Who audits the advertised odds on lotteries

September 20, 2007 1:30 PM

One in three chance you'll win a free pop, check under the cap! Does anyone audit these advertised odds?

I almost never win these things. I figured with one in three odds , maybe this time... but no. Then I started wondering - maybe they're cheating.

I'm imagining some kind of sweepstakes control board, randomly buying bottles of Dr. Pepper at convenience stores across the country to do samplings of who wins... crack statisticians ensuring that winners are evenly distributed across regions... yeah right. So how do I know they're not cheating?

posted by PercussivePaul to law & government (6 comments total)

Gaming Control Boards. When my old company wanted to run a trade show contest where you rolled dice to spell out our firm's name and could win a Porsche, we had to make sure gaming officials were present.
posted by acoutu at 3:07 PM on September 20, 2007

So how do I know they're not cheating?

Basically, you're asking, what's the chance that you'll buy x bottles of soda and not win anything.

This sort of thing can be modeled with a Binomial distribution.

You'd expect to get either 3 or 4 winning sodas every 10 bottles (your mean), plus or minus 2 bottles (your variance).

On the whole, every ten bottles, you'd want to win between one and six times. This should happen approximately 96% of the time you do a 10-bottle sample.

On the other hand, not winning at all should only happen 1.6% of the time you do a 10-bottle sample.

So the statisticians check whether random samples of product are wins, counting them and seeing if their ten-bottle samples deviate from this one to six bottle range.

In reality, they count more than ten bottles, but you don't need a whole lot to do testing.

If there's a lot of deviation from this range — let's say that you don't get any wins at all about 5% of the time — this indicates that the game is rigged or broken in some way, and would presumably raise a red flag with auditors.

Assuming the auditors are honest, other than a customer being able to file a fraud lawsuit, I'm not sure what sanctions those auditors can impose on the company that hires them.

In the case that the auditors are crooked, who audits the auditors? Good question. Perhaps a company would hire two independent auditors and look more closely if their results differ.
posted by Blazecock Pileon at 3:15 PM on September 20, 2007

> I almost never win these things. I figured with one in three odds , maybe this time... but no. Then I started wondering - maybe they're cheating.

The answer might be simpler than you thought. I had a friend who could tilt the bottle just so and discern whether it was a winner. He could buy just one and essentially drink free sodas throughout the entire promotion. Your odds might be a lot lower by the time you're picking a bottle from the shelf, and it's not because the companies are cheating.
posted by churl at 5:15 PM on September 20, 2007

I won 4 in the past 6 weeks where the odds said "1 in 2." I was lucky and didn't cheat... I won 4 times in 5 buys.

Churl... I don't understand how he could drink free pop... the caps almost always say "buy one get one free." He'd be getting half his pop free, sure.
posted by IndigoRain at 7:12 PM on September 20, 2007

IndigoRain: most of the caps used to be of the "redeem this for one free soda" variety. That's how it was when I was in high school in the late 90s, anyway, and I myself used the strategy churl described.
posted by Riki tiki at 2:53 AM on September 21, 2007

Yeah, the specific promotion I remember was 1 in 6 caps would win a free soda. This was 5-8 years ago. I don't drink as much soda now, so I haven't seen the 1 in 2 caps winning "buy one, get one free," -- that would obviously change things for my friend (but not necessarily improve the odds for the asker).
posted by churl at 5:46 PM on September 21, 2007