Hearthpwn

No daily quest, rerolling quests not available, quest menu takes over a minute to open, must be a new patch.

I have 80g quest to trade on Asia

Mungo#2613

I will show quest then you challenge me for trade.

I'd say 1% is way too high then. Games where I had lethal, missed it and then went on to lose are memorable because they happen so rarely. Games which are close are not the norm in this game.

Looks to me like that maths you present is based on the assumption that 1% of all game losses are due to having the opponent miss lethal and then you go on to win the game. Is that what you really mean? Or are you trying to say "given that the opponent has lethal on board, they will miss lethal and I will win 1% of the time" which would be very different.

I have challenge on US, mungo#2613, let's trade since none of my friends seem to be online.

EDIT: If it goes like an hour without trading just add me as friend and I will do the quest anyway.

EDIT2: Done, thanks Viniks (spelled it right 3rd time lol)

Quote from mungo2613 >>

Check out my post here https://www.hearthpwn.com/forums/hearthstone-general/general-discussion/205335-need-good-mathematicians?comment=39

which gives you the probability of drawing a 1-of from your deck by the time you have drawn D cards given that you mulligan M cards.

I simplify the argument in this post (same thread) https://www.hearthpwn.com/forums/hearthstone-general/general-discussion/205335-need-good-mathematicians?comment=53

The formula is

Going first:

p(I have card after D draws or in opening hand) = 1/10 + 9/10 * (M/27 + (1 - M/27) * D/27)

Going second:

p(I have card after D draws or in opening hand) = 4/30 + 26/30 * (M/26 + (1 - M/26) * D/26)

I can work out the odds for a card you have 2 of in a deck as well if you want.

EDIT: That assumes you always keep the card you want in your mulligan. If you toss the card away (or you didn't have it before you mulligan) you use

p(I have card after D draws) = M/27 + (1 - M/27) * D/27 (when going first)

when going second and mulligan M cards

p(I have card after D draws) = M/26 + (1 - M/26) * D/26

instead. (Which is the second half of the formula posted above). The first set of formulae are

p(I had it pre mulligan or I drew it after D draws)

= p(I had it pre mulligan) + p(I didn't have it pre-mulligan) * p(I draw it after D draws)

= p(had it pre-mulligan) + (1 - p(had it pre-mulligan)) * p(I draw it after D draws)

 I made a mistake here. I bolded the part if the quote which is wrong.

If you toss a card away (which is a 1-of) in the mulligan the probability of having drawn it by the time you have made D draws is D/27 (going first) or D/26 going first, since you can't get it back in the mulligan phase. The 2nd set of formulae in the post above are correct if you did not toss the card away during the mulligan.

Summary:

Chance to have a specific 1-of card before you mulligan is 3/30 going first or 4/30 going second.

If you did not see the card pre-mulligan use the 2nd set of formulae to calculate the probability of drawing it by the time you have made D draws when you mulligan M cards.

If you threw it away during the mulligan phase use D/27 or D/26 when going first/second.

The probability of drawing a specific card from your deck each time you draw a card is of course just (number of cards of that type left in the deck)/(cards left in deck).

The second set of formulae just linearly interpolate between M/27 (or M/26 going second) and 1 (probability of having drawn the card after drawing your entire deck). M/27 (M/26 going second) is the probability of receiving the 1-of card when you mulligan M cards if you did not have the card pre-mulligan. If you threw the card away during the mulligan phase, the linear interpolation goes from 0 (probability of getting the card back from the mulligan) and 1.

The more complicated first set of formulae is the probabilities you should see if you run a simulation many times for a fixed number of cards M being mulliganned when your mulligan strategy is - always keep the card if you see it before you mulligan. (Hmm, that means the formula is wrong if you mulligan all your cards, since that would mean you did not have the card before you mulligan. In  that case you'd just use the second set of formulae with M = 3 or M = 4 when going second since p(I had it before I mulligan M cards) is then equal to 0).

Check out my post here https://www.hearthpwn.com/forums/hearthstone-general/general-discussion/205335-need-good-mathematicians?comment=39

which gives you the probability of drawing a 1-of from your deck by the time you have drawn D cards given that you mulligan M cards.

I simplify the argument in this post (same thread) https://www.hearthpwn.com/forums/hearthstone-general/general-discussion/205335-need-good-mathematicians?comment=53

The formula is

Going first:

p(I have card after D draws or in opening hand) = 1/10 + 9/10 * (M/27 + (1 - M/27) * D/27)

Going second:

p(I have card after D draws or in opening hand) = 4/30 + 26/30 * (M/26 + (1 - M/26) * D/26)

I can work out the odds for a card you have 2 of in a deck as well if you want.

EDIT: That assumes you always keep the card you want in your mulligan. If you toss the card away (or you didn't have it before you mulligan) you use

p(I have card after D draws) = M/27 + (1 - M/27) * D/27 (when going first)

when going second and mulligan M cards

p(I have card after D draws) = M/26 + (1 - M/26) * D/26

instead. (Which is the second half of the formula posted above). The first set of formulae are

p(I had it pre mulligan or I drew it after D draws)

= p(I had it pre mulligan) + p(I didn't have it pre-mulligan) * p(I draw it after D draws)

= p(had it pre-mulligan) + (1 - p(had it pre-mulligan)) * p(I draw it after D draws)

You undeleted some of my old posts, thanks for that.

I notice however we are unable to see post history from our account anymore, to see this you have to go to a thread you posted in, find one of your posts, and use the drop down list under your username, which takes me here: https://www.hearthpwn.com/members/mungo2613/posts

which has tabs for

However if I just go to "My account" from the top right drop down (at the top of each page) it takes me here: https://www.hearthpwn.com/account which does not have a "posts" tab, it only has tabs for 

which seems like a bug

EDIT: Ok, I can get to my posts via the my decks dropdown from the top right menu instead, so it's not that major but it is a bit annoying.

Quote from FrostyFeetEU >>

Quote from mungo2613 >>

Thanks for the name change.

Are all my old posts permanently gone now then? There was some decent mathematics going on in some of them (mainly probabilities of drawing cards etc. by a certain turn when mulliganning certain number of cards, discover probability discussion, etc.) which would be a shame to lose, took me a long time to work out :(

 They have been deleted but we can undelete them. The downside is that there's no mass undelete so we'd have to do it one by one. For that reason I'm personally willing to resurrect single posts (if you can guide me to the thread in question) but not entire post histories.

 I had a thread called mungo's maths thread which hasn't been posted in in ages but had a lot of mundane probability calculation in it, but it showed how you would go about calculating mulligan odds, and went into a lot of detail. (EDIT: I think I did a big calculation about the probability of having a card after N turns when you mulligan M cards in that thread andcame up with a formula which took a while to work out and then I simplified how I approached the problem and came up with the same answer as the complicated version which was reassuring).

There was another good discussion I remember in one about building beasts which got moved into the hunter forum (even though discover is a generic thing really) which had some good info about discover probabilities. The maths in that was a lot more straightforward though.

I also had a few threads were I performed a few chi-squared tests on data to get evidence about whether data appeared to come from random distributions or not, you could probably search for chi-squared to resurrect those. That's handy if people present data and say it isn't random ;)

Dunno if anyone else would be interested but it is handy for me to refer back to in case someone asks a question I know I did a comprehensive answer about previously.

Thanks for the name change.

Are all my old posts permanently gone now then? There was some decent mathematics going on in some of them (mainly probabilities of drawing cards etc. by a certain turn when mulliganning certain number of cards, discover probability discussion, etc.) which would be a shame to lose, took me a long time to work out :(

Please change my username back to mungo2613

I can't see my collection because of my usrername change either.

So I clicked on the stream and got about 10 minutes of ads before any content

Well, has she asked you to craft Archbishop Benedictus, to increase the size of your deck?

But I haven't got Malygos (which seems like a good combo), does he get the C'Thun buffs as well if played the turn after you play C'Thun? If so I may try that in wild.

So it took them until the Global Games actually started to remember to press the button to run the script (5 weeks after voting closed I think).