I’ve been hearing for months that the Republicans would win this presidential election if only they hadn’t nominated Donald Trump. I’m not completely against this line of reasoning. I had hoped from last summer that that GOP base really had no ideological bearings and so would nominate him because I thought he would be a weak candidate. But I think this idea that “Hillary would be losing to any other candidate” is just a myth. I wrote about this back in May, John Kasich Would Not Beat Hillary Clinton. Remember: Kasich is about the most “reasonable” candidate the Republicans have.

One thing that bugs me about the whole “Hillary would be losing” game is that it is just that: a game. Although I think it is based on bad reporting over decades, Hillary Clinton has very low approval numbers. So what does it even say that if the Republicans had nominated someone more popular than she that they would win this election? It’s almost a tautology: Hillary would be losing if Hillary were losing. I think what we are really supposed to take away from this is that the country is really in a Republican mood. It’s a way for conservatives and (even more) moderates to whistle past the left turn the country is taking.

The Trump Tax

What most brought this all back up for me was Vox‘s Trump Tax. This is a very silly exercise where they look at six political science models, take a weighted average, and then decide where the race should be right now. Based upon this, they think it should be Generic Republican +1.8 percentage points. Also note: the range of the models is 8 percentage points (Generic Democrat +4.0 to Generic Republican +12.0).

Now, being as I’m a “fundamentals guy,” you might think I would like this approach. I do, after all, have my very own election model. Now my model is very simple. But there’s a reason for that. I’ve always thought that election models were mostly nonsense. The only thing you can really depend upon is the economic trend. The rest, I believe, is noise. It’s based upon the fact that these models are really nothing but correlations. And they are based on a very small sample size — in general, no one looks at elections before Eisenhower.

Professional “Hillary Would Be Losing”

If you really want to play the “Hillary would be losing” game, you can take it to the professional level. You can say, “If Trump weren’t so bad a candidate, the press would have spent more time picking over the Republican candidate instead of just shrugging and saying, ‘You know Trump!'” Or you can go back further and say, “If Democrats didn’t see the mess that was the Republican primary, they might have nominated someone other than Hillary.” You can play the “Hillary would be losing” game to the end of time. It means nothing.

Another thing that bugs me about all of this is exactly what a Generic Republican is. The models are based on elections when Republicans had better candidates. If it wasn’t Trump, it was going to be Ted Cruz — a man so vile even Republicans hate him. Or what about the much more likable Marco Rubio? I always thought that Hillary would have eaten him for lunch. So I’m not sure what Generic Republican tells us anymore. What my model tells us is that this election should give the Democrats a slight advantage.

So if conservatives and moderates want to make themselves feel better by playing the “Hillary would be losing” game, fine. But all it really says is that in a different universe, things would be different. And I think we can all agree on that.