The valuation YC offers -- a median of 250k for a two-person team -- would be absurdly high if it was being offered to any team of two recent college graduates; but YC doesn't make offers to just any team. YC only picks the best teams (about 5% of applicants if I'm remembering the statistics correctly)... and it's not unreasonable to expect that the top 5% of teams are worth far more than average.

Similarly, the chance of a YC-funded team succeeding (for whatever definition of "success" you choose) is much higher than the chance of a non-YC-funded team succeeding -- but this doesn't necessarily imply that YC is helping the companies it funds. It might just be that YC is very good at picking winning teams.

This isn't to say that YC is making all of its money by taking advantage of the people it funds, though: Simply by selecting those top 5% of teams, YC provides a very valuable service -- distinguishing teams which have a 50%+ chance of success from those with a 5% chance of success, and encouraging the teams with a 50%+ chance, is certainly economically useful; and the profit which accrues to YC as a result of that is not qualitatively different from the profits which any other smart investor makes by figuring out that a stock is undervalued -- in both cases, the investor is doing work to increase the efficiency of a market.

The economist in me wishes that we could experimentally separate the selection process from the 3 month YC program in order to determine their relative importance -- but somehow I doubt the YC crew would be interested in running an experiment where they either (a) pick good teams and only fund half of them (in order to determine whether the teams they identified as good were likely to succeed with or without YC), or (b) fund a "good cohort" and a "bad cohort" of applicants simultaneously (in order to determine if being funded by YC is enough to make anybody successful).

(In addition to the difficulty of running such an experiment, the YC crew might not want to have this question answered: They probably don't want to be confronted by teams saying "well, were interested in being funded at a valuation of $200k -- but knowing that YC thinks that we're a strong team has resulted in us increasing our self-valuation to $800k".)

But in the absence of experimental data, the question remains open, with all the attached room for anti-capitalist hatred: Does YC help the teams it funds... or are they just shrewd investors?

As for the question of whether we help people or not, I don't think many people who know what goes on inside YC would argue that we didn't improve startups' prospects by 6.4%, which is what we have to do to earn our keep. Considering the amount of effort we put into it, we'd have to be pretty ineffective not to manage that.

EDIT: Just reading the TC comments, it amazes me how often this basic misunderstanding comes up, even after an article that explains it.

No but explaining this might help HN readers to resist posting these types of stories.

Competition of any sort generates "trash-talk" from competitors and the like. Joel has a nice explanation in "Fire & motion" ~ http://www.joelonsoftware.com/articles/fog0000000339.html why competitors might do this. HN readers are ( "unnecessarily" ) getting caught up in the cross-fire.

In the case when journalists are those who "dislike" and not people, someone could think that their motives to write something bad are not so innocent, if not directed by someone who pays to create a bad impression.

Although to be honest the majority of press I see about YC is really positive.

In some ways, YC represents the purest form of the capitalist ideal - people risking their OWN money and resources to help develop a productive enterprise, and then reaping the rewards if it's successful (presumably the YC founders are risking their own money and their time with the startups they fund...).

I think most people reserve their anti-capitalist vitriol for things like hedge funds, where the managers of said funds can get insanely rich taking risks with other people's money.

I find the YC model very hard to hate.

distinguishing teams which have a 50%+ chance of success from those with a 5% chance of success"

Perhaps we should give them a few more years of operation before we assume those to be facts. So far the vast majority of funded teams haven't resolved one way or the other, which is actually a pretty good sign, but I think I'd still bet the under on 50%.

Pre-emptive note to anyone who replies to my comment above: I'm not likely to respond to anything you say, but please assume that I'm not being an idiot.

/me wanders off to find some ice for his wrists...

The startups offered YC-funding, who declined it... how did those guys work out, in comparison? (although the offer itself is an important YC input).

Also, the YC guide says they they decline funding to many very good applicants (because YC has limited places and they go to the very very good applicants). It sounds like these guys would have been selected, in some other batch... YC knows who the "almost selected" guys were and could do a comparison.

The only way to measure YC's ability in picking winners is to to have PG & co rate a bunch of non-YC founding teams at the early stage of their startups, wait a few years, and see how well their predictions held up against more or less random predictions. The experiment must involve non YC startups is that YC enhances their startups' chances of success, which can create the likely false impression that YC excels at picking successful teams rather than at guiding the teams it happens to pick to success.

In poker, if a pot is very large, calling a bet (ie investing some money) can have positive expectation even if the probability of winning the pot is quite low.

Similarly, YC's filtering ability only needs to be "good enough" to generate positive expectation, which can happen even if their predictive ability for selecting winners is quite low. Picking winners at a slightly better rate than the competition gives them an edge, even if they're not actually good at it. ie being "barely competent" at something extraordinarily difficult can often be enough to achieve considerable success.

Of course, I don't see any reason to believe that they aren't actually quite good at choosing winning startup teams -- it would certainly be interesting to be able to measure that. It will also be interesting to see if their hit-rate goes up over time.

none the less it's obvious by now that YC at least acts as a proven talent filter which I think is highly desirable by VCs

TechCrunch also usually covers every YC project as it launches.

This is the main value of YCombinator: It provides guidance and media exposure that would otherwise take years to achieve, if at all - how many people can say that every project they do will get a story on TechCrunch?

Judging it as a VC company is misguided. The money involved is less than a year's tuition in many US colleges, and the recipients are mostly young enough to just live with their parents for a few months, and not need it anyway. The funding is minor, but the other parts - startups pay, a lot, to get this level of publicity and consulting.

I'd go with the "record label" metaphor instead of Lacy's American Idol one (I'm older ;)): Find talent, give them some producer and studio hours to record a single, then use your connections with DJs to get it some airplay hope it will catch on. Watch "24 Hour Party People" for more details :)

The one piece missing here, though, is the fact that ultimately the public (pre-mp3) was paying money for the singles. YC's audience is big companies with money to spend - it may prove less sustainable in the long run.

No no no. It truly is not-- it blows me away how often people say this. Look at the traffic graphs of most YC companies and they have a healthy spike around Demo Day and then nearly flatline (and start the rough/long trudge up the hill).

If I were going to put things at the top of the list of values YC provides, I'd say it's:

- the (very) public declaration of intent-- it becomes DAMN awkward to quit once you're a YC company.

- the investor relationships, introductions, and acceleration. It's nearly impossible to pitch 150+ investors at once without YC. That setting results in faster investment oftentimes (but not fast enough).

- YC guidance. They've shepherded 100+ startups thru this now-- they have more data and better instincts about the "right" thing to do than most advisors... And invest way more time.

- The peer relationships. Very few early stage questions can't be (very effectively) answered by the collective wisdom of YC founders (who have all gone thru the same issues).

More importantly, I think is that it's giving young programmers a big leg up in understanding how the VC game is played and an opportunity to build a software career that is very independent. For really good programmers I suppose there is always demand, always a place to work and more or less define your own path. But if you sell your company for 15 mill, you get a nice piece of what I like to call FU money.

I've sort of followed this YC project from the beginning as an outsider and my sense is that the folks behind YC are putting back some of their good fortunes. It's a cool way to spend some of their money and possibly grow it even more.

All the negative press and trash talk about burned out programmers, etc. looks to me like a good measure of success. The others are looking at the backs of your sneakers

That's not exactly true. I think majority of the startups were launched or had extensive prototypes before they joined YC.

This makes sense, because it filters out the people who are too lazy or not motivated enough to build an amazing product. (But it also filters out very smart and motivated people with awesome ideas who are too busy, as many smart people are.)

In any case, the risk that YC takes isn't quite as big as investing in two kids with a pipe dream. All of the founders seem to be carefully selected for their propensity towards success.

- Are there any technology-based startups funded by YC?

- A proxy could be patent applications: how many YC startups have filed for patents?

I'm looking at Justin.tv, and I've never seen smooth live video online before. It's consistently smooth (though minor freeze-frames every so often). Your implementation is a better method for live video.

Is this because your implementation is coded better - tighter, more efficient, cleverer - or is there a fundamental new insight behind it, that makes it faster?

Of course, there have also been insights that I think may be fundamental, but they aren't patentable as far as we know.

But the two kinds of technology-based businesses make me think of Google's technology: server implementation (fast results) and pagerank (relevant results).

Most of the criticism that I've seen seems to come from, well, critics, and not people who actually build stuff so that may have something to do with it.

However, I think there is some legitimate discussion going on in these anti-YC articles.

For example, it is an interesting and open question whether it's wise or healthy for the market to develop so-called "built to flip" companies. While YC isn't exclusively in the built-to-flip business, it happens to be a vocal advocate for this model where there once was none.

Generally, PG (chief ideologist for YC) is interested in doing disruptive things in the realm of startup funding. Disruptive people tend to often be both spectacularly right and spectacularly wrong -- PG no exception -- and I think that traits make a person more polarizing.

I don't think it's necessarily a knee-jerk to dislike a company because you disagree with it's chief ideologist. Look at Apple and Steve Jobs -- it seems to me that everything about the guy permeates every nook of the company. If you don't like him, you're probably not going to love Apple products.

the original post on Sarah Lacy's blog is atrocious, I can't see why MA bothered to respond

It seems like publishing link bait never fails to garner an undue amount of attention.

I'd like to say that I have learned a LOT from reading the contents of this site. I'd imagine any YC hatred is a sign that YC is actually do it's job well!

It's true that YC companies receive more exposure than similarly-early non-YC companies. Well, that's one of the benefits. With YC acting as some sort of filter for minimum quality, blogs like TechCrunch know there's at least something worth looking at.

oh wait...