1.) means nothing without also knowing what percentage of companies seeking funding are from YC. Hypothetically, if there are 2000 companies seeking VC funding, 20 companies that get into YC, 100 companies funded, and 10 of those funded companies are from YC, then only 10% of VC A-rounds went to YC companies. Yet your chances of getting funding given that you got into YC are 50%, yet your chances of getting funding otherwise are 5%. Most entrepreneurs would give up 6% of their company for a 10-fold increase in the chance of getting funded.
It is much easier to figure out what percentage of VC A-rounds go to YC companies than it is to apply to YC: VCs publish their portfolios, often before companies even launch. We did pretty much that when we started:
(n.b.: large). Instead of YC A-rounds, we were looking for security A-rounds. It took a few hours. This is very OCD, but hey, you can spend a few hours doing nothing on IRC.
My point is, I think your numbers are wrong, and no way do 10% of all A-rounds go to YC companies. Do YC companies have a 50% chance of getting an A-round within 12 months? I guess Graham can tell us.
Most entrepreneurs would not give up 6% for a 10-fold chance in simply getting funded. You will find this hard to believe, but I'll report it to you firsthand: some companies turn down the opportunity to get funded. We haven't even gotten into the terms discussion. Say 50% of YC'ers get funded. What percentage of them get good deals, and what percentage get tranched, 3x prefs, tiny valuation, and loss of the board? 0% of bootstrapped companies have those problems.
Late edit: there are more than 100 VCs on that radial chart. So much for 10/100=10%.
Do YC companies have a 50% chance of getting an A-round within 12 months? I guess Graham can tell us.
Off the top of my head, I'd say the ratio is about that, of companies that want to raise series A rounds. Most don't actually want to. They don't need that much money, so they go with angels because it's so much less restrictive.
This is all besides the point --- I think it's interesting to think critically about YC, but really, who cares?
I'm saying, to the person who said "I wish it was that easy", it is exactly that easy. Most of the companies YC "funds" don't need VC to start up, which is good, because YC doesn't give you that kind of money anyways.
There are lots of paths to success. Look at virtually any Mac software startup; none of them get VC, and they accomplish ridiculous things. Which is harder to build, Reddit or Quicksilver? Reddit was Graham's "hello world" for Arc.
"You will find this hard to believe, but I'll report it to you firsthand: some companies turn down the opportunity to get funded."
Not all that hard: my last employer turned down VC financing. And I'm going a similar route with my own startup: I'm not averse to VC financing, but only after achieving "product/market fit". There's no sense taking other people's money for an idea that may not fly, and I can change direction more easily without employees or investors.
I still would like to see actual numbers for the 4 quantities I mentioned. My point is not that you're wrong, it's that without that data, you can't draw a sound conclusion either way.
"What percentage of them get good deals, and what percentage get tranched, 3x prefs, tiny valuation, and loss of the board?"
One of the advantages of YC, as I see it, is that this is far less likely to happen. YC brings bargaining power: if a VC screws over a YC startup, you can bet that YC will not be bringing any future deal-flow to that VC. You don't have this advantage as a lone startup, regardless of how good your product is. Never underestimate the power of incentives in getting people to behave properly.
Is there any evidence, at all, that YC "sponsorship" improves deal quality? "Most" YC startups don't take A rounds (Graham). About 50% of those that try to, fail (Graham). If YC takes 20, and we generously assume 9 of them go for A-rounds, and 5 of them obtain them, what percentage of them get good valuations?
How many YC "sessions" have there been so far? If each produces 4-5 funded companies 6 months later, this isn't a hard question to answer. For instance, most companies publish their boards of directors. Let's see which ones got screwed.
