Subject: Re: OT: S&P versus T-Bills?
Based on monthly data, starting the beginning of 1950, the rolling 5 year average SP500 (GTR1 ^S5T) return is 1.77. The median is 1.74. The range is 0.67 to 3.58.

To see how CAPE effects the performance of buying during new market highs, you'll need to have a lot more samples. By only looking at data where the CAPE is above 35 you only have two significant sample periods (the first in 1998-2001) and the second in 2020-2021). Though you are testing back to 1950, the CAPE ratio being below your target nearly constantly is causing your sample size to be so small as to lose its utility.

https://www.multpl.com/shiller...

That isn't enough data to draw any conclusions by looking at CAPE values > 35 in the chart above. But you can solve as follows: If you take all the dates where the market reached a new high (for all CAPE values) and then (from this subset of all returns) chart the 5-year forward annual on the Y axis, with the CAPE ratio on the X axis, you will then encapsulate all the data (rather than just two short distinct sample periods, which should strictly should not be extrapolated).

I'd go back to 1871 using the data on Robert Shiller's website.

What I am almost sure you will see is that you will have a distribution of returns that increases as the CAPE ratio decreases, however the 5-year returns will still be on average a little higher than the CAGR of the broad market - in consonance with your thesis.

By accounting for all data, and looking at the continuity of returns over different CAPE starting conditions, you can then extrapolate with more confidence what to likely expect as return returns when starting from today. I continue to expect this to be worse than bonds for 5 years (if we started many times from a CAPE of 35, though there is so much noise that I'd consider each case to still be fairly random). CAPE isn't magic, but more or less just reporting what you are getting in earnings when buying today. If you buy a smaller bundle of earnings (higher CAPE), it is going to be worth less on average (and thus likely less 5 years away) than if you buy a larger bundle of earnings (lower CAPE) for the same price.

For all periods near market highs, the very worst time to invest for 5-year returns (for any given market high) will nevertheless still be when the CAPE ratio is the highest, such as now. Though that doesn't mean you should sell all and move into bonds today, because we still have the advantage from the momentum.

Proximity to recent market highs (and any time within 6 months is pretty good) is about the most reliable momentum strategy out there (if thinking in terms of index investing).

However as Jim has pointed out, and I want to underline - the CAPE ratio isn't useful as a timing device, as it only works over 5-10 year periods, but it is excellent to narrow the range of expectations for you returns to help with your planning, not to mention your sanity.

I'm fully invested right now, for example, despite the CAPE ratio (or the price/book ratio of the broad market, or the price/sales) rarely being this high.

The CAPE chart above is for the S&P500 which is market cap weighted. As it happens, the S&P5600 (small caps) are in a radically different situation. I'm amazed how little press this gets. The small caps, and even the medium caps, I consider valued about correctly right now - neither expensive, nor cheap. I expect the S&P600 (such as IJR) to beat the S&P500 (such as SPY) over the next 10 years - that's not a prediction, but a central expectation (subtly different). The earnings have to go somewhere, and when you are paying a considerably lower multiple, and the EPS is growing faster (small caps historically have grown EPS much faster than large caps, see link below) then the intrinsic value gains pile up over time and eventually show up in the quotation. I wrote a fair bit about the S&P600's relative valuation to the S&P500 here:
https://www.shrewdm.com/MB?pid...

- Manlobbi