Hacker News Reader

Comment by pg | Hacker News Reader

I should know better than to make jokes that align with people's existing fears.

Once we funded someone who reminded me of Zuckerberg, and even then it was far from the only reason. And the fact that the YC partners have taunted me about it ever since shows that if anything we're probably extra skeptical about Zuckomorphs.

I realize it's possible to be unconsciously influenced by applicants' appearance and manner, but can anyone at this point believe we're not aware of that danger, and simply pick people based on matching some crude archetype? That we could remain that naively incompetent after 16 cycles of picking founders and then watching how they do?

View on HN

I'm going back and forth about how smart it is to reply to this seeing as the point I was trying to make was not "Wah YC rejected us because we're different!!!"

For the record, I've always really respected YC for being honest and up front about biases. I find it very refreshing, because most everyone else is too scared of being accused of discrimination to admit it. Everyone has biases. Every VC and angel is going to try to match what's in front of them to something they know when assessing the potential value. I have no doubt that YC will accept a great company that doesn't fit the "type", we spoke to lot of companies like that before we came out to SF.

What I'm curious about is YC's habit of "fund without idea" or "fund for new idea" ... how likely is it that someone against the traditional startup type would be given that opportunity? I don't know the answer because there's no data on what ideas people came into YC with, vs what they left with ... but it's an interesting question to me because cognitive biases are an interesting topic to me.

The people who think they are smart enough to think their way out of their own biases are usually the ones who fall victim to them the most often. So I've always really seen YC's honesty about this as a GOOD sign, not fodder for criticism.

The thing working with data has taught me is that humans are bad about guessing outcomes. Expectations are constantly trumped by the data, if you look at the data with an eye for your own biases.

In any case, thank you for sharing your experience.

This isn't a dig, so don't take it as one. But you mention in the article that you have felt like this your whole life - people have tried to fit you against a preconceived stereotype and you don't fit one. You also said that before the interview you were worried about the interview for that reason. Perhaps that's your bias?

1) No one is going to believe that you can accurately compensate for your biases. If you are in fact biased towards Zuckomorphs, meta-awareness is only going to help so much with that. You don't have to be naive, you don't have to be incompetent, but you're still biased. OH WELL, LIFE IS HARD.

2) Respectfully, pg, why are you bothering with this, and more with all the other "how to hack YC", "why we got past YC", "why we failed YC" blog posts? I'm surprised you don't have a bot that first-posts the following comment to every such submission:

"On behalf of YC, there is basically zero chance that TFA gives meaningful insight into YC's hyper-evolved process for seeing past such rudimentary schemes. -- the PGBot"

You're missing an 'I worry' in there somewhere.

Heh... I'm sorry, but I can't help but be slightly amused by the idea of a "pgbot". So much so that I started on capturing some "pg'isms" in AIML a while back.

https://github.com/mindcrime/pgbot

I'd love to play with this more, but I have to much respect for pg to do it. Any more and I'm afraid it would go from "cute, maybe amusing diversion" to the realm of actually offensive.

What might be fun though, would be to compile some stock expressions, advice, quips, etc. from a whole range of well known personalities in the startup & tech world, and create a bot that combines them all. You could call it "startupbot" and stick it out there somewhere for people to interact with.

You would probably find this amusing - http://www.braintripping.com/new/paul-graham

Be careful about assuming that everyone who could productively engage with you understands the world as you see it. Who else has done exactly what you're doing before? Nobody. Frankly, knowing how competently you do or don't pick teams is your job, not theirs. Your comment makes perfect sense, but ends standoffishly.

The main takeaway from this article for me was not the point of failing to look like Zuckerberg, but the importance of being likable and project a certain image of (the promise of) success. It's clear from the beginning of the essay that this was the main issue even among the group of founders themselves, so it was no surprise this deficiency translated straight to the interview.

I didn't really read that as the main take away of the post. And I didn't get the feeling that they were trying to say that you do simply "pick people based on matching some crude archetype". I read it as a more general analysis of the dangers of making those kind of snap decisions (note all the stuff about how the team-members didn't even necessarily care for each other, based on initial impressions). The bit about you and your Zuckerberg quote was just one element of a bigger picture, from what I saw.

Maybe it's just me, but what I took away from the story was "don't hold too strongly to rush judgements about people, and don't worry so much about what pg and the YC crew think about you, regardless of what you look or act like". But that could just be a reflection of my own biases.

Whether or not we (humans) can adequately compensate for our own cognitive biases, even when we are aware of them, is an interesting topic. Personally, I'm glad this post showed up and helped raise that issue.

This makes me curious -- let's say Mark Zuckerberg himself applied to YC with the objectively worst idea you've ever seen on an application.

Do you give it a shot?

Of course!

This reminds me of one of Warren Buffet's famous quotes:

"When a management with a reputation for brilliance tackles a business with a reputation for bad economics, it is the reputation of the business that remains intact."

Taking this thinking another step, one of the characteristics of good management is probably the ability to avoid bad businesses.

If Zuckerberg were to leave Facebook to start another startup, and analysis of the idea makes an investor think it's a bad idea, I'm guessing they would still invest in Zuckerberg based on the assumptions:

1. He knows something you don't. 2. He'd move from a bad idea to a good idea later and it's worth investing in smart businesspeople.

Tablets were a bad business before Apple got involved.

Tablets weren't a business before Apple got involved.

Go Corp and several successive teams at Microsoft would beg to differ.

https://en.wikipedia.org/wiki/GO_Corp.

From that link:

Founder Jerry Kaplan says that in its lifetime, the company generated "no meaningful sales".

Business => $$$.

Reminds me also of one of Marc Andreesen's rules for startup success:

  1. When a good market meets a bad team, market wins.
  2. When a bad market meets a good team, market wins.
  3. When a good market meets a good team, something special happens.

Also, Zuckerburg himself has said that the reason he refused all buyout offers for Facebook was "I'm unlikely to have an idea as good as this ever again." That was probably the quote that changed my opinion of him from arrogant douchebag who got lucky to someone who just might have his head screwed on. It displays an interesting level of humility that's largely missing from his public persona.

" "I'm unlikely to have an idea as good as this ever again." That was probably the quote that changed my opinion of him from arrogant douchebag who got lucky to someone who just might have his head screwed on. It displays an interesting level of humility that's largely missing from his public persona."

You know, Zuck actually seems like a nice guy. Someone I'd love to grab a beer with. He's obviously smart, incredibly humble, (supposedly) easy to work with and doesn't really fit the existing molds of a wunderkind CEO like Jobs or Gates.

Note that a16z invests in startups at a much later stage than YC does. For YC, the quality of the team may be the most important because they're at a stage early enough to significantly change the market they target. By the time a startup is looking for series A (or later) funding pivots are relatively more expensive.

Yeah, I suspect that YC is willing to invest based on team because they figure the idea can easily change anyway. I've heard that one of the points of the YC interview is to weed out inflexible founders that are not willing to radically change their approach. And Andreesen has recommended that "Until you achieve product/market fit, keep the team as small as possible.".

Wait, do you mean NOW or do you mean when he was 18? :)

If possible I would like to ask one more question: let's say Larry Elision applied to YC with the bad idea.

Do you give it a shot?

%s/Mark Zuckerberg/Elon Musk/g

10 years ago, an electric car company would sound like objectively the worse idea to many. Tesla was founded in 2003, the same year that General Motors officially cancelled its EV1 program. Toyota stopped selling its Rav 4 EV. Honda had long since ceased production of the EV Plus. If 3 of the largest car companies in the world tried and failed to make a profitable electric car, what hope did a newcomer with no automaking experience have?

I knew almost for sure pg or someone from YC was going to answer something like this.

This man has probably overseen the early days of more successful startups than anyone else in the world. That huge amount of heuristics (pratical experience) loaded in a single head is the reason I listen to what he says. He's wise.

But wise does not mean "he who gets it right every damn time". I guess there's likely a psychological explanation as to why we love opportunities to pick on the flaws of successful people. "Hey look, pg made a mistake! Ha, I knew he wasn't god!"

It would probably help if you shared more anti-zuckomorph stats / anecdotes so people could chew on those endlessly instead.

Got any?

1. There should be some kind of control group.

2. If the world is in fact biaised in favor of Zuckomorphs, then your own biais would be a good thing, but a biais nonetheless.

Reply on news.ycombinator.com