wow look how much better we are doing compared to 2016, rather than focusing on what unique unknowns exist in 2020.The Juggler said:
You were not worried four years ago??static111 said:
This is the only election I have ever felt this level of insecurity around since I turned 18 and could start participating. Hopefully I am wrong and this election can be measured using the same metrics as previous elections and there is nothing to worry about.The Juggler said:
ohmygodstatic111 said:
My real question is how do you poll the people that are trying desperately to hold on to the white privilege system and will say all day that they aren’t voting for trump,especially to pollsters, but still check his name in the privacy of the voting booth. I think Trump has a much better chance than anyone thinks based on passive white supremacy, but since it can’t be easily quantifiable and put down in numbers it’s something that could prove to be the ultimate November surprise.Lerxst1992 said:mrussel1 said:
Great... there's better than a 23% chance. So what is it, 24%? 88%? The beauty of your argument i suppose is that you can't be wrong even if you're not using math, only old articles. If Trump wins, you say...I was right! Same thing if he loses. Neither would be empirically true.Lerxst1992 said:mrussel1 said:
Dude.. sorry, you don't have the background or analysis in this topic to declare that Silver's 75 or 77% is way off. Your assumption also discounts the massive move of suburbanites toward the D. That was clear in the analysis post 2018 and the numbers are bearing out the same this time. Trump won the suburbs in 16 and he is losing them today. They are not exactly chock full of NCWs. Women have also jettisoned the R party dramatically since 16 as well. So the statement that NCW's moving from 55 to 60% turnout wins the election for Trump ignores every other demographic move.Lerxst1992 said:mrussel1 said:
This is exactly correct. Somehow people (particularly Trump supporters) think something less than 100% is equal to 101%.dignin said:
The issue I have is you calling 538's math foolish. You seem to want to say that 538 should be calling this more 50/50, when the math actually calls for around 75/25. What you don't seem to understand is that in 4 simulated coin flips Biden would win 3 times while Trump wins once. Nobody here is excited about those odds, of course we want Biden winning 4 of those coin flips but the math doesn't show Trump winning 2 of those coin flips either. Biden is still the odds on favourite.Lerxst1992 said:Lerxst1992 said:Cook political report has an interesting tool where you can adjust turnout by demographic and see how that impacts electoral college.
It’s starts with trumps best demo, non college whites at 55% (Biden win w 307 electoral votes). Increasing that demo to 60% gets trump a win with about 280 EVs. 59% turnout, reduces it to 230 EVsI cant imagine a 1% turnout difference in trumps base nets a 50 EV difference, but it underscores how difficult a job pollsters have to get the polls accurate and account for even a tiny increase in trump base turnout. It ain’t helping my confidence with pollsters and if it’s this much of a difference 538 is really off their rockers w .77
HOW IT WORKS: Start with the results of the previous election, adjusted fordemographic change since 2016. Then, adjust the sliders below to see how shifts inturnout and support among five demographic groups could swing the Electoral CollegeSorry but this is damn interesting, as is the Supreme Court vs House analysis to decide contested elections. Clearly I’m in the wrong place.
This election isn't another 2016, way too many variables have changed.Not foolish, just underestimating the trump effect on elections.
Most significantly, elections are not about flipping a coin. According to Cook, Trumps 2016 performance at getting NCWs is about 4% from a huge statistical advantage, where one percent of this demo can net trump 50 EVs.He was at .55 NCW 2016 turnout. At .59 Biden still wins with @ 280 EVs. Thats great, we can all exhale. At .60 turnout, it goes to 230 and trump wins. THATS the tipping point here, this potential wave trump has access too. Being by far The largest demo (by far can not be emphasized enough), it is foolish to discount this phenomenon. That’s not even close to flipping a coin. It’s extraordinarily biased and a weighted coin.I’m hoping trump never gets there, but I am not shutting my eyes about it either. He proved in 2016 he knows how to get new voters from this demo to vote. No reason he can not find new voters there again. That’s why 77% is way off.Dude, the mistake Silver is making is this is not statistics, it’s political science. Trump has tapped into something pollsters did not identify last time. There is a better than 23% chance he can do it again.
lets fire up that time machine...again,
....PoliticsClinton Dominates in Key Philadelphia Suburbs, Bloomberg Poll Finds
Losing Pennsylvania’s 20 Electoral College votes would sharply curtail Trump’s paths to the White House.October 13, 2016, 5:00 AM EDTThe funny thing is I’d estimate Biden at .63, maybe .14 lower than 538, based on tipping point states. I think Biden is the favorite, but there is a stronger chance pollsters are making the same mistake again than .23; my main concern is trump is looking much stronger lately in these states, NC FL AZ PA. With these states within or near the moe, if having trouble believing .77 is accurate. If there is an “unpolled white vote” as there was in 2016 trump could be tied or ahead in those states.
the 2 main tipping point states, 538 has Biden at .77 in PA and .64 in AZ. The two closest tipping point states have a .13 spread. I find that debatable.
Its interesting that gauging demographic accuracy in polling is part of 538s process, but if we look at PA, they actually give biden better rating on that category (Biden 53/ trump 46). That just seems wrong, but I can not prove it, except to point out this is where pollsters got trump wrong in the past.
“ Average with demographics-based vote share projection”
Groundhog Day. All of these threads. lol
There is no evidence of this happening. There have been hardly any difference in the live phone polling and online polling this year. Both are anonymous, by the way. But for the people who seem convinced that Trump supporters are lying to pollsters over the phone for fear of being shamed or something (I don't know), how do you explain that Biden's doing just as well with the even more anonymous online polling?
This theory makes zero sense.
Folks---it's okay to be scared. But my god, it's not all doom and gloom. Biden's in very good shape with just over a month to go. That point cannot be disputed.
i mean I’ve already seen and heard coworkers talk about how the $750 tax is no big deal because rich people make their money in a different way and it gets taxed differently.There are a lot of things going on with framing the election in a way that Don can challenge the results in a close election. So the polling may well be right on and Don still “wins” I’m hoping for a high turnout to offset this.www.myspace.com0
Categories
- All Categories
- 148.9K Pearl Jam's Music and Activism
- 110.1K The Porch
- 274 Vitalogy
- 35K Given To Fly (live)
- 3.5K Words and Music...Communication
- 39.2K Flea Market
- 39.2K Lost Dogs
- 58.7K Not Pearl Jam's Music
- 10.6K Musicians and Gearheads
- 29.1K Other Music
- 17.8K Poetry, Prose, Music & Art
- 1.1K The Art Wall
- 56.8K Non-Pearl Jam Discussion
- 22.2K A Moving Train
- 31.7K All Encompassing Trip
- 2.9K Technical Stuff and Help