Impossible odds

Please someone that isn’t a complete retard calculate this (bc i obviously am, losing with the best arena deck i drafted in a long time):

1x corsair cache
3x arcanite reaper

All in last 6 cards.

Odds?

P.s. yes i am salty as hell bc i’m so sick and tired of my streak of bad luck

let’s pretend it’s 12.43%, would you feel any better? :thinking:

2 Likes

Well, there are 6 possible spots out of 30 for 4 cards to fit in there. The chances that those 4 cards are in the last 6 of the deck is hard to calculate.
Each of those cards could be in any one of those last 6.
The chance that 1 card is in the last 6 would be 1/5 (aka 6/30).
That then leaves 5 cards of 29 left.
The odds of the 2nd card is 5/29.
The odds of the 3rd card is 4/28.
The odds of the 4th card is 3/27.

So, 6/30 x 5/29 x 4/28 x 3/27 = 0.00054734537.
So it’s 0.054734537% chance.

Those are the same odds for any 4 cards to be in the bottom 6 of the deck and it happens every single game (they’re just not always the cards you wish you drew).

If you also think of how many people play this game, it’s not too uncommon. If the odds were 1% instead, lets say, then 1 out of every 100 matches this would happen to some player somewhere. If it was 0.1% chance, it would happen to 1 out of 1000 matches for some player somewhere. If it was 0.01% chance, it would happen to 1 out of 10,000 matches. Since it’s about 0.055%, then it would happen to about 5.5 matches out of 10,000. If 100,000 matches are played in Hearthstone in just one day, then 55 people would have this happen to them. It has to be SOMEONE. And this time it was you.

5 Likes

The odds don’t matter, if the game dictates its your time to lose, you will, be it either via RNG or card draw.

I played 2 games in a row with my bomb warrior yesterday against 2 priests. One of them ran mindrender ilucia (who plays that card nowadays anyway, but w/e) and stole my bulwark, then next turn, I ate 1 bomb of his deck as he had drawn none during the game (while he had about 15 cards on his deck and 3 bombs among them). Then next turn, while he had the bulwark equipped and his deck back to him, he pulled the other 2 bombs, negating all the damage. Conceded.

Next game: priest plays thoughtsteal while my deck had about 20+ cards still, stole the exact 1 bloodsail pirate I had and removed 1 durability of my weapon, destroying it. Later on, he plays mind vision against my hand of 5-6 cards IIRC and again, guess what - got a copy of my bulwark.

I don’t care about numbers, you can’t convince me this is for real when this kind of impossible shenanigans happens every other game.

2 Likes

And there’s a 100% chance that that SOMEONE makes a forum post about it.

Sampling bias at its finest, with a side helping of negativity bias.

3 Likes

Good mathing there. If you don’t mind, what would be the odds of not drawing a card you have 9 duplicates of in the first 10 turns? Was running mill, copied nine Coldlight Oracles into my deck over the first few turns and did not draw a single one til after turn 10 when my opponent already had lethal. The statistical improbability of that happening has to be off the charts.

I’m not quite sure how many cards are in the deck and how many you’d drawn and what turn you duplicated the cards.

Can you tell me how many cards were remaining in the deck at the time you duplicated the card (after they were shuffled in)?

Don’t have the exact number but somewhere between 20-24 were left.

According to McEnany, 1 in 4 quadrillion to the 4th power. I don’t think she maths.

1 Like

Alright, so I guess I will take the average of that range, 22 cards. I’m assuming the 9 were shuffled in at this point, which brings the total deck size to 31. At this point, I am calculating that there are 9 total Coldlights out of 31 cards and you hadn’t drawn at least 1 of them until AFTER turn 10. If you had 22 cards it’s hard to say whether you went first or second, but I will say you duplicated them on turn 4. This leaves a total of 6 turns, or 6 drawing chances until turn 11 occurs (the next turn after 10 mana for both players), since you said you hadn’t drawn one until AFTER turn 10.

You have 6 drawing chances.
31 cards - 6 = 25. There are a total of 25 spots that those 9 cards must be in for you to NOT draw ONE for 6 turns (The bottom 25).
So, probability for the 1st card to be in the last 25 would be 25/31. This takes up a spot, so there are 24 left for 8 Coldlights out of a total of 30.
2nd card is 24/30.
3rd = 23/29
4th = 22/28
5th = 21/27
6th = 20/26
7th = 19/25
8th = 18/24
9th = 17/23

25/31 x 24/30 x 23/29 x 22/28 x 21/27 x 20/26 x 19/25 x 18/24 x 17/23 =
0.10133766863
or
10.133766863% chance that you would NOT draw one.
You got pretty unlucky, 9 out of 10 matches with the same situation you should have drawn at least 1.

1 Like

How fun. Elementary school math!

Each turn is a SpecificCard (SC) out of RemainingCards (RC) to draw. (in your case, 9 dupes out of 30 for turn 1 and then (f) RC = N–1; or RC minus 1 for each turn).

If you have 30 cards and you have 9 of something in there, your first draw is 9/30.

For the mulligan there’s a chance, at each card to draw 1 in X (or, card 1 has a 9 in 30 chance). HS shouldn’t give you the same card back that you mulligan, so realistically, without the coin, if you didn’t have one of the dupes, you should have started with a 9/27, then a 9/26, then a 9/25 (or a ~35% to ~38% chance to draw a card, just over 1 in 3 to draw a dupe). If you didn’t, the next card would be a 9/26 again, because the pool went back to 27 cards vs the mulligan where you can drop the pool to 27 cards to begin with.

So you have that and you’re on your second card and you HAVE NOT drawn your 1/9 dupe cards. The 2nd card is now 9/26. This repeats each time until you draw your card.

9/N+9/N-1+9/N-2 if you have not drawn your card by the 21st card you will have 100% chance for each card (9/9, 8/8, 7/7).

If you do draw the card, the odds to draw another reduce for the following card.

9/27 and you draw one. The next one will be 8/26, then 8/25 until you draw one.

roughly speaking 9/30 = 30% chance to draw, or just under 1 in 3 chance to draw that 1 of 9 dupe. Those odds go up each time you draw a card and DON’T draw a dupe, but not by a lot. Every card is 1/30th a chance to draw.

Your odds of drawing all 9 in the first 9 draws: 6.98951224e-8… pretty slim. That’s the near astronomical number.

To not draw a card you have 9 copies of each turn starts at ~70% and drops to ~57% by the 10th turn. Meaning, you have a ~43% chance to draw a card of the 9 dupes in the remaining 21 cards. You do not hit 100% until the 21st card, when there are 9 left. or 9/9=1 represented as a fraction.

I know it doesn’t sound astronomical… because it isn’t.

1 Like

I think that’s where people get confused. The phrasing creating a logical fallacy inferring an absolute over a probability.

Sure I have a 49.9% chance to land a coin on heads. Each flip I have that same probability; however, there also exists the probability that I would not flip it on heads… ever.

So even if my odds were slim to not flip it on heads (or highly in my favor) through the greater probability of probabilities (flipping a coin 1000 times and one of those being heads), each chance is still almost 50% I miss. So while we can ‘imagine’ that I should have definitely flipped a heads one of the last 999 times, that means this next one is definitely heads!

It’s that basic math confusion that gets gamblers in trouble. Ok, I haven’t won in the last 300 bucks, this next bet should absolutely win. Then you get gamblers remorse because… well… math and probabilities. The odds ALWAYS favor the house.

10 games in a row of not drawing 1 of the 9 cards you have in a 30 card deck (which gets worse as you increase the size of the deck) overall has a low probability, but that’s not how you calculate it effectively - you calculate it that way for how it happened on the greater scale, but it’s still an XX/YY card draw each effectively. For every card you add on another .1% chance to NOT draw it.

We create the idea that it’s low odds to do in a row, because that’s how math works, but probability isn’t a globally averaging number. The averages should work out on a long enough scale, because there may be a game where all 9 are drawn consecutively in the first 9 draws! (maybe ;P)

1 Like

Yes, that’s all true. And right about the gambler’s fallacy. The numbers I made don’t represent anything but the average if the exact average were to happen.

2 Likes

Brandiddley and Twykster, thanks to you both. That game had been really sticking in my craw. I just couldn’t fathom the likelyhood that so many copies of a card (especially one that’s in my opening hand, easily 80% of the time) wouldn’t be drawn over 10-12 turns. Guess there just wasn’t enough copies to get over the statistical hump of a deck stacked against me.

It’s an Arena deck, bub.

You are welcome.

Have you read the TOS? Not really sure that’s an appropriate word to use.

Arena? Bad command or file name sorry. Fine I will delete it. I hope you are happy proving my nubbness. Im going to actually calculate it now to make up for it. Answer coming in 2023. Stay tuned. Reading is fundamental. The more you know. :rainbow:

Why wouldn’t it, I can’t imagine someone mentally slow being helpful here .

Yes exactly, if something unlikely happens in a game with lots of rng and thousands of matches being played daily, it’s because the game specifically wants you to lose because it doesn’t like you.

2 Likes

Whatever you say, simp

1 Like

Yikes.

1 Like