Of course, that's what all job interviews are like. If they get a bad vibe from you or think you're not a "culture fit", they can quickly reject you based on a short impression. And that's not necessarily a bad thing.

What bothers me is the reptition of the idea that any one person can be some kind of magical oracle of character judgment, bordering on having psychic properties. Especially without strong scientific evidence. I'd rather PG use less black-and-white language.

The thing is, however, that everything you do will also have this metric applied to it, by everyone you interact with, and it is a metric that impacts success. Any sales interaction you have, any hiring process you implement, any investor meeting you have, etc. All human interactions will be judged with this metric in addition to the other metrics. Is it "fair"? I don't know. But, it is reality. Given that, having someone who is good at it, on a team that is otherwise not good at it (and having read pg on nerds and popularity, I can surmise he considers himself not terribly good at it; nerds in general are famously socially inept), is worthwhile.

Which raises the concern that the social ineptitude of a technically brilliant founder might prevent them from getting into YC (and I think that is the fear being expressed in this thread). I can say that most YC founders I know are charming people; nerdy, mostly, but still charming and socially adept, at least when interacting with similarly nerdy peers. Is this a prerequisite for startup success? And, is it actually what is being selected for when we talk about this "social radar"? I don't know. I don't think I have this particular skill strongly enough to recognize a hit and a miss on these metrics, though I can spot technical fakery a mile away (and there's a surprisingly high number of applicants who are technically incompetent trying to pass for competent; I suspect few make it past the application process).

>We nerds (and I am among this bunch) don't really like the fact that popularity, "emotional IQ", and social influence are not quantifiable, but we are judged on them, anyway.

It's not exactly that such things are not quantifiable, but that PG is unable to express what is being judged other than in the same few, undefined words, at least in this essay. An example of such judgement is never given, it is only asserted that correct judgement can be dispensed by Jessica. That is a magic oracle. No basis is provided for the judgement (other than its source), only the judgement itself.

Now, as others have pointed out, I doubt this is exactly how it happens. Most likely Jessica brings to light some contradiction, however, the essay doesn't go into what those contradictions have been, or might be. The essay really only provides the view that the contradictions are blessed and therefore automatically accepted.

The concern here is that PG is propagating concepts like "character", "Social Radar", etc., without being able to define what they are. Others may try to replicate this and start their own cargo cult, which has applied its own secret definition of those words, in order to work the magic sauce.

What is so hard about the concept/metric being applied here that it can't be put into words?

In every one of your responses, for starters. ;-)

"Others may try to replicate this and start their own cargo cult, which has applied its own secret definition of those words, in order to work the magic sauce."

As I mentioned I would like for more investors to try to behave like YC, even if they're unsuccessful in the attempt.

And, there are already many investors cargo culting the YC process. I don't think that's a bad thing; they aren't as successful (I guess TechStars is the nearest analog so far), but they're trying to replicate the winning formula. They may fail in a variety of interesting ways, because it is cargo culting in many cases, but by trying to do things more like YC they're likely making the world a little better for early stage founders.

"What is so hard about the concept/metric being applied here that it can't be put into words?"

Again, I would guess pg doesn't understand it, and so can't quantify it, but trusts that Jessica's correct more often than not. It doesn't seem like pg was even trying to show everyone how to do what Jessica does, just to clarify that she does many things within YC and that she has often been forgotten in the telling of the YC story; explaining that does not require him to explain how it works.

> Again, I would guess pg doesn't understand it, and so can't quantify it, but trusts that Jessica's correct more often than not. It doesn't seem like pg was even trying to show everyone how to do what Jessica does, just to clarify that she does many things within YC and that she has often been forgotten in the telling of the YC story; explaining that does not require him to explain how it works.

I can have a stab at an explanation here. I'm sort of straddling the border (being a "socially inept" nerd by nature, and having invested considerable amounts of time over the last 15 years in getting better at the "social stuff") and know enough people who are on the other side, and have spoken with them often enough, to have formed some ideas about this.

First, most people who are really good at this tend to be at a stage of unconscious competence. It's not really something you teach to others very frequently, so it's rare to find someone who spends the time to analyse their own analysis of others, which helps explain the "mystery factor" here. As they say, if it takes 10 years to get good at something, it takes another 10 years to get good at teaching it to other people!

In my experience people who get very good at this tend to be (like Jessica is described in this article) people who have a strong discomfort around conflict and/or other people's distress. I don't know what the chicken/egg situation is - which comes first, the sensitivity to other people's distress, or the skill in reading other people's emotional states? Either way, the two combine and lead to someone who spends a lot of time being aware of how people around them are feeling, what they're thinking about, etc, so that they can detect potential conflicts very early and head them off before they cause distress. Like anything else, you get good at what you do a lot, and someone who spends a lot of time thinking about what's going on in other people's heads is, over a few decades, going to naturally develop an incredible (from the outside, almost magical) skill at reading people from what are almost unnoticeable cues like tone of voice, body language, the content of what they say, what they don't say, etc.

Ultimately what this boils down to, imho, is building a model of the other person in your head. Think of your best friend, the one you understand most - you probably have a model of them in your head. You could have a conversation with that model and, if you have known them for a while and are not totally insensitive to people (which is possible... don't beat yourself up about it!) you can probably predict how they would react to a given situation with a fairly high degree of accuracy. Sure, you'll get it wrong from time to time, but you'll get it mostly right. People who are good at this "social stuff" just build those models much, much more rapidly, and much more accurately, than you do, through practice and habit, and they do so constantly throughout the day with everyone they bump into, so they get to practice that over and over again, and keep getting better.

Does this help a little?

This social radar is the ability to spot people that would "stab you" if needed.

That intuitive part of the brain represents the vast majority of what we do day to day. Our rational brains activity feeds into it, too. However, cues about people's speech, behavior, emotions... these are naturally all picked up best by the emotional brain. Jessica may have been doing what Paul describes for much of her life with that part of her brain soaking in details she can logically spot and some that are unconscious impressions/feelings. Trial, error, and external observation correct inaccuracies over time if one lets them. At some point, the models in her mind were honed so well that they can spot significant positive or negative traits very reliably.

So, there's nothing wrong with Paul's description and it's likely Jessica couldn't fully explain her mental model because she doesn't know it. She will certainly, with introspection, have elaborated out many specifics that she could explain and you could train yourself to work with. However, as I said, this is mostly a subconscious process that can at best only be partly elaborated. Like any black box, you can only assess its reliability by looking at the quality of outputs that come from inputs.

There will always be false negatives, false positives, and occasionally WTH!?'s from intuitive decisions about people. What's good in this space is quality of results whose accuracy is consistently good with relatively-low, error rate. Paul's statements indicate her emotional/intuitive brain uses very-effective models of people far as their character goes. The results speak for themselves. That's all you should need.

Note: Only way to learn such skills is experience. A job where you deal with lots of people in ways that makes their ethics show can accelerate the process. Still takes years and years, though.

To be clear, I did not say it is cult-like, only that it is unclear.

If YC does operate on a magic oracle, then I might argue that it might be a cult.

>Of course, that's what all job interviews are like. If they get a bad vibe from you or think you're not a "culture fit", they can quickly reject you based on a short impression. And that's not necessarily a bad thing.

I'm not sure what it means that a YC interview may be thought of as a job interview, but I share your concerns. I'm also concerned that the essays may encourage such behavior. While the teams of the sort YC are organizing may benefit from such a selection process ("culture fit"), without any logical explanation of the process, the essays essentially encourage people to go out to start their own cults.

Unfortunately, none of the descriptions demystify the magic. Maybe I am reading the descriptions incorrectly, but the actual logic behind the magic is never explained. The descriptions only contain an appeal to a magic oracle power.

>But after the interview, the three of us would turn to Jessica and ask "What does the Social Radar say?" [1]

The footnote for that quote says:

>...

>"She was always good at sniffing out any red flags about the team or their determination and disarmingly asking the right question, which usually revealed more than the founders realized."

I'd be interested in reading or knowing more about the questions which were asked, instead of a direct appeal to authority (right question), what the questions revealed, and what the founders actually realized.

As far as I can tell, these details never get discussed publicly.

Imagine if Jessica ran a VC firm with two other people like her, plus Paul, and they were trying to fund companies that required immense social skills. The founders would all be excellent conversationalists and promoters, and could probably fake the tech talk enough that Jessica couldn't tell. PG would be able to spot the buzzwords and call them on it, and it would seem magical to the rest because they don't know SQL from C++.

I try not to believe in magic as well, so I assume there are reasons why a particular founder was judged unsuitable. It's just that those reasons are not explored and all the language used to described the YC process is essentially an appeal to a magic oracle.

>Imagine if Jessica ran a VC firm with two other people like her, plus Paul, and they were trying to fund companies that required immense social skills.

I've never been inside a VC firm, so it's difficult for me to imagine precisely what that would be like. Are you saying that the scenario in this hypothetical is different from YC?

>The founders would all be excellent conversationalists and promoters, and could probably fake the tech talk enough that Jessica couldn't tell. PG would be able to spot the buzzwords and call them on it, and it would seem magical to the rest because they don't know SQL from C++.

It sounds like a decent strategy. Certainly not foolproof, but at least you can filter out the people who understand the language, but not yet the technicals...I think.

My grandfather was a pub owner -- he could talk to anyone about anything, and was often able to de-escalate people and situations. I remember him being a fluent reader of body language... he could size somebody up and immediately & accurately tell you that the person was trouble.

Which says, to me, that pg doesn't understand it, either. He just trusts it based on historical reliability.

I'd also be curious to know what some of those magically insightful questions were, but I don't think it nullifies the point of the article to not include them.

She is not a silent Oracle, she just isn't particularly outgoing when it comes to press and such, and so she is not seen when looking in from the outside. She is extremely friendly, entirely sincere, and not at all judgmental (unless, of course, there is something negative to judge...character of founders does matter, and if pg has noticed Jessica has better skills in judging character it would be silly to ignore her opinion). pg and the boys seem imminently capable of spotting technical fakery, but maybe are weaker at spotting character fakery. I dunno. They've obviously had great success with the way they operate. I would be unwilling to bet against them, and I know they do it while adhering to pretty high ethical standards.

So, call it a cult if you like, but if I had my druthers, more of the industry would imitate YC.

However, your claim nearly directly contradicts this essay:

>PG: One of the things she's best at is judging people.

>You: She is ... not at all judgmental ([exception])

So, yes, we have this exception where the key value is all in the judgement of individuals (I'll put aside the seeming contradictions in this line of thought for now), but the basis of the judgement is never given. Instead all the descriptions just refer to generic "negative character attributes" (my own paraphrasing). There is never a description of what a negative character attribute is (other than the very vague, "faker"), how it was determined that a person possesses such a character, nor how the decisions are followed up to see if the judgments were accurate.

The essays also make it sound like Jessica's feedback is never given to those who are interviewed. From the descriptions, it sounds like Jessica either gives you a thumbs up or a thumbs down. If you get a thumbs down you just get the boot, but likely no feedback on how your character was judged.

>call it a cult if you like

I didn't call YC a cult. I said that some of the descriptions given in this essay, and others, are vaguely cult-like.

Being judgmental is usually meant as someone who is quick to criticize even mundane things and who focuses on the negative. Its a filter that ignores the positive and amplifies the negative - someone who blows things out of proportion. It often is driven by a desire to feel superior to another and comes from a place of insecurity.

Being a good judge of someone is being observant and weighing the complexities of what someone presents of themselves. Think lady justice with the scales - taking a whole picture before making judgment, and avoiding bringing in one's own motivations, emotions, or concerns in the process. Lady justice's blindfold is the symbol for this objectivity.

As you can see they're totally different in how they filter data, when they make judgment, and their relationship to the other.

You're right, the criteria used doesn't seem to be visible. While feedback on particular aspects of an interview NOT related to social cues would be helpful, feedback for social cues & character is just not something anyone owes you - no matter how emotionally difficult an interaction may be to participate in.

There are likely ways you can get feedback on these kinds of things - friends or coaches - but ultimately its not the kind of thing that one studies directly for. If it were, everyone would come in an actor, rather than themselves (and lets face it, the psychopaths would be the winners). You can try all you want to study how to project the right microexpressions and whatever but that's endlessly complex and you'll lack sufficient data to optimize, not least of which is because every person and situation is unique. This may go against some notions of fairness but lack of visibility and feedback is just part of the domain of social interactions.

Fundamentally the character & social cues you give off are an output of what you do with other inputs - how you view and treat people, what you're motivated by, the time you take to strengthen your own observational skills (such as through mindfulness practice) and how well you know yourself, etc. If you're concerned about giving off good social cues, you'll get a heck of a lot farther focusing on building an awareness of whats going on in your mind, in the conversation, and with the other person that is triggering those cues. That's the best way to be genuine, and in turn the best way to come across as genuine.

Isn't this basically the feature that is being claimed here? PG relates that even when the other partners are all go, Jessica has an amplifier that finds the negatives and blows it up? The only thing left to determine would be whether the proportion was correct. Unfortunately, the aftermath of the explosions aren't really discussed in these essays, so we can only speculate.

>Think lady justice with the scales - taking a whole picture before making judgment, and avoiding bringing in one's own motivations, emotions, or concerns in the process. Lady justice's blindfold is the symbol for this objectivity.

It sounds fine in theory, but unless this lady can put down the logical argument that connects the premises to the conclusions, and which can be argued against, I wouldn't want such a lady deciding justice for anyone.

>Lady justice's blindfold is the symbol for this objectivity.

I think any time objectivity is being symbolized, you need to be very, very careful. As far as we can tell, true objectivity cannot be attained. Fetishizing is dangerous.

>There are likely ways you can get feedback on these kinds of things - friends or coaches - but ultimately its not the kind of thing that one studies directly for. If it were, everyone would come in an actor, rather than themselves (and lets face it, the psychopaths would be the winners). You can try all you want to study how to project the right microexpressions and whatever but that's endlessly complex and you'll lack sufficient data to optimize, not least of which is because every person and situation is unique. This may go against some notions of fairness but lack of visibility and feedback is just part of the domain of social interactions.

I agree. The problem is that this essay essentially describes YC's process as being exactly this situation. Jessica sits and analyzes the situation, then brings down judgement and it is decided. No details given (so, one must assume that the decision could be based on such things as microexpressions, though PG labels it character, and Social Radar).

>Fundamentally the character & social cues you give off

And here is the problem. People don't give off character and social cues, those are entirely constructs in the perceiving mind. Assuming that people are communicating some thing that they may not be is a recipe for problems. Relagating an oracle to interpret such communications (which may or may not exist) is something that is difficult to comprehend.

He also gave at least one example (Airbnb) where they didn't like the idea, but funded them because they liked the founders. Who do you think was the primary judge of "liking" the founders in that case?

> unless this lady can put down the logical argument that connects the premises to the conclusions, and which can be argued against, I wouldn't want such a lady deciding justice for anyone.

What if the logical argument doesn't exist? You seem to be ignoring that possibility completely. We have to make choices all the time with insufficient information; in fact the situations in which we actually can articulate a logical argument for doing or not doing something are rare. The fact that you appear to be very uncomfortable with this does not make it false.

> People don't give off character and social cues, those are entirely constructs in the perceiving mind.

"Character and social cues" just means "information about what the person will do in situations other than the one they're currently in." All of us do give off this information, whether we like it or not. Everything you say and do is information about the internal processes that determine what you say and do, and therefore is information about what those internal processes will output in other circumstances. It's certainly not complete information, but complete information is unattainable anyway.

> Assuming that people are communicating some thing that they may not be is a recipe for problems.

People aren't consciously communicating character and social cues; in fact they might be consciously trying to hide them. That doesn't mean they can't be valid information.

This only really shows that Jessica (assuming your implication of who the primary judge was in that instance) may also advocate for particular founders, not just veto some.

>What if the logical argument doesn't exist?

Then lady justice should not make a decision.

>"Character and social cues" just means "information about what the person will do in situations other than the one they're currently in."

The problem is that this is nearly equivalent to tea reading. What theory allows you to accurately predict what I will do, purely based on your observation of me, or even under cursory verbal examination? I don't think that there is one. That people in this thread believe that there could be one is strange.

Unless you are going to accept everybody or reject everybody, there are going to be valid reasons to advocate for some founders and advocate against others. I don't understand why you insist on focusing on the latter but ignore the former.

> Then lady justice should not make a decision.

So you only make decisions when you have a logical argument that justifies a particular choice? You must lead a very...interesting life.

Also, your use of the word "justice" is not, um, justified. Whether or not someone gets funded by YC is not a matter of justice. Nobody has a right to YC funding.

> What theory allows you to accurately predict what I will do, purely based on your observation of me, or even under cursory verbal examination? I don't think that there is one.

You're right; there isn't one. So what? Do you have a theory that allows you to accurately predict how a food you've never tasted before will taste to you? How a color or material you've never seen before will look to you? How an experience you've never had before will feel? Yet somehow you manage to taste new foods, see new colors and materials, and have new experiences.

You don't need a theory of how something works in order to do it. People accurately threw spears long, long before anyone discovered Newtonian physics. People ate nutritious food and got energy from it long, long before anyone had a theory of how metabolism works. And people make accurate judgments about other people even though nobody has any very good theories of how the human mind works.

> That people in this thread believe that there could be one is strange.

That you believe that you need logical arguments and theories in order to do anything at all is strange.

I don't think I have focused on either. I'm primarily just responding to the points others are bringing up. I made my main points in my OP's.

>So you only make decisions when you have a logical argument that justifies a particular choice? You must lead a very...interesting life. >Also, your use of the word "justice" is not, um, justified. Whether or not someone gets funded by YC is not a matter of justice. Nobody has a right to YC funding.

It's not my use of justice. It is the word the other person in this thread brought in. If you read the thread, you will see that they were making an analogy of this situation to "lady justice". I agree that it is a rather inept analogy.

>People accurately threw spears long, long before anyone discovered Newtonian physics.

And sometimes they failed. The essay treats Jessica as fail-proof.

>That you believe that you need logical arguments and theories in order to do anything at all is strange.

Complete strawman. Show me where I claimed, or even implied as such.

If you wish to convict someone a la the lady of justice, then yes, I do believe you need logical arguments and theories before you can do anything (in regards to punishment).

I don't see that at all. It says she contributed something important to the YC evaluation process that none of the other founders could. It doesn't say she never made any mistakes. Nor does it say that her input was the determining factor in every choice. I think you are reading things into the article that aren't there.

> It's not my use of justice.

You didn't use the word first, but you are treating YC's process as though the word was appropriate. See below.

> If you wish to convict someone a la the lady of justice, then yes, I do believe you need logical arguments and theories before you can do anything (in regards to punishment).

But if you believe the justice analogy is "inept" (your word), why would you buy into someone else's interpretation of YC rejecting an applicant as "punishment"? It's not. As I said before, nobody has a right to YC funding. Their money, their choice. They don't have to give a logical reason; they don't have to give a reason at all. They could choose among applicants by throwing darts, and nobody would have any right to complain.

Of course, choosing by throwing darts wouldn't work well, which is why YC doesn't do it. But what they do do, including the "Social Radar", does appear to work well, even if no one can construct a logical argument for why it does. Since it's their money, and it's working well for them, nobody else has any right to complain that they're being "punished" if YC rejects them. They certainly don't have a right to do so on the basis that there isn't a logical argument backing up YC's choices.

Who is complaining? Obviously YC can do whatever with its money. It doesn't even need to be said, let alone, twice.

>Of course, choosing by throwing darts wouldn't work well, which is why YC doesn't do it. But what they do do, including the "Social Radar", does appear to work well, even if no one can construct a logical argument for why it does.

Right, so as far as anyone on the outside can tell, YC operates on a type of magic oracle.

>You didn't use the word first, but you are treating YC's process as though the word was appropriate.

My only point here is that the essays describe YC as essentially a magic process. The person I was responding to tried to make an analogy to the "lady of justice", while simultaneously making a claim to objectivity (in addition to others). My sentences with containing or referring to "lady" (including my comment about punishment) were only in response to that person's claims. Those sentences contain no information about what I think about YC's process (or rather, it's description in this essay).

Best example https://www.youtube.com/watch?v=yG4DvM0wxdk

So, maybe the scene above will illustrate the difference between cult-like leadership and leadership that's more like a family. If anything, the essay just describes a down-to-earth, genuine, tightly-knit approach to business and evaluating business partners. There's many companies like that albeit not as famous as YC. You don't see people suing them, writing exposes, and alleging all kinds of abuse like with the cult-like companies. Huge difference.

I agree, so let me try to rephrase for you. For a bunch of people (YC et al) who actively promote making decisions with data, resorting to some ill-defined intangible such as "Social Radar" as being the secret sauce for success just seems noticeably contrived.