I'm an undergraduate taking my first-ever upper division math class. It's Combinatorics, which is pretty

There are $\binom{52}{5}$ different hands possible in a single poker deck. In a double deck, however, there are more than that, because some cards can be repeated.

The way that I've been attempting to enumerate this is by going through the different possibilities,

  • abcde
  • aabcd
  • aabbc

and trying to enumerate the number of different combinations of cards that can be rearranged to those specific permutations.

So, for example, aa-bcd has $\binom{52}{1}$ possibilities for the "aa" pair, and a further $\binom{52-1}{3}$ possibilities for the "bcd" part of the hand. So would there be $$\binom{52}{1} \binom{51}{3}$$ possiblities?

By this logic, I come up with the summation

$$\binom{52}{5} + \binom{52}{1}\binom{51}{3} + \binom{52}{2}\binom{50}{1}$$,

(Edit: Slight mistake, the last one should be $\binom{50}{1}$ not $51$. Pointed out by Long, below.)

but since I can't find a hard answer, I'm finding it difficult to double-check my work.

What do you think? Is this the right way to "think through" the problem? Have I made an important mistake (likely)?

Edit: Thank you for all of your help. I took the confirmation and generating function technique I learned here to write an extension to this, on how to count the number of 5 card hands from a double, triple, and quadruple deck.

  • $\begingroup$ You should get $3\,748\,160$ if you make the correction as suggested by Kevin Long's answer. Another way to look at it is using the generating function $(1+x+x^2)^{52}$ then taking the coefficient of $x^{5}$. With a bit of algebraic manipulation this gives $\binom{56}{5}-\binom{52}{1}\binom{53}{2}$. $\endgroup$ – N. Shales Feb 1 '18 at 18:31
  • $\begingroup$ Yes, now you have edited your answer is correct! $\endgroup$ – N. Shales Feb 1 '18 at 19:27

Your reasoning is correct. The different cases are just counting how many pairs of repeated cards you have (either one, two, or three). You should just note that you've made a slight mistake on the third case- see if you can spot what it is.

EDIT: With the edit, your answer is correct.

  • $\begingroup$ Thank you for the confirmation and mistake-catching. $\endgroup$ – Andrew Quinn Feb 2 '18 at 1:31

We have that for your hand, you could have any particular card appear in it either no times, once, or twice, so the solution can be found as the coefficient of $x^5$ of the expanded function $$f(x) =(1+x+x^2)^{52}$$

  • $\begingroup$ Very cool! Generating functions are an incredible tool when it comes to things like this. $\endgroup$ – Andrew Quinn Feb 2 '18 at 1:30

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.