11-03-2020, 02:12 PM
11-03-2020, 02:49 PM
(11-03-2020, 09:32 AM)Eric the Green Wrote: [ -> ]The Sabato prediction does jive with JDG's, so it looks like Mr. Glick has looked beyond bias this time and agrees with brower. My salute as well.
???????????????
Although I appreciate being considered a good sport, my predictions are not in line with 538, Sabato, or Cook. My original prediction from MAR 2016 (most of it produced in JAN 2018) gives Trump 48.63% of the vote and about 293 EVs. The "48.00%" and "45.93%" scenarios are for informational purposes. I suspect that that Trump's share of the nationwide popular vote is more than 45.93% (538 gives him 45.4%) but less than 49% (I'm pretty sure Trump will get better than Carter's 41% and 97 EVs al la PBrower2a). It all depends on how accurate the polls are. Silver rightly points out that the polls in 2016 were off by less than 2%. I point out that 1% made a pretty big difference. It's a little like doing brain surgery and being off by just an inch. Oops. Have the pollsters learned their lesson, or made them worse? We shall have a better idea in about 10 or 11 hours. I'll be going over to friends to watch. There will literally be popcorn.
BTW, Silver puts the miscellaneous vote at only 1.2% instead of my predicted 1.6%. If accurate, that would knock Trump's chances of winning a state or district by a plurality somewhat. I don't have time to do the math.
----
This next isn't original analysis on my part, but my prediction for the Senate is 50 R's, 48 D's, 2 I's. My reasoning: If you check 538 from 2018:
https://projects.fivethirtyeight.com/2018-midterm-election-forecast/senate/?ex_cid=midterms-header
... and 2016:
https://projects.fivethirtyeight.com/2016-election-forecast/senate/?ex_cid=2016-forecast
and 2014 (you'll have to take my word for it), R's with over 50% on election day don't seem to lose, but there are usually 2 or 3 R's with less than 50% who do win. Has 538 improved their methodology? I'd bet only so far. Thus, 50Rs.
And BTW, what's with "this time"? My analysis is always based on facts. Would you like to provide any contrary examples? I have a long track record of unbiased analysis. I used to get paid it. My suspicion is that you just don't appreciate it.
11-03-2020, 04:48 PM
11-04-2020, 12:56 AM
11-04-2020, 01:42 AM
11-04-2020, 02:11 AM
11-05-2020, 01:24 AM
(11-03-2020, 02:49 PM)JDG 66 Wrote: [ -> ](11-03-2020, 09:32 AM)Eric the Green Wrote: [ -> ]The Sabato prediction does jive with JDG's, so it looks like Mr. Glick has looked beyond bias this time and agrees with brower. My salute as well.
???????????????
Although I appreciate being considered a good sport, my predictions are not in line with 538, Sabato, or Cook. My original prediction from MAR 2016 (most of it produced in JAN 2018) gives Trump 48.63% of the vote and about 293 EVs. The "48.00%" and "45.93%" scenarios are for informational purposes. I suspect that that Trump's share of the nationwide popular vote is more than 45.93% (538 gives him 45.4%) but less than 49% (I'm pretty sure Trump will get better than Carter's 41% and 97 EVs al la PBrower2a). It all depends on how accurate the polls are. Silver rightly points out that the polls in 2016 were off by less than 2%. I point out that 1% made a pretty big difference. It's a little like doing brain surgery and being off by just an inch. Oops. Have the pollsters learned their lesson, or made them worse? We shall have a better idea in about 10 or 11 hours. I'll be going over to friends to watch. There will literally be popcorn.
BTW, Silver puts the miscellaneous vote at only 1.2% instead of my predicted 1.6%. If accurate, that would knock Trump's chances of winning a state or district by a plurality somewhat. I don't have time to do the math.
----
This next isn't original analysis on my part, but my prediction for the Senate is 50 R's, 48 D's, 2 I's. My reasoning: If you check 538 from 2018:
https://projects.fivethirtyeight.com/2018-midterm-election-forecast/senate/?ex_cid=midterms-header
... and 2016:
https://projects.fivethirtyeight.com/2016-election-forecast/senate/?ex_cid=2016-forecast
and 2014 (you'll have to take my word for it), R's with over 50% on election day don't seem to lose, but there are usually 2 or 3 R's with less than 50% who do win. Has 538 improved their methodology? I'd bet only so far. Thus, 50Rs.
And BTW, what's with "this time"? My analysis is always based on facts. Would you like to provide any contrary examples? I have a long track record of unbiased analysis. I used to get paid it. My suspicion is that you just don't appreciate it.
I just don't see it much, considering your previous posts. Your premises are often biased and off the mark.
But I thought in a previous post you predicted Trump would get 217 electoral votes, which agreed with brower's. You didn't mention that in the above ramble. I was glad you didn't automatically predict a Trump win. Right now that's just about the electoral votes that he has, but he stands to gain at least 15 more from North Carolina and maybe more. But he won't get 293 unless he can get Barrett to give them to him. That would be one person deciding the election instead of 150 million Americans. Real democracy? NOT.
Trump and the Republicans are doing better than I and brower thought, no doubt. From my point of view, that is tragic. So far, Biden is winning the popular vote by a greater margin than Hillary, barely (over 3%), but California ballot counting is stuck because of lots of drop-off mail-in ballots that have to be verified. New York has been coming in slow too, as well as Nevada and PA. Having been a poll worker many times, I can testify how many people drop off their mail-in ballots at the polling place on election day. It has been much more this time. So, only 66% of the CA vote is in. It has been trending 65% Biden, so I expect more votes from the largest state will add to Biden's popular vote margin. It could end up as about 5%. That would mean Trump gets about 46-47%. Your prediction said more that 45.93%, so it will be just above that.
11-05-2020, 10:00 AM