> And the other other half are undervalued then.

You must be responding to the current HN title and not anything from the actual paper? This comment doesn't make any sense otherwise.

The paper literally has an entire section titled "All unicorns are overvalued" (section 4.2) and figure 3 shows the distribution of overvaluations given their methodology.

It's not a zero sum game. Company valuations are independent.

The other half could be overvalued for other reasons.

That would be true if 1 + 1 = 2 but stock pricing changes by the minute/day/good news/scandal/hiring/firing, etc. Some of their valuations may have been on the money for the minute that it was determined but so many factors come into play even a huge over-valuation could even out next week based on a great quarterly report or a contract signed or if the right CEO is hired on an otherwise adrift ship.

You could at least read the paper abstract instead of relying on the made up HN title...

The actual claim is "almost one half (53 out of 116) lose their unicorn status when their valuation is recalculated". So, yes, if half lose unicorn status, then the other half do not lose their unicorn status. Is that really such a keen insight?

If we assume a simplistic log-normal distribution over the valuations of startups, and overvaluation is exactly half, this would mean that as a whole, startups are overvalued (median<mean)

Nope, because none of them are calculating valuations using their lowest priced shares are they?