Hacker News Reader

Comment by jsjohnst | Hacker News Reader

> I saw a billboard for a lottery, it said "estimated lotto jackpot 55 million." See, I did not know that shit was estimated.

It’s estimated because they don’t know exactly how many tickets will be sold leading up to the next drawing. I’ve not tracked it, but the times I’ve seen, were under 1-2 percentage points. If it bothers you that they said “estimated $55M” and instead you win $54,333,187.42, is it really that bad?

View on HN

It was a joke by the inimitable Mitch Hedberg.

The correct answer would be to sell it as a $54M jackpot then. Or require the organizers to pay the difference out of pocket.

Or just label it as estimated until tech is developed to determine how many tickets will be sold in advance?

Is this an attempt at humor?

No?

If you don't know the answer, underestimate and overdeliver. Or consider whether you should be doing it in the first place.

The problem is that underestimating will cause fewer tickets to be sold.

A business being less successful if they don't misrepresent reality is perfectly fine to me.

> A business being less successful if they don't misrepresent reality is perfectly fine to me

We are talking about a state government backed lottery, not a business.

I can imagine a perverse population of customers whose elasticity function prevents the estimate from ever being correct by always buying too many tickets.

> I did not know that shit was estimated.

> If you don't know the answer, underestimate and overdeliver.

Make up your mind, is estimating bad or not? The example I gave isn’t the only possible outcome, some times it’s 1-2% more than quoted when more tickets were sold.

Err, what? I never posted that first quote?

And even then, it's perfectly consistent to say that estimating is only OK if it doesn't make you look better than reality.

Reply on news.ycombinator.com