Both multiples divide price by earnings, and they routinely disagree by a third. The difference is not precision — it is which period the denominator describes, and why that matters most exactly when you care most.
A price-to-earnings ratio is price divided by earnings per share. That much is not in dispute. What is in dispute is which earnings — the ones a company has already reported, or the ones analysts expect it to report next.
The trailing multiple answers a question about the past: given what this business earned over the last twelve months, how much is the market charging for it? The forward multiple answers a question about the future: given what it is expected to earn over the next twelve months, how much is the market charging?
These are not two estimates of the same quantity, one more accurate than the other. They are answers to different questions, and treating them as interchangeable is where most of the damage gets done.
Markets are not paying for earnings that have already been banked. Those are gone; they are in the share price already. What gets bid on is the claim to future cash flow, which is why forward multiples generally track prices more closely than trailing ones.
This is also why the two diverge most violently at turning points. When earnings are about to fall off a cliff, the trailing denominator is still describing the good times, so the trailing P/E looks reassuringly low right up until it does not. When earnings are about to recover, the trailing denominator is still describing the wreckage, so the trailing P/E looks absurdly expensive at precisely the moment the asset is cheapest.
The trailing multiple is not wrong. It is late. And it is late by an amount that peaks exactly when the decision matters.
There is a wrinkle in "the next twelve months". Companies report in fiscal years, and the next twelve months from today almost never line up with one of them. A company nine months from its fiscal year end has three quarters of that year ahead of it and nine months of the following year.
Using the next fiscal year alone creates a discontinuity: the moment a company crosses its year end, the reference period jumps forward twelve months and the multiple steps with it. Across an index of hundreds of companies with staggered fiscal calendars, those steps are constant background noise.
A blended forward twelve-month multiple fixes this by weighting the two fiscal years in proportion to how much of the coming year falls in each. It is a small piece of arithmetic that removes an artefact large enough to be mistaken for a signal — in our own testing, using the next fiscal year alone shifted index multiples by up to a fifth.
Forward earnings are a consensus of analyst forecasts. They are not a measurement. They are systematically optimistic at the start of a year and get revised down as reality arrives, and they miss inflection points more or less by construction, since a forecast is an extrapolation of what is currently understood.
So the honest description of a forward P/E is not "the true valuation". It is "what the market is paying, divided by what the market currently believes". Both halves of that can move, and telling them apart is the actual analytical work.
That is why a multiple should never be read alone. A multiple rising while expected earnings fall is the market re-rating the same profits upward — expansion. A multiple rising while expected earnings also rise is growth being paid for. These look identical on a P/E chart and mean opposite things.
Turn the ratio upside down — earnings divided by price — and you get the earnings yield, expressed as a percentage. It carries exactly the same information, but it behaves far better at the extremes.
As expected earnings approach zero, a P/E ratio runs toward infinity and then flips sign; at the crossing it is not merely large, it is undefined. Any chart containing such a point is unreadable, and any percentile or average computed across one is destroyed. The earnings yield passes through zero without drama.
It also has the advantage of being denominated in something comparable. A 5% earnings yield sits naturally next to a bond yield in a way that "20 times" never will.
Education
What bank reserves are, why there is no fixed dollar threshold for "enough", which indicators actually tell you scarcity is arriving, and why DLI keeps the reserve level out of the score while showing it on the page.
Education
A practical daily routine for crypto investors who want to stay macro-aware without becoming macro-obsessed. Seven indicators that capture the full liquidity picture, when each one matters most, and how to read them in under 10 minutes a day.
Education
Most retail investors hold a few cached beliefs about Fed policy that don't survive contact with the data. Here are four of the biggest, with what the historical record actually shows for each.