Markets at all-time highs with limited forward returns due to high starting valuations

As I write on Thursday, the S&P 500 is building on yesterday’s new all-time high. If you held through the Iran-related correction—and perhaps even took advantage of lower prices by adding to high-conviction holdings—congratulations.

While the S&P 500 has returned to its January high and pushed above it, the article highlights several developments since January that investors may be underpricing: an ongoing war with Iran that has already triggered a 10%+ market drawdown; failed peace talks with restarted negotiations; oil trading well above early-year levels, with Goldman Sachs warning that a severe supply shock could drag the S&P 500 as low as 5,400; and inflation pressures, including headline CPI reaching a two-year high of 3.3% in March, driven by a record 21% monthly spike in gasoline. The piece also notes that traders have pushed rate-cut expectations to mid-2027, a major shift from what was expected in January.

Valuation risk: what investors may be paying

The article argues that the more important question may not be whether the market’s new all-time high prices in these risks, but what kind of returns investors can realistically expect from current, elevated prices.

It points to valuation as a key long-term risk, focusing on the cyclically adjusted price-to-earnings (CAPE) ratio, also known as the Shiller P/E ratio. The CAPE compares stock prices to 10 years of inflation-adjusted earnings, developed by Nobel laureate Robert J. Shiller.

The core message presented is straightforward: high starting prices tend to be associated with lower future returns, while low starting prices tend to be associated with higher future returns.

CAPE near 41 and what history suggests

According to the article, the CAPE ratio is at almost 41, described as the second-highest reading in more than 140 years, surpassed only by the peak of the dot-com bubble.

It cites a chart (not reproduced in the text) that plots starting CAPE ratios since 1871 against annualized real returns over the subsequent 10 years. The article says that when starting CAPE ratios are above 35, forward 10-year returns have mostly clustered between 0% and 5% annually, with several periods producing negative real returns.

The piece also references a 2024 academic working paper titled “Estimating Long-Term Expected Returns” by researchers at La Trobe, Massey, and Auckland University of Technology. Using an improved version of the CAPE ratio, the authors find that starting valuations explain a meaningful portion of 10-year return outcomes, reinforcing the historical pattern described.

“This time is different” debate

The article addresses common counterarguments to valuation warnings, including claims that interest rates are structurally lower than in past periods (supporting higher valuations), that technology companies have higher margins (also supporting higher valuations), and that old averages may not apply.

It attributes a review of these arguments to Research Affiliates, which the article says concluded that some points have merit and that equilibrium valuations may be modestly higher than the 20th-century average. However, it states that none of the arguments explains a CAPE of 40.

The article’s conclusion is that even if elevated valuations do not revert to the mean, the implication is still low future returns: paying a high price for future cash flows reduces the math of what investors earn going forward.

Not a timing signal, but a long-term return constraint

The article clarifies that CAPE is not presented as a timing tool. It notes that CAPE signaled expensive conditions in 2018, and investors who acted on that signal missed a significant bull run. Instead, it frames CAPE as a long-game lens rather than a “sell” trigger.

It then describes an “uncomfortable dilemma”: the market’s momentum is bullish and stepping aside has been costly in recent years, but starting valuations suggest underwhelming 10-year returns at best.

Alternative approach: focusing on specific stock signals

To address the dilemma, the article introduces a system promoted by its corporate affiliate, TradeSmith, which it says scans thousands of stocks each day for specific setups that have historically led to high-probability trades.

The piece says the approach is not based on forecasts or narratives, but on patterns in historical data that TradeSmith engineers call “signals.” It describes the system as detecting “tiny anomalies” in stocks’ historical data and identifying statistical connections that a human analyst would not find.

Examples cited in the article

The article includes examples attributed to TradeSmith CEO Keith Kaplan:

• Palantir Technologies Inc. (PLTR): The system signaled a 5.8% move in nine days, with a stated historical accuracy rate of 95%, followed by a 15.1% gain in seven days.
• Lam Research Corp. (LRCX): The article says two conditions had to align: the stock had to close above its 200-day moving average, and that close had to fall two trading days before a market holiday. It states that when the signal fired on August 28, 2025 (two trading days before Labor Day), LRCX gained 11.4% in 15 days.

The article also states that the system runs each stock through 847 individual calculations daily, compiling more than 2 million trade evaluations every 24 hours, and that it is designed to work regardless of whether markets are in bull or bear phases.

Event and access mentioned

The article says Keith Kaplan will walk through the system in more detail during an “AI Signals Trading Event” next Wednesday, April 22, at 10 a.m. Eastern. It also states that a working version of the platform will be available ahead of the event for users to test in real time, including searching stocks, viewing active signals, and exploring how the system works.

Bottom line

The article concludes that the market is at all-time highs even as risks accumulate and valuations remain stretched. It says that this does not necessarily mean investors should step aside, but it argues they should be thoughtful about how they participate.

Have a good evening,

Jeff Remsburg