101
Dominion General Discussion / Re: Rats?
« on: January 30, 2015, 12:33:44 am »My point was never to say that Rats is never good as a curse trasher. It was to say that it's very rarely good. You can introduce edge cases all day long. But each additional constraint you add just lowers the odds even further that it will actually happen in real life.
Remember, according to you, you need: 1. Rats, 2. Curse attack, 3. No curse trashing, 4. Light trashing, and 5. Strong engine components. Even if you count all of your additions (most of which I personally consider to be real stretches) among the possibilities, you're still looking at about a 1 in 1200 chance of all those pieces coming together. The notion that it's "just as likely as any two card combo" (which is 1 in 400) is just patently false.
I blame you for the half hour I just spent figuring out the formula required to find the probability of getting a d-sized combo in a k-sized sample from an n-sized set. For your information, it's k!(n-d)!/(n!(k-d)!) . Which, for a 2-card combo in a 10-card kingdom (ignoring Banes because fuck that) from a set of 206 kingdom cards is exactly 9/4223. Which rounds to 1/400.
EDIT: Fun fact! With the release of Adventures (30 kingdom cards), this will decrease to ~1/550 !!! ALL OF OUR TIME HERE IS WORTHLESS
This is why I computer science:
Code: [Select]
import random
cards = range(206)
count = 0
trials = 1000000
for i in xrange(trials):
kingdom = random.sample(cards, 10)
if (1 in kingdom) and (2 in kingdom):
count += 1
print float(count)/float(trials)
0.002133
off by <0.1%