This blog post could be interpreted as a way of providing more information to founders, so that they can't then say "But I didn't know that accepting another accelerator would lower my chance of getting into YC." Now you do.

Also, it's a true statement that AirBnB and Dropbox succeeded because they were great companies, not because they went through YC. At least in the early days, YC had a very strong bias towards selecting companies that would succeed without them. You'd have to ask someone affiliated with YC if that is still true, but given comments from YC partners here, I suspect it is.

That statement is not mutually exclusive with YC being able to help companies that would've succeeded without them, and indeed, both Brian Chesky and Drew Houston have said that some of YC's advice was very helpful to them in critical early stages.

Interesting, my perception is that it's gone the other way; that not only is YC now primarily accepting companies at a later stage when they've already been substantially derisked, but it's also investing in less risky companies to begin with. Several YC partners have said this as well, e.g. Jared Friedman:

"Companies are joining YC at a much later stage. When I started YC, most companies wrote their first line of code in the first week in the program. Today, many of the companies have been working on their business for a long time and some even have substantial customers and revenue before applying. If the companies in my batch applied to YC today, I doubt that many of them would get in." http://blog.jaredfriedman.com/2015/08/18/nine-years-of-demo-...

While I'd like to believe that YC is still willing to fund companies like JustinTV or Loopt, where there is substantial risk in terms of both the technology and the sociology, the reality is that if you go down the most recent batches of demo day companies most of them look like Dollar-Shave-for-X or Uber-for-Y. Maybe that's a slight exaggeration, but perhaps the most striking thing about the most recent batches is how few of the companies are obviously terrible ideas. Unlike back in 2006 and 2007, where probably a third of the companies failed to even launch, and another third probably never got more than 100 users or whatever.

Obviously YC is now also funding energy and biotech companies that are at least as risky as anything from 2005, but if you look at the verticals where they've been active the longest it certainly seems like the trend in those areas is toward later stage and lower risk.

I am sure this is a reflection of the companies they get to look at. If we use a dating analogy they seem to be getting more interest from supermodels, but this doesn’t mean they are getting more interest from potential life partners.

I think it's the opposite actually; there are an effectively infinite amount of startups, but the bandwidth of YC is (currently) finite. What we're seeing now is exactly what you'd predict just from the math, even without any data.

No, not necessarily.

Every startup in the batch had a unique end goal, but they're not necessarily traction in the way YC appears to view it. Sometimes the end goal is a pre-traction milestone, like shipping our first set of physical chips. In this sense, it's a blurred line between pre-accelerators and early-stage accelerators

Personally, I'd contend that YC today does not really like funding "very early-stage companies". Or at least, the definition of very early stage has been inflated from 2005 to now. The bar has been raised too high [1]; there are too many good applicants with significant product and/or traction already.

1: https://www.ycombinator.com/whynot/

I wish more people would apply this type of skepticism to tech investment in general. I think there's a lot of shady stuff going on that doesn't really get discussed -- imo, there are significant problems with the conventional startup investment model.

That said, in this case, I think it makes sense to say "Why wasn't that help sufficient? Were they really bad accelerators or are you just wasting money?" I also think it's important to recognize that a lot of small local accelerators are just bad accelerators. Since the most competent people are often the most keenly aware of their inadequacies, it makes sense that some really good companies would go through a local "starter" accelerator program, feeling unworthy for YC out of the gate.

The more cynical view would be founders going through multiple accelerators are more focused on appearance and "playing the startup game" than actually building a real business.

Cargo culting seems very prevalent in the current startup culture -- especially in those that are "playing startup".

These founders see all of these external things that startups do and assume they must be necessary for success -- or that they will at least will help cover up any problems they have.

Unfortunately this is just like carving wood into the shape of a radio and expecting it to work. It might look the same but it has none of the things that make it a radio. You can even attach a real antenna to it, but it is still just a block of wood.

Actually I think cargo cult thinking is more of a problem at the accelerator/angel/VC level than at the startup level. We have all these funders copying the YC model and relatively few trying to do things differently. We need much more diversity in funding and acceleration models. All roads might lead to rome, but we don’t all need to travel there on the same bus.

Yes, these people could exist, and there are legitimate reasons for doing multiple accelerators as well.

The end goal for a very early stage startup accelerator or incubator can be a lot different than a later stage accelerator.

Different accelerators for different specific purposes at different stages in the lifecycle of the company.

Because they're not claiming credit for the success for their alumni, and they are also giving other accelerators the benefit of the doubt that they are good too, and acknowledging that the primary factor is the founders themselves. If they don't thrive after going through another accelerator than why would YC be any different? I don't see the conflict here at all.

Founder team B with resources (M+other accelerator) accomplish Y.

If not (Y >> X) => Founder team B did something wrong or squandered the extra resources. That's a bad signal.

This holds unless you make the assumption that the other accelerator provided zero value. And if it did, then that reflects poorly on the founder team that chose that accelerator, baring something exceptional happening.

Pretty straightforward to expect more from teams who have had access to more resources.

We found out that having left an accelerator program, it is difficult take a potential deal with YC, as the cost of further dilution has to go with potential higher valuations in SF.

In fact, apply for YC first. If you don't get in, but have a respectable accelerator program and network, consider moving on and push it. Many roads lead to Rome.

Not all accelerators are equal. Going to college may or may not change your life. Going to Harvard, Stanford or MIT probably will

[1]even if you had really good financial or location-based reasons for it...

Occam's razor...

Or that the idea was less than successful and the executives could not execute even given resources.

that being said, I'm sure there are outlier scenarios where a company can plausibly say "we signed up for accelerator X and they sent us down the wrong path for x, y, and z reasons. we'd like to do it right this time around." and YC would take that into account.

edit: words r hrd