Actual Lotto (181 views)

[eric]

So the NBA lotto irl is an actual lotto where there are fourteen balls of which four are drawn, and each team has a certain number of combos they win with, with worse records having more combos. Now that we're doing a weighted lotto, I could simulate it by comparing an RNG to the overall chance or I could simulate the actual drawing of lotto balls and people could consult a chart (that I'd make) to see if they were still alive, something like this but with actual numbers:



Or like this but much bigger:



Would people be interested in that?

[Deleted]

i dont know what any of that means to know if i would be interested or not

[Deleted]

I want scratch offs

[eric]

Apr 18, 2019 20:50:46 GMT @odin said:

i dont know what any of that means to know if i would be interested or not

so look at the second graphic, that'll be easiest

Seattle had an 88/1000 chance of winning the 2007 lotto.

Specifically, they would win if the four lotto balls drawn were 4-6-9-14, or 4-9-11-12, or 5-6-7-8, etc.

I can run lotto one of two ways. I could do an RNG and if it comes up between .000 and .088 then Seattle wins, and so on down the line, OR I could simulate pulling the actual lotto balls. If the first one I "pull" is a 4 then Seattle knows they're still alive, but if it's a 1 they already know they're eliminated because none of their winning combos has a 1 in it.

I thought about it more and I would probably have a Google Doc rigged up that would automatically gray out areas of the chart and team names so people would be able to see immediately if they're still alive. Let me get that set up and maybe it'll be easier for people to conceptualize too.

[TimPig]

Apr 18, 2019 21:08:30 GMT eric said:

If the first one I "pull" is a 4 then Seattle knows they're still alive, but if it's a 1 they already know they're eliminated because none of their winning combos has a 1 in it.

I think this makes sense and sounds like fun but I don't want you to put the work in when Fecta and HT aren't even checking in.

[Deleted]

doesnt seem worth the effort

[skrouse]

Would be cool but only if it's not a lot of work.

[eric]

ok so check this out

the first line is how it would have looked before pulling any balls for last year's standings

the second line is if i pulled a 1 first, at which point four teams are eliminated and the oaks have 87% of the remaining combos

the third line is if i pulled a 3 first - bullets and bulls eliminated, hornets become the favorites to win. but then i pull a 2 and the hornets are out, and then i pull a 4 and the mavericks have 10 of the remaining 11 combos but the oaks still have a shot.

the fourth line is if i still pulled a 3 first, but then a 5 - now the hornets are still alive and increase their odds, and even more so if i pull a 14 after that

[eric]

it started out because i was curious how the NBA came up with the # of combos each team got, and it turns out the answer is probably they just made it up out of whole cloth BUT since i already had those it was easy to go from there to a spreadsheet that adjusts for any given # of teams. it is extremely low work load for me to run these numbers at this point. in practice i would probably upload it in code tags rather than pngs though

[TimPig]

This actually seems really cool, and I get the impression you want to do it (and therefore it probably isn't an excessive amount of work).

Sounds good to me.

[Deleted]

Yeah but like who is this even for. It builds suspense for what, a couple of seconds? Dont see the point.

[Deleted]

Insider it a maintenance subroutine