A few thoughts on where the presidential race stands and how to think about it in the short- and medium-term future. Maybe I should call this whole thing “Note to Myself to Keep from Going Crazy,” but if you find any of it helpful, it’s a note to you too.

Thought 1: Spend less time looking at poll numbers. They don’t tell the future. The history is full of candidates who looked like losers in May and ended up winners, and vice versa.

The latest polls are estimates only of the evanescent present, and they are inherently fuzzy estimates of that. If you take the margin for error seriously (which, of course, we all should, but seldom do), they generally show a statistical Trump-Clinton tie. Even when they don’t, at their best they measure a passing moment that has passed by the time the data reaches you.

Not that long ago, the polls suggested Trump was unelectable. Some Democrats even hoped he would be nominated because he couldn’t possibly win. Be careful what you hope for.  Now it’s a margin-of-error race, according to most polls.

But a bigger error, outside the margin, is to forget that even at their best, they are not predictive.

Thought 1A: If you must obsess on the polls anyway (which I recommend against), here is the big factor that renders the current Trump-Clinton numbers particularly evanescent:

Donald Trump, against all expectations (which underscores the main theme about thinking you can know the future), has consolidated the support of most Republicans. This happened incredibly quickly after the previous understanding — that many Republicans could not and would not unify around Trump — was the common wisdom (and I use the term “wisdom” facetiously).

On the other hand, among Democrats who currently support Bernie Sanders, a growing portion are telling pollsters that they will not support Hillary Clinton if she is the nominee. According to Nate Cohn, the New York Times political number cruncher, in April 71-82 percent of Sanders supporters said they would support Clinton if she is the Dem nominee. In May, that range has dropped to 55-72 percent.

That’s a reasonably big drop and is obviously being produced, to a significant extent, by anger among Sanders supporters over the feeling that they and Bernie are not being treated fairly as they press to keep alive Sanders’ narrow path to the nomination. This is understandable and normal.

History suggests that many or most of them will eventually, perhaps with little enthusiasm, vote for the Democratic nominee. (Some will vote the Green Party ticket. Some will not vote. Very few will vote for Trump.) But most will, and Sanders will encourage them to vote strategically, which will mean an anti-Trump vote for Clinton. Until that shakes out, the current polling tells us even less than current polling usually does about what will happen in November. This was one of the main points of the Cohn piece I cited above.

Thought 2 is captured by two Churchill quotes: To fools like me, these two somewhat contradictory quotes have been taking up space in my brain for decades, so long, in fact, that I’ve gradually come to learn that one of them Winston Churchill never said, and the other he did say, but attributed it to a source that he didn’t specify. In other words, neither one of them is really a Churchill quote. They capture the blessing and curse of democracy.

The first one of them (the one no one can authenticate as having ever been said by Churchill) goes like this:

“The strongest argument against democracy is a five-minute conversation with the average voter.”

Trump is a liar, an egotist, a bully, a sexist and a racist. Or maybe he’s just pretending to be. He has never communicated a set of policy ideas that makes sense or that survives follow-up questions, which he generally deflects with bluster.

When one of his previous utterances becomes inconvenient, he either changes the policy, refuses to accept responsibility for it, or changes the subject. It’s hard to tell whether this is the real him, or just a role he developed for present purposes.

He now has the support of many normal Republicans who once swore they would never be for him and millions more who have always been for him but cannot, in the course of a five-minute discussion, make a coherent case for voting for him based on actual policies or actual things he has done or said. Perhaps the best explanation of this is captured in a smart piece by Walter Russell Mead. The core of Trump’s support are Americans who feel that they have seen who Democrats favor and seen who establishment Republicans favor, and none of it is good for them. They want something else that will be good for them and, for the moment at least, Trump has convinced them that he gets it and will do that.

Neither Trump nor any of his growing body of more normal-seeming surrogates can make clear how that will work,  but the rules of our discourse somehow get in the way of making this as plain as it needs to be made.

My life’s training as a journalist tells me that the most helpful way to handle a situation like this is to keep asking substantive questions and pointing out the problems with his previous earlier answers. But this doesn’t work in this case. Plus that method depends on the now-shredded assumption that a candidate in this condition will either develop some answers or be forced to slink off the stage and not come back until he has some answers. That assumption, apparently, is soooo 20th century.

The second quote, which Churchill actually did say, on the floor of Parliament, but of which he didn’t identify the source, is this:

“Indeed it has been said that democracy is the worst form of government except for all those other forms that have been tried from time to time.”

Translation: Notwithstanding the deranged obloquy and vituperation above, if Trump does win 270 electoral votes, he gets to be president and gets control of the nuclear launch codes and we will just to have to hope, notwithstanding everything above, that he has just been pretending to be a reckless jerk.