U.S. Stock Market has reached its most expensive valuation in history after the Warren Buffett Indicator crossed 230% for the first time ever
...
The Space X IPO and other IPOs will push the indicator up even higher.
However, I have always wondered why a company going public which makes the indicator higher, or a company going private which makes the indicator lower suggests whether a market is more or less an expensive valuation.

I think the implicit assumption is that over time the set of all firms that are publicly listed account for a fraction of total US business activity that is within a relatively narrow range. Each take-private reduces it and each IPO increases it, but the range is probably pretty narrow if only because such a large fraction of the largest businesses are always public firms. SpaceX might be the first time in history that a kink in that graph is visible, but the overall total is still likely within a range.

The indicator is intended as an "approximately good enough" thing, not a precise metric. If it's within 5% of an all time high, that's all you really need to know: prices are high, so forward returns will necessarily be lower on average as soon as the multiple expansion stops.

Here's another graph that doesn't need labels on the axis...approximately right is again good enough to get the general idea across, even if (say) the mix of businesses has changed a bit over time.
multpl.com - S p price to sales

Jim