Election 2020

[pbrower2a]

My estimates for the results for intermediate probabilities based on near-linear differentials  for three and two months out.

Time to election  |1 point|5 points||10 points|20 points|
one day............. |...64%|....95%|.....99.7%|.99.999%|
one week........... |...60%|....89%|.......98%|...99.97%|
one month......... |...57%|....81%|.......95%|.....99.7%|
three months......|...55%|....72%|........87%|......98%|
six months..........|...53%|....66%|.......79%|.......93%|
one year.............|....52%|...59%|.......67%|.......81%|

I can only estimate the value of margins between 1 and 5, 5 and 10, and 10 and 20 with estimates of differentials. and then using other differentials for interpolation between 1 and 3 months.  


lead  likelihood, three months and two months

0   50    50  10  87  91
1   55    56  11  88  92
2   59    64  12  89  93
3   64    69  13  90  94
4   69    73  14  91  95
5   72    77  15  92  96
6   76    78  16  93  97
7   80    80  17  94   97
8   83    84  18  95  97
9   85    87  19  96  97

Note that the probability that a leader wins more precipitously at 5% (9% delta) than at 1 (2% delta), so the increase from my estimate at three months cannot be linear in the range between a 1% and a 5% lead.

Now my estimates based on very recent polling (after the Party conventions), mostly in States decided by 10% or less in 2016 (today those are the only states). Numbers are percentage likelihood of a Biden win:



NC (Monmouth) Biden +2 (64%)
FL  (Quinnipiac) Biden +3 (69%)
PA  (Quinnipiac) Biden +8 (84%)
WI  (FoX News) Biden +8  (84%)
AZ  (FoX News) Biden +9  (87%)
NV (University of Nevada) Biden +5 (77%)
TX (Morning Consult) Trump +1 (44%)

CO (Morning Consult) Biden +10 (91%) -- new September 1
MI (Hodas) Biden +11 (92%)
MN (PPP) Biden +8 (84%)
TX (it is a D poll) Biden +3. Average with Morning Consult and you get +1 (56%)

Biden 0-4% saturation 8
Biden 5-20% saturation 6
Biden 21-35% saturation 5
Biden 36-40% saturation 4
Biden 41-49% saturation 2

50/50 white

Biden 51-60% saturation 2
Biden 61-70% saturation 4
Biden 71-80% saturation 5
Biden 81-95% saturation 6
Biden 95%+ saturation 8

With Texas shifted as barely as I can into positive territory for Biden, I can easily imagine a 400-EV win for Biden. COVID-19 has begun to hit Texas hard, and Trump has mangled the response to COVID-19 badly. This may explain the horrid poll for Trump in Arizona that seems anomalous. FoX News does and gets good polling even if its broadcasts are mostly right-wing propaganda. 

....This is different from approval and disapproval. Disapproval above a certain level kills a politician's chances of winning a place (in this case a State)


Here are some approval/disapproval polls. This pollster may not be completely accurate; no polling is demonstrably perfect. 

Civiqs - thru 9/2

National: 41/56 (-15)

Alaska: 43/55 (-12)
Arizona: 43/53 (-10)
Colorado: 38/59 (-22)
Florida: 45/53 (-8)
Georgia: 44/53 (-9)
Iowa: 46/52 (-6)
Kansas: 52/45 (+7)
Maine: 37/60 (-23)
Michigan: 41/56 (-15)
Minnesota: 39/58 (-19)
Montana: 47/50 (-3)
Nevada: 36/60 (-24)
New Hampshire: 37/58 (-21)
New Mexico: 43/54 (-11)
North Carolina: 45/53 (-8)
Pennsylvania: 42/55 (-13)
South Carolina: 50/48 (+2)
Texas: 48/49 (-1)
Virginia: 38/59 (-21)
Wisconsin: 43/54 (-11)

https://civiqs.com/results/approve_presi...oomIn=true

It is difficult, if not impossible, to win a state in which one has 50% disapproval. It is nearly impossible to win a state in which one has 52% disapproval. Beyond 52%? Basically, forget it unless something weird is going on like a three-way election  or an opponent mired in a major scandal. In that zone the results might be close, but you know the saying: close matters only with horseshoes, hand grenades, and atomic explosions. 

I find the Alaska number astonishing... Most of the rest isn't. That may say something about Alaska (how do you say "Press '4' for Inuit" in Inuit?) polling.