ckelly

Perhaps. But what defines working hours?

This article is arguing for working hours that equate to about 65% of waking hours. So your range seems too high.

"Housing as a vehicle for building wealth and housing becoming unaffordable for the younger generation are two sides of the same coin." That's a great insight.

It has always surprised me that many technology professionals (and business professionals in general) don't have a strong intuition for the power of sampling. For example, in this case, the author states: "With 100 samples, our estimates are accurate to within about 5%. The magic of sampling is that we can derive accurate estimates about a very large population using a relatively small number of samples. In the last scenario (100 billion M&MS), we have 1% accuracy despite only sampling 0.00001% of the M&Ms."

I bet many would think n=100 would be worthless once the population reaches millions, or especially billions.

One HN-related piece of evidence for that is when I pointed out what margin of error would be for a n=164 survey sample, I got downvoted hard! https://news.ycombinator.com/item?id=8050801

But I saw this hundreds of times talking to customers when I ran a survey sampling product out of YC.

Yes, I’ve heard Michael Bloomberg say the secret weapon was adding chat, which built the network effect.

Yes, I wasn't commenting on the original "taking it seriously" language.

If the market price reflected the probability, then an arbitrage strategy should not be profitable The market doesn't reflect the probability of an event happening.

No, the market's implied probability could be right, on average, across all deals...and the top merger arb funds could absolutely still be profitable by selecting deals when they think the market is mispricing the probability (for the reasons you mention: better experience, knowledge, etc.)

It's like the sports betting market: you can roughly impute a team's win probability from the (opening) betting line...and even if that's right on average, the top gamblers are still profitable.

And, of course, sometimes things with a say, 40% chance of happening do happen...so that doesn't mean the market was "wrong" about the chance (i.e. your LinkedIn mispricing exmaple).

But sounds like we're in full agreement you can't look at the implied probability from the market price and draw some conclusion about it definitely happening, or definitely not happening (e.g. the market not taking it seriously).

I wouldn't put too much weight into any sort of imputed probability from the price.

It's absolutely fair to impute a rough probability of deal closure from the stock price. The whole "merger arbitrage" industry works around that premise.

Sometimes the market doesn't think a deal has a 100% chance of closing (like MSFT and LinkedIn) and it still closes. There were valid antitrust concerns circling that deal, e.g. https://thehill.com/policy/technology/298573-salesforce-rais...

The answer is less attribution measurement, more incrementality measurement. Incrementality solves the famous "handing out coupons outside the pizzeria door" problem: https://www.adweek.com/programmatic/lower-ad-fraud-will-be-a...

A root cause of this is an overreliance on Multi-touch Attribution (MTA) models, instead of true incrementality experiments: https://www.adweek.com/programmatic/lower-ad-fraud-will-be-a...

Goolge had a similar product called Google Website Optimizer that it shut down years ago:

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

https://support.google.com/analytics/answer/2661700?hl=en

Some screenshots still floating around:

http://imgur.com/a/eReWQ

http://imgur.com/a/7XydI

tl;dr Yes

This is fantastic news for founders. Survata (S12) was pumped to have Initialized in our Series A last year. Garry worked the closest with us of all YC partners, and Alexis had already been a customer! They make valuable customer intros, always offer time to help, and have such a pro-founder view of the world.

In April of 2005, they tested their first upload. By October, they had posted their first one million-view hit: Brazilian soccer phenom Ronaldinho trying out a pair of gold cleats. Weeks later, Google paid an unprecedented $1.65 billion to buy the site.

This article misstates when Google acquired YouTube. It was October 2006, not October 2005: https://en.wikipedia.org/wiki/YouTube#Company_history

Garry has been incredibly impactful to us, even after we graduated YC. Can't wait to see what's next!

Enjoy your time abroad, Garry.

Correct. I had tried to clarify the margin of error calculation wasn't relevant here by saying "Even if this had been a quantitative study"...sorry if that wasn't clear.

Hi Max, Survata co-founder here. 164 respondents is actually a fine sample size for a quick read on qualitative feedback. Even if this had been a quantitative study, that would represent a 7.7% margin of error at a 95% confidence level. Counter-intuitively, you don't always need thousands of respondents to measure an audience's behavior!

I want a content marketing platform that let's you easily turn blog posts into other units of content (videos, infographics, slideshows, whitepapers) without going to 9 different systems.

Hi Nathan, Thanks for your comment. And no offense taken. :)

Most people are surprised how "few" respondents it takes to get to 5% margin of error at 95% confidence levels: https://en.wikipedia.org/wiki/Margin_of_error#Different_conf...

In this case, .98/sqrt(365) = 5.1%

So, perhaps counterintuitively, 300-400 respondents gives good read on large populations (like Snapchat users...or the US population!).

Hi Max, Survata co-founder here. The 365 respondents were used as the denominator for calculating the % of teens that use a service (like Snapchat) regularly. So that's the sample size for the first chart shown in the link. (The margin of error on 365 respondents is about 5%). Naturally, we had to exclude the respondents who used no social services from the questions that asked about their favorite social service, like the second chart shown in the link.

Hope that helps.

Hi - I'm a Survata co-founder. Thanks for catching that - it was an error we made in producing the chart, not a problem with the underlying study. The image has been updated. Sorry for the confusion!

Survata co-founder here. The first graph does not include those respondents. As the footnote says, "Respondents who answered that they 'never' use public transportation were excluded."

Survata co-founder here. You're right, and we mentioned that in the article:

"One obvious caveat is our survey coincided with the BART strike in the Bay Area, and might have reached Bay Area public transportation riders at their most frustrated."

Survata (YC S12) powers surveywalls on publishers like magazines and news sites around the web.

Edit: For disclosure - I'm a Survata co-founder

I'm a co-founder of Survata (YC S12) which also runs consumer surveys. We're lower cost than Google Consumer Surveys and give professional survey design advice for free. Email us at contact at survata and we'll hook you up with a Hacker News discount. :)

YC-backed Survata (another consumer survey service) released a logo testing tool last month: http://survata.com/logo-surveys

Startups considering new logo options should check it out!

Disclosure: I'm a Survata co-founder

We're back. Sorry for the downtime.

Survata co-founder here. Yes, we unfortunately went down right after posting. We're working to get back online. Sorry for the hassle.

Quite addictive. Fun game.

Survata (YC S12) is considered a competitor to Google Consumer Survey's core product. I agree this new product looks more like Qualaroo, though.

(I'm a Survata co-founder)

Hi Larry - I'm a Survata co-founder. Our system screens out respondents who answer too quickly, answer in a suspicious pattern, fail "trap" questions, etc. So individuals not taking the survey seriously do not impact our data collection.