FORTUNE -- Kleiner Perkins partner Mary Meeker today published her annual Internet Trends report, and included a section on whether or not current tech valuations are excessive. She acknowledged some artificial inflation, but then cautioned that current conditions are nowhere near as frothy as they were in the dotcom peak of 2000.
To help prove her point, Meeker provided the following data on venture capital investments:
No doubt, there is much less venture capital investment today than there was in 2000. But deal volume only tells part of the story.
Using the same data source, Meeker would have found that later-stage deal sizes in 2014 only were 16.5% lower than they were in 2000. Moreover, later-stage deal sizes in the first quarter of 2014 actually were higher than the 2000 figure.
One could argue that the less-robust IPO market means that these late-stage companies are more mature than their 2000 counterparts and thus are actually being more efficient with their money. But it also is worth noting that the ratio of IPO pre-money valuations to the amount of VC investment fell markedly in 2013 to just 5.5x. That is much closer to the 6.3x in 2000 than to the double-digit marks we saw between 2008 and 2012 (including a whopping 18.2x in 2010).
More importantly, Meeker completely omits any mention of broader VC valuation trends. We know that 2013 valuations were at a 10-year high across the board in 2013 (thanks to Pitchbook), but finding adequate market data doing back further has proved elusive. But Kleiner Perkins should have outstanding historical data on such things -- even if just within its own portfolio -- and Meeker would have plenty of anecdotal evidence for her three-plus years as a venture capitalist. So not including it here is both disappointing and curious.
To be clear, I'm not arguing that we're in 2000 redux. But I'm also cognizant that Meeker has a dog in this fight, since her day job is to make later-stage investments for a Silicon Valley venture capital firm. Moreover, that firm is in the midst of raising a new fund out of which she'll invest. So if she's going to use venture capital data to argue for today's relative moderation, it would best best to not cherry-pick.
Sign up for Dan Primack's daily email newsletter on deals and deal-makers: GetTermSheet.com