21

Donor lottery details

On January 15 we will have the drawing for the donor lottery discussed here. The opportunity to participate has passed; this post just lays out the details and final allocation of lottery numbers. If you regret missing out, I expect there will be another round, and it would be useful to know that you are interested.

There were 18 participants who contributed a total of $45,650. We will take the first 10 random hexadecimal digits from the NIST randomness beacon at 12pm PST on January 15 and interpret them as a random integer between 0 and 16^10-1. The interval [0, 16^10-1] has been allocated amongst the 18 participants in proportion to their contribution, as indicated in the table below. The random number will fall into the [Low #, High #] range of exactly one participant, who is the winner.

I will set aside $45,650 from my DAF, to be granted at the winner's discretion at any time. They can also choose how that money should be invested in the meantime.

We originally stated that the prize pool would be $100,000, but have decided to adjust it to $45,650, guaranteeing that there will be a winner and reducing my personal risk to zero. The winner is welcome to take a double-or-nothing bet in order to get up to $100,000 if they prefer the larger scale (and can probably find a way to gamble to even larger amounts if they want to).

Because I no longer bear any risk, I am not going to charge a 1% fee (which was my original plan). Organizing and thinking about the lottery still took 3-4 hours of my time, but I think that I can offer lotteries with minimal labor in the future, and I am happy to put a little volunteer time into making the first one happen. (Some other donor may be a more natural provider over the long run though.) 

Contributor Amount ($) Low # (in decimal) High # (in decimal) Probability
Timothy Telleen-Lawton 5050 0 121632721144.5 11%
Gregory Lewis 5000 121632721144.5 242061157921.5 11%
Ajeya Cotra 2200 242061157921.5 295049670103.5 5%
Rohin Shah 2800 295049670103.5 362489594699.5 6%
Helen Toner 2500 362489594699.5 422703813087.5 5%
Nicole Ross 500 422703813087.5 434746656765.5 1%
Howie Lempel 5000 434746656765.5 555175093542.5 11%
Rebecca Raible 2000 555175093542.5 603346468253.5 4%
Pablo Stafforini 2000 603346468253.5 651517842964.5 4%
Aaron Gertler 500 651517842964.5 663560686641.5 1%
Brayden McLean 5000 663560686641.5 783989123418.5 11%
Benjamin Hoffman 100 783989123418.5 786397692154.5 0.2%
Catherine Olsson 500 786397692154.5 798440535832.5 1%
Eric Herboso 500 798440535832.5 810483379509.5 1%
Ian David Moss 2500 810483379509.5 870697597898.5 5%
Glenn Willen 500 870697597898.5 882740441576.5 1%
Jacob Steinhardt 4000 882740441576.5 979083190997.5 9%
Brandon Reinhart 5000 979083190997.5 1099511627775 11%

Lessons

Donating to popular charities is a lot easier than contributing to a DAF; future lotteries should probably be implemented as donation swaps. For example, if I wanted to make a $100k contribution to MIRI, then participants could donate $X to MIRI and tell me to reduce my donation by $X. This makes participating in the lottery roughly as easy as donating to MIRI, which has good payment infrastructure. I think donation swaps are also useful when employers offer donation matching, though donation matching didn't come up this year. (I think matching lottery entries is compatible with the spirit of employer donation matching.)

We got more participation than I initially expected. Some of the participation was based on the novelty of the idea, but nevertheless I expect there will be a larger lottery next year. That should also be helped by a smoother user experience---no $5k minimum, implemented as donation swapping so very easy, and accompanied by an upfront explanation of how to participate.

Now that the drawing is going to happen, I do expect the lottery winner to make a materially better decision (in expectation) than they would have made otherwise. Moreover, I think the existence of the lottery was bottlenecked on the kind of work that Carl did in advocating for the idea and contacting possible providers (rather than on the existence of customers). So I've increased my estimate for the value of entrepreneurial spirit in the EA community.

Sanity check

3c0ade0f0490dff240b5b4a97c522c14cfd1490a2d40b4dddc535e0bd238c6fb

This is the SHA-256 hash of the first section of the post (original text here, I've edited it since then but not changed the substance of the agreement), which I will post on twitter and other people are free to store for their records. Hopefully this is a cheap measure which can make it difficult to manipulate the terms of the auction after the random number is revealed.

I've emailed these hopefully-final terms to participants and no one has objected so far. If there are any last-minute revisions, then we will hopefully have time to get things in order prior to January 15. I will tweet the updated SHA-256 hash at that time.

Comments (2)

Comment author: Paul_Christiano 25 March 2017 11:03:36PM 3 points [-]

I owe Michael Nielsen $60k to donate as he pleases if [beacon.nist.gov](beacon.nist.gov/home] is between 0000000000... and 028F5C28F5... at noon PST on 2017/4/2.

Comment author: Carl_Shulman 15 January 2017 08:09:04PM *  11 points [-]

Looks like Tim Telleen-Lawton won, as the first ten digits of the beacon at noon PST were 0CF7565C0F=55689239567. Congratulations to Tim, and to all of the early adopters.