Why looking at financial metrics for early stage VC firms isn’t as revealing as you might expect

It’s pretty hard to tell who’s actually good at venture capital.

Sure, I’m being mildly facetious, but I’m mostly serious. I came from covering the public markets, where you’re swimming in numbers—revenue, Ebitda, you name it. But in the private markets, as a reporter, brass tacks numbers on how startups and the VC firms that back them are performing are a whole lot harder to come by.

Thanks to the Freedom of Information Act, however, I can sometimes get a public records peek into how funds are performing. Recently, I got back a FOIA request from the Los Angeles County Employees Retirement Association (LACERA) and the set of numbers I found most interesting revolved around Primary Venture Partners, Storm Ventures, and Canaan Ventures—all early stage-focused firms.

So, we have some numbers. Cool, right? Not so fast. So, this is ITD IRR—inception-to-date internal rate of return. That means, in short, that these returns are specific to when LACERA invested. Taken a step further, there’s actually only one vintage here we can really assess, since there’s only one fund that’s been totally completed: Canaan VII.

That ITD IRR is 14.62%. Now, that number might look low if you’re expecting a 20% or 30% return, but taken in context for that 2005 vintage, Canaan VII is a high-performing fund, per IRR. According to PitchBook, the top decile for that specific vintage was 12.59%, and according to Cambridge Associates, the upper quartile for 2005 was 10.71%.

Still there’s only so much we can assess here. As one investor reminded me: The “investment multiple” metric is squishier than it looks, since we don’t know if it’s TVPI (total value to paid-in) or DPI (distributed to paid-in capital). And there’s a pretty big difference: “DPI is actually cash returned, TVPI is just paper valuations,” the investor said.

From there, for every other fund on this table, it’s a wait-and-see game. Though some may think of IRR as a holy grail metric, it’s not really that simple. As one LP explained to me: IRR shouldn’t be used to compare funds that are less than ten years old, as it can provide misleading results. And for the really new funds that have barely called any capital, looking at their IRR right now is kind of like looking at the score in the first quarter of an NBA game—it takes time for investments to generate returns, and in most cases there’s still a lot of capital to call. Obvious, but bears repeating when looking at all this in context. I looked to PitchBook analyst Kyle Stanford, questioning the entire premise of this project: Are these numbers actually useful?

“It’s a really good way to describe the nuances of venture,” said Stanford. “You can talk about a 14% IRR in 2005, which was great, whereas maybe a 20% IRR in 2012 was mediocre…And when you talk about this, it speaks a lot to why LPs are looking for vintage diversification.”

There is, of course, some interesting stuff going on in this data. Take that Primary IV ITD IRR, of a whopping 91.26%.

“Despite the tough environment, this has been the fastest we’ve seen one of our early stage funds take off,” Primary cofounder and general partner Brad Svrluga told Term Sheet via email.

Though that number will shift over time, that’s a killer start to the first quarter of a long game. But that’s just it, right? For essentially all of these vintages, the game is still very much in progress.

So, ironically enough, the more numbers I’ve gotten, the more they reinforce for me just how long of a game venture really is. And the funny thing about it is: It strikes me that even if venture were more numerically transparent, it would still be tricky to quickly tell who’s winning the race—because the race itself is in slow motion.

See you tomorrow,

Allie Garfinkle
Twitter: @agarfinks
Email: alexandra.garfinkle@fortune.com
Submit a deal for the Term Sheet newsletter here.

Joe Abrams curated the deals section of today's newsletter.

This story was originally featured on Fortune.com