This is the third installment of a multi-part series. You can read the second part here.
We cannot give the manosphere too much credit. Although we were significant during the primaries for influencing Conservatives nationwide, especially in the Northeast, once we got to the general we found ourselves battling the well-established and entrenched TV media. Older voters, women, and liberals who did not read within our circles and whom we had no influence over, were now part of the pool of voters we needed to compete for and could not rely on the internet to sway. It was a much harder game.
Thus, during the general, it was mainly up to Donald Trump and his team to beat the machine, which was one of his big selling points I had for him during the primaries. We understood that his independent wealth, mastery of the media, and tight game was only way to shine the sunlight down into the cave and get voters to deprogram themselves away from old media and into new media, the internet.
My gut instinct Trump had what it took to be successful in the general was proved as soon as the LA Times/Dornsife poll came out, and we saw him beating Clinton by 7 points after the RNC. Most were scared by the other push-polls showing Trump behind Clinton, but those polls had weak methods. Anyone who took a high school stats class could have told you why. The LA Times/Dornsife university poll started with a statistically representative sample of America’s demographics, and then polled that sample. The LA Times only replaced the people polled if they dropped out of the survey due to personal reasons.
Conversely, just about every other poll took the opposite approach – calling up random people, polling them, and then applying “filters” and “weights” to make the sample demographically representative. In other words, the LA Times poll worked forwards, while the others worked backwards. Doesn’t take a genius to figure out why working backwards in statistics is a horrible idea. The potential for bias that can be introduced with such backwards methods is enormous.
The other polls that called the election correctly are People’s Pundit Daily and the Investor’s Business Daily. The PPP poll used a mixture of LA Times methods with a bit more random sampling, while the IBD used the traditional method of pure random sampling. I am not sure why the IBD poll was accurate while all the other random-sampling polls were wrong. The IBD does not share its methods so I cannot comment.
Also, the LA Times poll said it would predict the popular vote, and not the electoral vote, but it turned out to be predicting the electoral vote more than the popular vote. The reason for this, I surmise, is that they most likely picked their initial sample demographically across the states in a way that mirrors the electoral college distribution of the USA, since the races are mainly located in key states, therefore they had Trump +3 at the end. Had they gone with a heavier popular vote model they would have ended up like the PPP poll of +.8. That said, they should stick with the electoral college model because that is how presidents are actually selected.
A lot of guys kept getting fooled by the shitty polls, leading them to bad conclusions, and all I have to say is: be more careful reading how stats work.
Eventually I made the decision to get involved with the ground campaign in NH. I understood that, according to the best polls, we had earned the loyalty of about 40% of the USA’s voters, and Trump himself could get around an extra 8-9% with his media savvy, but to get the last yard to the touchdown, to move the needle that extra 1-2% and cross the 50% threshold would require good old fashioned brute-force door to door sales skills. All the game I had learned here over the years was put to use in NH. And in NH, believe it or not, is where we won the election, which will be the subject of the next installment.