Plurality Threshold Analyses 10/26/24
A FIASCO OF FLAWED POLLS
Today’s projected Electoral College Vote is: Harris-Walz 262… Trump-Vance 218… and Toss-Up 58.
That’s a major shift from last week. In both Pennsylvania and Michigan, Harris-Walz are 3% above the Plurality Threshold.
In the other big move, Trump-Vance dropped below the Plurality Threshold in Arizona and Georgia. They led Harris-Walz in AZ (+1%) and tied them in Georgia (+/- 0%).
Nationally, the RealClear Politics average for the six-way race has Trump and Harris tied at 47.5%, Kennedy at 1.3%, Stein at 0.9%, West at 0.6%, and Oliver at 0.7%. So the national Plurality Threshold equals 48.25%.
BUT, BUT, ONLY THREE OF THE ELEVEN POLLS IN THE RCP AVERAGE INCLUDED ROBERT F. KENNEDY, JR. IN THEIR HORSE RACE QUESTION.
Yet RFK, Jr. is on the ballot in 29 states, according to NBC News. Stein is on 31 states. And West is on the ballot in 16 states.
In the twenty mega-Democratic states, RFK, Jr. is on 11 ballots: CA, IL, MI, NJ, WA, MN, WI, MD, CO, OR, and CT.
Please note that he’s on only two battleground states: MI and WI. That lowers the Plurality Threshold in those Democratic-leaning states, a clear advantage to Trump-Vance. Stein is on all battlegrounds but Nevada.
In the remaining 4 mini-Democratic states and 27 MAGA states, Kennedy’s name appears on 18 ballots: AL, AK, AR, DE, DC, ID, IN, KY, LA, MT, NM, OK, SD, TN, VT, and WV.
Since RFK, Jr. suspended his campaign and endorsed Trump, most pollsters, erroneously, have left him off their statewide surveys. So, if this was a poker game, RFK, Jr. is a wild card — maybe a trey (3) of clubs or a sore spot (4) of hearts — but not much more.
But this we do know. The Quislings — Kennedy, Stein and West — will shrink the net margin between Harris and Trump in every state where their names are on the ballot. And, cumulatively, that means a narrower net margin in the popular vote nationwide.
A FIASCO OF FLAWED POLLS
Five good reasons exist to doubt the polls this year. First, MAGA pollsters are inflating Trump’s percentage of the vote — perhaps, by that “sore spot” of four points — in order to drive the averages down to almost dead even.
Second, some pollsters are using “weighting by recall vote” to over correct for the “missing” Trump vote in 2020 surveys.
Third, too many pollsters are simply not tracking who is and is not on the ballot. See discussion above of RFK. Jr.
Fourth, the margin of error (MoE) explodes for small subsets off a poll’s sample. For example, a high quality poll with 1,600 respondents has an MoE of 2.5%. But when a subset — often called a cross tab — has only 105 respondents, the MoE grow to ± 9.6%.
Drawing any conclusion from a subset of, say, 105 young Black men, is insane. Far better to devote an entire poll to their support for Kamala Harris than rely on an MoE that could be wrong by twice that 9.6%, or 19.2%!
Finally, if a poll adds up to 100% or 98% with only two candidates, then the pollster is “pushing” respondents to make a binary choice. It ain’t that simple in presidential politics. Never. Ever.
So, please, ignore the polling averages, partisan pollsters, and the political hacks touting their “results.” Instead, poll yourself. Ask just one question:
What’s my gut instinct on who is best for my family?
Then trust your gut and..
Vote! Vote!! Vote!!!