Nate Silver of 538 is out with his prediction of how the mid term elections will go in November. According to Silver, there is a 74% chance that the Democrats will take control of the House. That's really high. It's not as high as the 96% chance that Hillary Clinton would beat Donald Trump which was what Silver predicted on election day in 2016. In other words, Silver thinks the Dems will win, but after 2016, he's hedging his bets a bit.
These polling analyses are fun to read for political junkies like me, but they are really nothing meaningful. There are two and a half months to go until the election, so there's a big chance that there will be an important event that will shape the elections which has yet to occur. Polls today won't tell us anything about such events.
Think for example what will happen if the special prosecutor puts out a report that says that there was no collusion between the Trump campaign and Russia. Then add in that the report points out that the Democrats and the Clinton campaign were the ones who colluded with Russians. That seems to be the truth. How would that affect the votes in November? Would anyone care? Would the anti-Trump crazies be so disappointed that many of them wouldn't vote? Would they all be encouraged to vote to show that Mueller's report didn't matter to them? Who knows the effect.
Imagine also that the USA makes a deal with China next week and most of the tariffs go away followed by a burst of positive economic activity. Would that lead more people to vote for the GOP to continue the great economy under the Republicans rather than the stagnation we had under Obama? Or how about if there is some sort of terror attack near election day? How will that affect the voting?
The point is not that this, that or some other thing will move the voters this way or that. No, the real point is that it is way too soon to make a mathematical prediction based upon polling as to which party will control the House. These predictions went too far in 2016 and the pollsters all got burned. You would think they would have learned something from that, but apparently, they did not.
These polling analyses are fun to read for political junkies like me, but they are really nothing meaningful. There are two and a half months to go until the election, so there's a big chance that there will be an important event that will shape the elections which has yet to occur. Polls today won't tell us anything about such events.
Think for example what will happen if the special prosecutor puts out a report that says that there was no collusion between the Trump campaign and Russia. Then add in that the report points out that the Democrats and the Clinton campaign were the ones who colluded with Russians. That seems to be the truth. How would that affect the votes in November? Would anyone care? Would the anti-Trump crazies be so disappointed that many of them wouldn't vote? Would they all be encouraged to vote to show that Mueller's report didn't matter to them? Who knows the effect.
Imagine also that the USA makes a deal with China next week and most of the tariffs go away followed by a burst of positive economic activity. Would that lead more people to vote for the GOP to continue the great economy under the Republicans rather than the stagnation we had under Obama? Or how about if there is some sort of terror attack near election day? How will that affect the voting?
The point is not that this, that or some other thing will move the voters this way or that. No, the real point is that it is way too soon to make a mathematical prediction based upon polling as to which party will control the House. These predictions went too far in 2016 and the pollsters all got burned. You would think they would have learned something from that, but apparently, they did not.
No comments:
Post a Comment