Hacker News Reader

Comment by velodrome | Hacker News Reader

I calculated the average rate after current dropoff. The funding will fail and the total will be roughly around $20 million dollars. Still, that is quite a bit of money and at least some sort of validation.

They should have kept it at $600 and increased the total needed for the project.

Some quick calculations:

($3476017 - $3180762) / 12 hours = $24605 / hour

$3180762 + $24605 / hour * (30 days * 24 hours / day) = $20,896,062

View on HN

You do realize that this makes every statistician cringe? Your sample size for calculation is too small to base any kind of reliable projections on it making this a pure guess.

By my calculations from this first month, we can reasonably conclude that the subject will give birth to nine babies in eight months.

Of course. There is so little data to work with many assumptions. This is why I said it was a quick calculation.

If you can do better, please share :P Unfortunately, I just don't have the time.

Besides, the only conclusion that can be drawn at this time is that this is going to be a one of the largest kickstarter-style funding efforts ever.

http://www.youtube.com/watch?v=c8Ezwg5LKd4

Yes, if we had this kind of data for several campaigns, then you could compute a (normalized) <% of goal> vs. <time>, and from that estimate the final <% of goal> for this campaign.

The problem with your original calculation, is you're assuming the rate is constant, which it may be, roughly, for day 3 through day 27, however in these campaigns there's always an uptick right before the deadline. Really, it's that uptick that you want to estimate, based on previous campaigns.

I don't think kickstarter or indigogo release this kind of data though, unfortunately.

There's a graph in this blog post showing the phenomenon you describe:

http://www.kickstarter.com/blog/shortening-the-maximum-proje...

Your right, I did assume the rate would be constant going forward. However, I assumed that the rate of change (RoC) would be the highest now than later.

This is the data from the largest kickstarter "Pebble: E-Paper Watch for iPhone and Android":

http://www.kicktraq.com/projects/597507018/pebble-e-paper-wa...

If you notice, the earliest part of the campaign has nearly the highest RoC. Towards the middle, it kind of flattens out. And, as you mentioned, towards the end (the uptick) the RoC resembles that of the beginning of the campaign.

If everything remained constant for the next 27 days, that "uptick" would have to be >$10 million dollars on that last 3 days.

Still, they will beat the largest kickstarter campaign ($10 million).

"They should have kept it at $600 and increased the total needed for the project." - The reason the curve levels off isn't (entirely) because the cost is $200 more. It's because the 'limited time' offer caused a lot of people to buy it FAST, people who may have not bought at all. Personally I did not purchase, but I was very close to almost entirely because of the "limited time" appeal that $600 offer had. If it was a permanent $600 price point I wouldn't have even considered it.

In the last 5 hours when I looked (roughly 4pm - 9pm in the eastern U.S.), they had raised $59,000. That's about $290,000 a day they are currently running at. They are going to need a very big push at the end if they stay at that rate for most of the month (since they need about $1 million a day).

This calculation also looks to begin at midnight. I don't think we can assume an even spending pattern throughout the day. You will have to take into account multiple days and weekends to test to see if there is a change in buying behavior by hour or day before you can make these assumptions.

And then there are press events most likely planned that could accelerate the buying. Or another event that could slow buying.

My numbers suggest they're getting over $96K/hour having them hit their 32M dollar goal on August 4, 2013 at just shy of 9pm.

$/Sec 26.7404

32M on 8/4/2013 21:40:03

> They should have kept it at $600 and increased the total needed for the project.

I suspect the profit margin on the $600 price was too small, hence the small number offered.

I don't think it can be that easily predicted. It all depends on Canonical's press efforts. There will also be a surge at the end of the project.

Reply on news.ycombinator.com