I wrote about the Shiller CAPE ratio recently, when it was flashing a warning most investors were ignoring. It's still flashing. And as The Motley Fool noted this week, the ratio is now closing in on its dot-com era record, the highest valuation in the history of the measure. So let's talk about what that actually means, because the honest answer is more complicated than the headline suggests.

First, a quick refresher for those who missed the earlier piece. The CAPE ratio, built by economist Robert Shiller, compares the S&P 500's price to its average earnings over a stretch of years, smoothed to strip out the noise. Why smooth at all? Because a single year of earnings can be wildly misleading. A recession year, when corporate profits crater, can make the market look absurdly expensive on a one-year basis even when the long-term picture hasn't changed. The price you pay for a decade of earnings, that's the question CAPE asks.

The answer right now: expensive. Historically expensive, near a level that, before the late 1990s, the market had never seen. I lived through that era as an investor with real money on the line, and I remember doing the mental gymnastics to convince myself the rules had changed. They had not.

Now, before anyone heads for the exits, here is the part that matters. Expensive valuations are a poor market-timing tool. In 1996, Alan Greenspan warned about irrational exuberance years before the dot-com peak arrived. Anyone who sold then missed the enormous gains that followed through the end of the decade. High valuations tell you the market likely offers a lower expected return over the next decade or two. They do not tell you the market will fall next quarter. Those are different statements, and confusing them has cost investors a lot of money over the years.

But. And this is the part I'll keep repeating. A high starting valuation has historically been the most reliable signal we have that long-run returns from here will likely be lower than what a cheaper market would deliver. The logic is simple: the more you pay per dollar of earnings, the less those earnings have left to pay you back. It isn't a crash prediction. It's an expectation reset.

So what does a 62-year-old who has watched this movie a few times actually do? Nothing dramatic. You keep buying, because time in the market still beats timing it, and I have never once met the person who sold at the top and got back in at the bottom. You keep your allocations honest, which means if you've been riding the gains and your stock weighting has crept from 70 percent to 85 percent, you trim back to plan. And if you're within a few years of needing the money, this is exactly the moment to make sure that money isn't fully exposed to whatever comes next.

When did you last check what you're paying for the market, not just what it has paid you?

Usual disclaimer: I'm sharing perspective built on forty years of my own mistakes, not professional financial advice.