Its interesting to see how various people think about "true" value and "intrinsic" value.

My own thoughts are that "true" value or "intrinsic" value IV are not the most interesting way to look at an investment. What I would replace them with is IRR, Internal Rate of Return.
...
But here is why I like IRR:

Ticker       Company          2006 Price   2026 Price (Current)     Annualized IRR     Estimated Value of a $10,000 Investment
AMZN      Amazon.com Inc        ~$1.61           ~$254.98                 29%               ~$1,580,000
AAPL        Apple Inc           ~$2.55           ~$324.96                 27%               ~$1,270,000
GOOGL     Alphabet Inc          ~$9.61           ~$337.12                 20%                 ~$350,000
BRK.A  Berkshire Hathaway  ~$96,100.00       ~$758,500.00                 11%                  ~$79,000

I agree with you that the prospective actual realizable rate of return over your prospective holding period range is what matters. I refine that to the "pretty darned sure" rate of return, but it's the same idea.

The problem is, of course, knowing in advance what a good pick might be. The three in that table are picked, one assumes, based on their past results. But apparently you didn't pick them 20 years ago.

As an example, I looked up the expected returns for your three "A" listers, Alphabet, Apple, and Amazon, from a few years back. Value Line's forecasts are famous for being very close to industry consensus. Five years ago, they put out 3-5 year annual total return forecasts for 1635 firms. I don't have 10- or 20-year forecasts from 10- or 20-years ago, but the insight is the same. Five years ago:
Amazon was their 436th from top pick. It has returned 8%/year.
Alphabet was their 762nd from top pick. it has returned 18.9%/year.
Apple was their 1377th from top pick, at 2%/year. (yes, lower than 84% of the stocks in the list). It has returned 16.7%/year.
So, not only were those three not expected to do particularly well, but the one they expected to do the best did the worst among the three. Yet the analysis method was eminently sensible: what are the best metrics of value for this specific business, what is a likely rate of growth of that metric, and what multiple is the market likely to assign to that metric a few years in the future?

The sad truth is that merely looking at a list of really great long term compounders just doesn't work as a way to pick winners in advance, since it's backwards looking. In the late 1990s it's obvious you'd hold GE, right?

So, I think discussions about intrinsic value estimates are extremely worthwhile, for precisely the reason you mention: to *identify* the best long term IRR picks. IV estimates (good ones) are based foursquare on estimates of the future earnings and other future observable value of a business. So yes, it's only the rate of return that matters, but personally I think the best way to get that is to get in at a price that's low relative to your [future] IV estimate. Assuming that your IV estimate is prudently based on the best available "pretty darned sure" future for the security in question.

Very much like Manlobbi's IV10 approach: look ahead a long time, and hold the things with a good ratio of future likely value to price today. Sell and switch to better opportunities during overvaluations or not, to taste. I'm big on switching horses, and sometimes that's better and sometimes not. This has been a good stretch: my portfolio total return has beat the S&P index by 11.45% in the last year, while staying over 60% in cash. Other years I lose a bundle, marked to market anyway.

Jim