Subject:
|
Re: Preposterously simple stat question
|
Newsgroups:
|
lugnet.off-topic.geek
|
Date:
|
Tue, 21 Jun 2005 16:11:35 GMT
|
Viewed:
|
2072 times
|
| |
|
|
In lugnet.off-topic.geek, Dave Schuler wrote:
> Assume a weekly lottery that draws six numbers from the series 1 through 40,
> with no duplicates. Statistically speaking, am I better off repeatedly playing
> the same six numbers selecting six numbers at random each week?
Statistically, assuming it's a fair draw, there should be no difference, as this
is a draw with replacement (each time you start drawing 6 numbers, all 40 are
there) at the start of the draw, between any 6 numbers.
Non statistically, that is, taking psychology of yourself and other players into
account, if we assume it's a fair draw, then you would want to pick numbers that
OTHER people are less likely to pick, (since the prize is shared out) to
increase the prize value. But that's pretty imponderable unless you know what
gets picked a lot and think it's a trend likely to continue.
Also, it may be more convenient for you (and thus of slight economic positive
value) to pick teh same numbers each time to reduce the time you spend on
playing... OR, it may increase your pleasure in playing by thinking about which
numbers to pick each time suggesting that you should pick different ones
If it's an unfair draw, your best bet, of course, is to pick the numbers more
likely to be drawn, but you'd need to know how exactly it was rigged. Most mobs,
er, state governments, don't share that info out.
> It seems to me that if (in a different example) you randomly pick a number
> between 1 and 10 and I have to guess it, my chances of getting it right are one
> in ten regardless of whether I stick with the same choice or re-shuffle it each
> time (assuming that you likewise randomly re-select your number each time). Is
> this correct?
Yes.
> And does it also apply to a 40-digit lottery pool?
Yes, if it's a fair draw.
> I’m always embarrassed that I don’t know this stuff, but I never had a course in
> statistics, and I tend to second-guess my reasoning.
Don't let that stop you from playing the lottery, it makes the prizes bigger for
me when I play... (which is only when the expected present value is larger than
the bet)
|
|
Message has 2 Replies:
 | | Re: Preposterously simple stat question
|
| (...) Often, I'm told, birthdays and anniversaries are the source of a lot of lotto-style picks. If so, numbers 1-12 are very common, and 13-31 are more common. 32-40, not being easy date numbers, are thus more rare. (19 years ago, 22-Jun-05, to lugnet.off-topic.geek)
|
Message is in Reply To:
| | Preposterously simple stat question
|
| Assume a weekly lottery that draws six numbers from the series 1 through 40, with no duplicates. Statistically speaking, am I better off repeatedly playing the same six numbers selecting six numbers at random each week? It seems to me that if (in a (...) (19 years ago, 21-Jun-05, to lugnet.off-topic.geek)
|
6 Messages in This Thread:
 
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
This Message and its Replies on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
