oldbloke's mutterings

got a return from 4 of the last 5 attacks on the lotteries.
i'd offer shares in a syndicate, but it is of course a fluke of the sort we statisticians expect to see every now and then.
however, it's amusing that it's happened _exactly_ coincidentally with my becoming bored with getting buggerall from the Lotto and branching out into the Euromillions and Thunderball. the 4 returns have been 2 from the Lotto and 1 each from the other two.
a spreadsheet i worked up seems to show that both the Thunderball and Euromillions are better value - as far as lotteries have any value - than the Lotto. yet although the DailyPlay looks even better on the same criteria (or less bad, I should say), it doesn't tempt me. this is something to do with what some philosophers of gambling call Utility Value - which attempts to measure something rather vague that lets people justify bets they know are theoretically bad. i've never seen a formula for it though...[1]
UV must of course include the Return and the Probability, but in nothing like as simple a fashion as R*P - that just measures the degree of rip-off. UV has to include some relationship between the Return and your annual income or the cost of living (the "never work again" factor), and between the entry cost and your disposable income. it's the fact that almost anyone can afford to blow a quid on the Lotto, yet the jackpot would mean they never have to work (or draw benefit) again, that makes the ridiculously long odds playable.

[1] Haven't actually looked, of course. Perhaps I should.