2 ms·

In general, prizes awarded by judges in contests are not a very good predictor. Judges are usually busy people. They're not going to work that hard to figure

by pg 17y ago

In general, prizes awarded by judges in contests are not a very good predictor. Judges are usually busy people. They're not going to work that hard to figure out who's the best. They'll tend to be fooled by superficial polish. I was a judge at TechCrunch 50 this fall. A few weeks later we had YC interviews. I don't think I was any more slack than the other TC50 judges, but there was no comparison between how hard I worked to pick the best startups in the two cases. Not only was I going to have to put money into the YC startups, whoever we picked was going to become part of the YC community from then on. In fact, after having interviewed so many startups under conditions where getting the right answer really mattered, I found I had to consciously suppress my tendency to ask probing questions. Even so I think I seemed unduly harsh compared to the other judges.

jasonlbaptiste 17y ago
I watched the sessions where you judged. Some people thought you were harsh, but you were asking legitimate questions. Seatgeek was one, where you were asking about their algorithm. That's important stuff, especially when judging a startup. Some might feel startups on the stage there should not be put under the gun, but that's silly. If these are truly 50 of the best startups out there, they should be prepared for questions like this.
robg 17y ago
Aren't they two entirely different formats? In one, you're asking probing questions but it's a small group setting and with your longtime partners. In the other you're in front of an audience and judging with folks you may or may not know. You've also screened the applicants for YC for ideas you'll want to probe deeply whereas the contest seems more about general business models and fit to markets. You'll invest your money just based on one meeting whereas the contests seem more like a chance to simply get a second meeting. It's the difference between rational dissections in YC versus blink assessments in the competitions. I don't see very many parallels, actually. In general, prizes awarded by judges in contests are not a very good predictor. Do you know of specific data here or is this your general impression? Some contest winners and close losers have seemed very impressive to me. And that's not just at TC50 but also from the DFJ and MIT competitions too.
- pg 17y ago
  They're definitely different formats. That's why I toned down my questions for TC50. But it's interesting that both took about the same amount of time. That means we could have dug a lot deeper at TC50 if that had been the point of the event.
  - robg 17y ago
    You mean the actual "interview", right? As a judge were you also expected to review and filter applications?

by pg 17y ago

jasonlbaptiste 17y ago

I watched the sessions where you judged. Some people thought you were harsh, but you were asking legitimate questions. Seatgeek was one, where you were asking about their algorithm. That's important stuff, especially when judging a startup. Some might feel startups on the stage there should not be put under the gun, but that's silly. If these are truly 50 of the best startups out there, they should be prepared for questions like this.

robg 17y ago

Aren't they two entirely different formats? In one, you're asking probing questions but it's a small group setting and with your longtime partners. In the other you're in front of an audience and judging with folks you may or may not know. You've also screened the applicants for YC for ideas you'll want to probe deeply whereas the contest seems more about general business models and fit to markets. You'll invest your money just based on one meeting whereas the contests seem more like a chance to simply get a second meeting. It's the difference between rational dissections in YC versus blink assessments in the competitions. I don't see very many parallels, actually. In general, prizes awarded by judges in contests are not a very good predictor. Do you know of specific data here or is this your general impression? Some contest winners and close losers have seemed very impressive to me. And that's not just at TC50 but also from the DFJ and MIT competitions too.

pg 17y ago
They're definitely different formats. That's why I toned down my questions for TC50. But it's interesting that both took about the same amount of time. That means we could have dug a lot deeper at TC50 if that had been the point of the event.
- robg 17y ago
  You mean the actual "interview", right? As a judge were you also expected to review and filter applications?