I am a statistician and I buy lottery tickets • Hacker News Board

Re: I am a statistician and I buy lottery tickets

#91

Post by epo » Sat, Oct 20, 2012, 6:54 PM UTC

jimrandomh wrote: post #80

Interesting that he claims this lottery has positive expected value, but I don't think it's true. He forgot about income tax. This is one of the standard tricks for making lotteries look better (or less bad) than they really are. (The other common trick is paying an annuity instead of money, and advertising its nominal value instead of its actual value, which is substantially greater. This one doesn't seem to be doin…

In the UK, lottery winnings are tax-free and paid in a lump sum. It may be the same in Australia.

Re: I am a statistician and I buy lottery tickets

#92

Post by epo » Sat, Oct 20, 2012, 7:04 PM UTC

wilfra wrote: post #30

MichaelApproved wrote: post #9

Earlier quoted context omitted.

He doesn't think imagining winning the lottery is fun, it actually is fun for him. He's actually entertained and is comfortable spending money for a little imaging and escapism. What's wrong with that?

Nothing is wrong with that. When I go to Las Vegas, I play craps for the same reason. But I didn't make a massive blog post that implied I could beat craps, get it on page 1 of HN and after 20,000 words and numerous graphs and tables announce to the World that it's ok that I donate some money to Bellagio once a year because it's my money and the waitresses are hot and it's fun. I wouldn't do that because no amount of…

"But I didn't make a massive blog post that implied I could beat craps ..." He implied nothing of the sort, the headline is a good summary of the post. You are annoyed that the post is not what YOU thought it should have been, learn to read.

Re: I am a statistician and I buy lottery tickets

#93

Post by ced » Sat, Oct 20, 2012, 7:12 PM UTC

emmett wrote: post #82

ced wrote: post #78

Earlier quoted context omitted.

Evolution is often all-or-nothing --- no one produces "half a child". This is especially true of men, if you believe the 40-80 figure [1]: 40% of men have descendants vs 80% for women. We are the descendants of A) successful people [2] and B) unsuccessful people who took a big gamble (winning the lottery; robbing a bank; crossing the ocean to find better opportunities) that made them wealthy enough to reproduce. [1]…

Actually people have "half a child" all the time. Every time a sibling reproduces, for example, that's half a child. So there's a perfectly viable strategy of helping your siblings have more children instead of having more children yourself. If you can create more than 2 successful sibling-children for every child you don't have, your inclusive genetic fitness is increased. This doesn't mean your conclusion (people g…

Good point. Evolution is a complex balance.

Re: I am a statistician and I buy lottery tickets

#94

Post by Evbn » Sat, Oct 20, 2012, 7:36 PM UTC

muppetman wrote: post #20

brianwillis wrote: post #11

Looking back, there seems to be a correlation between the times in my life when I've bought lottery tickets, and the times in my life when I've felt hopeless. If I ever feel trapped by circumstances, depressed, lost, or I don't have optimism for the future, the less logical part of my brain will somehow rationalise the remote chance of winning millions as a perfectly justified excuse for buying a ticket. The author o…

That's a very interesting view point/experience. For me, it's been the opposite. If I'm really happy, feel a bit like I'm on top of my fiances for this fortnight, I'll "splash out" and buy a Powerball ticket. If I'm down, the idea of giving away money on such a silly endeavour seems stupid. Anyway, here's hoping you don't buy too many powerball tickets in the immediate future!

Conspicuous consumption.

Re: I am a statistician and I buy lottery tickets

#95

Post by Evbn » Sat, Oct 20, 2012, 7:40 PM UTC

aes256 wrote: post #44

brazzy wrote: post #23

Earlier quoted context omitted.

The difference is that insurance is not about the expected return at all - it's about managing risk. The "payout" from insurance is tied to events that can happen whether or not you have insurance , and which would ruin you financially without it. The expected payout from insurance is only relevant when the above does not hold, i.e. you have enough liquid funds to cover the loss against which you're insured. A good e…

Indeed. Also, most people are risk averse, so they are happy to purchase insurance with an EV (expected value) equal to less than the premium simply to avoid the gamble.

Not indeed. Variance matters. Look at the hedge fund industry.

Expected return is insufficient if you can't recover from death to regress to the mean. So yeah, zombies shouldn't buy insurance

Re: I am a statistician and I buy lottery tickets

#96

Post by Evbn » Sat, Oct 20, 2012, 7:46 PM UTC

JoeAltmaier wrote: post #72

graeme wrote: post #70

Earlier quoted context omitted.

The cumulative impact of a lifetime of spending $5 per week is much less than winning a lottery. Most people can sustain a $5 per week hit, especially if they get the mental benefits described by the parent article. There are better things to spend $5 on, but it's not a huge deal either way - as long as you stay out of the addictive/compulsive zone. The losses covered by insurance are very impactful. Most people can'…

Hm. Most people collect on insurance for very-sustainable costs - drug benefits, dental for checkups, routine visits to the doctor. Where do those fit into this argument? Personally, I would like insurance that pays only for large expenses. I know its a sucker game to pay insurance; since the insurance companies profit I must be losing. But 'group insurance' is not something I can change.

When you buy catastrophic insurance it usually comes with free preventative care, which benefits the insurer as well as the insured.

You can often get a lower premium for a higher deductible, and everyone who knows how to multiply does this.

Re: I am a statistician and I buy lottery tickets

#97

Post by Evbn » Sat, Oct 20, 2012, 7:47 PM UTC

lylepstein wrote: post #21

paulsutter wrote: post #16

Insurance has an interesting comparison to lottery tickets. In a lottery ticket, the expected return is usually about 50%. Which is exactly the same as insurance (the premium is about double the expected payout). Yet people regard insurance as prudent, and lottery tickets as foolish. Me included (as long as its a risk you cant easily cover). But our explanation, that "Lottery tickets are foolish because they have a n…

I think you can classify lottery tickets as investment, wheras insurance is, well, insurance. I don't play the lottery because it's extremely unlikely to make me positive returns. I pay for health insurance because I want to cover the unlikely possibility that I have a serious health problem. There's also the amortization to consider. Over a lifetime I may lose 50% of what I put in to insurance -- but I don't have 50…

"Investment" is traditionally reserved for activity wth positive expected returns compared to baseline behavior.

Re: I am a statistician and I buy lottery tickets

#98

Post by Evbn » Sat, Oct 20, 2012, 7:49 PM UTC

pbhjpbhj wrote: post #59

mrb wrote: post #26

Earlier quoted context omitted.

It does mention why he buys tickets: not for the expected return, but for the pleasure of entertainment.

Well it mentions why he 'says' he buys tickets. > The standard argument of people who think that lotto is for suckers is based on expected return. They are taught at school that a rational investment is one with an expected value greater than the price paid. I have some problems with this argument, but I’ve never really calculated it before, so let us consider the numbers. // This para to me says he's not really look…

You seem to be projecting.

Re: I am a statistician and I buy lottery tickets

#99

Post by Evbn » Sat, Oct 20, 2012, 7:50 PM UTC

jimrandomh wrote: post #80

Interesting that he claims this lottery has positive expected value, but I don't think it's true. He forgot about income tax. This is one of the standard tricks for making lotteries look better (or less bad) than they really are. (The other common trick is paying an annuity instead of money, and advertising its nominal value instead of its actual value, which is substantially greater. This one doesn't seem to be doin…

If playing lottery is worthwhile fun even when you lose, taxes don't matter.

Re: I am a statistician and I buy lottery tickets

#100

Post by Evbn » Sat, Oct 20, 2012, 7:55 PM UTC

wilfra wrote: post #37

staunch wrote: post #7

There is that Stanford math PhD that won four lotteries. http://www.forbes.com/sites/kiriblakeley/2011/07/21/meet-the...

That's really interesting. Seems pretty obvious she cheated in some way. If something only happens once every 'quadrillion years' that's pretty much evidence right there, but given all of the other suspicious things as well...ya - I think there is basically no chance those were legit wins.

There is a similar story of a Canadian or somewhere that cracked a weakness in the mechanics of a scratch off game (Monty Hall paradox exploited in real life!) for consistent average winnings.

And of course the famous MIT Blackjack team.

Of course, a lottery loves publicity, and they don't care which player wins, so win-win.