It's insane to use fundraising as a proxy for success, especially for venture capital. If YC grads are raising less money, it could easily mean their businesses are more successful.

The problem is that YC itself uses # of companies that receive additional funding as a proxy for its own success.

This could be worse news for YC than it appears.

Honestly this is a very strange metric for me. It doesn't account for things like the frothy conditions of 2011 - 2014 where people were raising money like crazy.

It doesn't speak to trends where investors were actively looking to nourish alternatives to YC since there was a perceived monopoly on investor mindshare at YC.

It doesn't explain what the series A cumulation means for investors, nor startups, nor the incubators themselves.

If this article is for investors, wouldn't it better to understand and compare the total valuations of startups from each accelerator (by batch would be best.)

If this is for startups, wouldn't it be best to compare percent of batch that went on to raise series A?

edit: I don't mean to suggest that the data isn't interesting. But given the extremely human nature of startup investing, I think it's worth speaking to qualitative factors when trying to bring insight.

The article offsets the data by the year the incubator started. This is not relevant fort a start-up having to pick an incubator today. I think it would be more interesting to see what percentage of current graduates raise series A. My guess is that this is highest for YC but still only one in five. Of course there are also cash flow businesses that don't need to raise but these are the exception rather than the rule. Disclosure: winter 2015 YC graduate and a big fan of YC

Huh? From the article, 500 Startups is not even close:

"Y Combinator companies comprise 4% of Series A deals"

"500 Startups produced 1.8% of Series A recipients"

Really what matters is rate of return. It doesn't surprise me that a company called 500 Startups gets a lot of Series A deals.

I understand what Mattermark is trying to do here, and applaud them for trying, but the way the data is presented doesn't withstand scrutiny. There are far too many interceding variables for this analysis to have any merit. Namely (1) total deals done/varying market conditions in a given year, (2) yearly cohort size, (3) the fact that most Series A deals happen within 3 years of accelerator graduation, (4) not accounting for large seed deals, later stage deals, and exits. I don't quite understand how number of Series A deals equates to success in the first place, and I'm having trouble finding what this analysis proves, if anything.

It's really am issue with problem of induction writ large for measuring VC returns w/ averages(say for ROI/ROR(Like this better)) asking for averages when the entire point is that averages are likely not stable for comparisons.

Do averages of returns across VC stay stable even if they are supposed to manufacture black swans, which is what it actually is, and not "Unicorns".

Also most people calling it Unicorns is probably an indication that that segment of detractors is undereducated and should be ignored as there are plenty of other more accurate terms.

If the means are an irrelevant metric to compare across VC's/YC-Like seed funding groups, then we should look for a more stable metric to compare performance as in this field and in the market, as always, the past is not indicative of future performance.

This is really difficult for the human mind to perceive isn't it? Nevertheless, let us hold still, but still make haste.

Which recent YC companies do you believe will become the most successful? (omitting the obvious Airbnb, Instacart, Gusto, Stripe and Reddits)

It would be more interesting to see aggregate metrics by yearly cohort, like rev, valuation, etc.