A lot of people seem to have the intuition that you can’t, in ordinary circumstances, know that a particular lottery ticket will lose. Dana Nelkin, in a 2000 _Philosophical Review_ paper says this is because the belief you have is based solely on statistical evidence. (She says you can’t even have a justified belief to this effect.) Duncan Pritchard (in “this paper (PDF)”:http://www.philosophy.stir.ac.uk/staff/duncan-pritchard/documents/KnowledgeLuckLotteries.pdf) says that it is because the possible world in which the ticket wins is too similar to the actual world. As he says…

bq. After all, the possible world in which I win the lottery is a world just like this one, where all that need be different is that a few coloured balls fall in a slightly different configuration.

I rather doubt both of these explanations. I think the intuition that you can’t know you’ll lose is a bit of bad scepticism. To test that, I want to see how intuitions go on the following kind of lottery. This is a real-world case by the way, a bit of found philosophy. On Monday through Saturday, the lotteries in Australia are based around coloured balls falling in distinctive configurations. But on Sundays things are different, as “this site”:http://www.ozlotteries.com/play.php?lottery_id=6 explains.

bq. Sunday Lotto (or ‘Soccerpools’) is based on Australian and European soccer matches. You don’t need to know anything about Soccer though – it can be played just like a normal lottery game. Each week, 38 matches are listed and numbered 1 to 38 inclusive. For a standard lottery, you choose 6 numbers from the range of 1 to 38. The 6 matches that accumulate to the highest total drawn scores are the winning numbers (e.g. Match A with a final score of 4-4 has a higher total score than Match B which finished 3-3). The 7th highest result is the supplementary number.

Imagine I’m looking at a particular ticket, say the ticket of someone who does play the Sunday Lotto as a lottery, and I believe it won’t win. Could this be knowledge?

Pritchard says that it isn’t knowledge if there is a nearby world in which it wins. But imagine that (unbeknownst to me or the buyer) for this ticket to win requires there to be a score draw between Chelsea (at home) and a relegation threatened team. Chelsea normally win these one-sided games, and they very rarely concede a goal. So the worlds in which this ticket wins are rather remote. Is that sufficient for my knowing the ticket will lose? Do I have to believe it will lose because of these facts about Chelsea to know it will lose? Immediately following the quote above, Pritchard says

bq. Crucially, however, the _nearness_ of the relevant possible worlds has an impact on our judgements about the presence of luck.

That doesn’t seem right. If I believe this ticket will lose, and it turns out that (because one of the numbers corresponds to the Chelsea vs scrubs game) that the nearest world in which it wins is a long ways away, our judgments about the luckiness of my belief don’t seem to change. Or at least they don’t to me. Known distance from the actual world matters more than actual distance, I think for determining whether my belief is true by luck or not.

If I just believe this ticket will lose for standard lottery reasons, then Nelkin will still say I don’t know it will lose. But by her standards, all I need to do is have a minimal amount of knowledge about the underlying games in order to genuinely know the ticket will lose. And that doesn’t seem right either. Unless I’m deeply involved in fixing the games or some such, I think intuitions about the cases are that I can no more know a particular Sunday Lotto ticket will lose than I can know a Saturday lotto ticket will lose, even if I know a little bit about football.

If this is right about the intuitions, one of three things follows.

# We reject both intuitions (the one about Sunday and the one about Saturday) as being bad sceptical intuitions; or
# We find a way to distinguish Saturday from Sunday lotteries; or
# We find a new explanation for what is wrong with beliefs about lotteries.

I’m all for option 1, but obviously it isn’t the only option on the table.