3
$\begingroup$

I feel confused about selecting suits and cards that each suit corresponds to.

Suppose we have a standard deck of cards, and we want to form a 6 hands with at most 3 suits.

I understand this needs to be spliced into 3 cases to discuss, which are 1 suit case, 2 suits case, and 3 suits case.

But I feel confused when it comes to select cards.

For example, I am not sure if I understand this correctly, but I feel like if we only need to form cards with a single suit, then this can be done by $4(13 * 12 * 11 * 10 * 9 * 8)$ ways, since each suit has 13 cards, and we have 4 suits here.

And when it comes to the 2 suits case, I feel like we only need to draw from the total of 26 cards to form such hands, and there are 6 ways to do this, but I am not sure if my understanding about the 26 cards is right or what.

And for the 3 suits case, I feel like it's just choose 3 suits out of 4, and draw cards from total of 39 cards.

Is my understanding towards the total number of cards in each case correct? If not, why so?

$\endgroup$
3
  • 2
    $\begingroup$ Could you try to edit this down to a more specific question? It's extremely difficult to reply to this kind of thing, especially with all the "I feel confused" type of remarks and the vague multiple questions. $\endgroup$ Commented Sep 21, 2022 at 21:37
  • $\begingroup$ Welcome to MSE! Consider reading the "how to ask a good question" guidelines; your actual question is vague here, and your write-up has the same issue. For instance, do you mean $6$ hands, or hands of $6$ cards? Are you looking for total possibilities? Etc. $\endgroup$ Commented Sep 21, 2022 at 22:50
  • $\begingroup$ The answer given by JMoravitz, which is more elegant than my answer, employs Inclusion-Exclusion. See this article for an introduction to Inclusion-Exclusion. Then, see this answer for an explanation of and justification for the Inclusion-Exclusion formula. $\endgroup$ Commented Sep 21, 2022 at 23:27

2 Answers 2

3
$\begingroup$

There is no need to approach by breaking into cases. Case work will just become too messy and does not generalize nicely to other sized hands. Consider answering the question instead as: "How many ways are there to form a six-card hand such that at least one of the suits is missing?"

Let $A$ be the event that clubs are missing, $B$ that diamonds are missing, $C$ that hearts are missing, and $D$ that spades are missing respectively. You are tasked with finding $\Pr(A\cup B\cup C\cup D)$

This expands by inclusion exclusion to be:

$$\Pr(A\cup B\cup C\cup D) = \Pr(A)+\Pr(B)+\Pr(C)+\Pr(D)-\Pr(A\cap B)-\Pr(A\cap C)-\dots - \Pr(C\cap D) + \Pr(A\cap B\cap C)+\dots + \Pr(B\cap C\cap D) - \Pr(A\cap B\cap C\cap D)$$

By symmetry, this simplifies to be:

$$ = 4\cdot \Pr(A) - 6\cdot \Pr(A\cap B) + 4\cdot \Pr(A\cap B\cap C)$$

noting that we could leave off the intersection of all four since it is impossible for all suits to be missing simultaneously.

The probability that clubs are missing? That is answered simply by a hypergeometric distribution argument and is $\Pr(A) = \dfrac{\binom{39}{6}}{\binom{52}{6}}$. Similarly for the probability that both clubs and diamonds are missing as $\dfrac{\binom{26}{6}}{\binom{52}{6}}$ and so on.

Your probability is then:

$$\dfrac{4\cdot\binom{39}{6} - 6\cdot\binom{26}{6}+4\cdot\binom{13}{6}}{\binom{52}{6}}$$

$\endgroup$
7
  • $\begingroup$ I am unsure whether the original poster is presenting a Probability problem, or an enumeration problem. I didn't see anything in the posting that asked for the probability that not all $(4)$ suits are involved. $\endgroup$ Commented Sep 21, 2022 at 23:23
  • $\begingroup$ It's actually an enumeration problem, it's probably my problem describing the question unclearly, but I don't see why you wrote $4 * {39 \choose 6}$ to minus the later part, I thought it was going to be the sum of those 3 identities to obtain the total number of possible arrangements. $\endgroup$
    – MeTAZOIS
    Commented Sep 21, 2022 at 23:39
  • $\begingroup$ @user2661923 "at most 3 suits are present" and "at least one suit is not present" are equivalent statements. $\endgroup$
    – JMoravitz
    Commented Sep 22, 2022 at 0:35
  • $\begingroup$ @JMoravitz My previous comment was suggesting that the original poster was not trying to compute a probability, but was instead trying to enumerate. $\endgroup$ Commented Sep 22, 2022 at 0:37
  • $\begingroup$ @MeTAZOIS The enumeration problem and the probability problem are one and the same, the probability problem just takes it one step further and takes the ratio. Ignore the division if you want the enumeration. As for why you see some subtraction going on, that is simply how inclusion-exclusion works. And yes, inclusion-exclusion works just as well for enumeration as it does for probability. You alternately add and subtract increasing intersections, adding all singles, subtracting intersections of two, adding of three... $\endgroup$
    – JMoravitz
    Commented Sep 22, 2022 at 0:37
-1
$\begingroup$

There are two ways of enumerating the number of possible $(6)$ card hands, from a $(52)$ card deck, when the cards are drawn without replacement, and where you want to reject any $(6)$ card hands that involve all 4 suits.

One way is to enumerate the number of $(6)$ card hands that do involve all 4 suits, and deduct that from

$$\binom{52}{6} = \frac{(52)!}{(6!)[(52-6)!]}.$$

The alternative method is the direct approach, which is detailed below. I have chosen this approach because, while it is more difficult, it is also more educational.


You have a $(3)$ step process.

First, you have to identify all of the mutually exclusive ways of partitioning the number $(6)$ into the sum of three non-negative integers.

These are

  • 6,0,0
  • 5,1,0
  • 4,2,0
  • 4,1,1
  • 3,3,0
  • 3,2,1
  • 2,2,2

Then, one at a time, you have to enumerate each of the three mutually exclusive patterns above. In such an enumeration, the second step is to enumerate the total number of different ways of assigning suits, to fit the pattern. I refer to this as computing the suit-factor.

Then, the third step is to enumerate the number of ways of choosing the cards within each of the pertinent suits. I refer to this as computing the combination-factor.

Throughout this discussion, you need to know that if $~n \in \Bbb{Z^+},~$ and $~k \in \{0,1,2,\cdots,n\},~$ that the number of ways of choosing $k$ things out of $n$ things, sampling without replacement, where order of selection is regarded as irrelevant is

$$\binom{n}{k} = \frac{n!}{k![(n-k)!]}.$$


$\underline{\text{Case 1: Pattern = 6-0-0}}$

There are $(4)$ ways of choosing which suit to draw the cards from. So, you start with a suit-factor of $(6)$.

Then, you have the combination-factor, that represents the number of ways of selecting the cards, once the suit assignments are made. There are $~\displaystyle \binom{13}{6}~$ ways of choosing the $(6)$ cards within one suit.

Therefore, the overall enumeration here is

$$4 \times \binom{13}{6}. \tag1 $$


$\underline{\text{Case 2: Pattern = 5-1-0}}$

There are $(4)$ ways of choosing which suit to select $(5)$ cards from. Once this is done, there are then $(3)$ ways of selecting which suit to draw the single card.

So, the suit-factor is $(4 \times 3 = 12).$

Then, the combination-factor is $~\displaystyle \binom{13}{5} \times \binom{13}{1}.$

Therefore, the overall enumeration here is

$$12 \times \binom{13}{5} \times \binom{13}{1}. \tag2 $$


$\underline{\text{Case 3: Pattern = 4-2-0}}$

There are $(4)$ ways of choosing which suit to select $(4)$ cards from. Once this is done, there are then $(3)$ ways of selecting which suit to select $(2)$ cards from.

So, the suit-factor is $(4 \times 3 = 12).$

Then, the combination-factor is $~\displaystyle \binom{13}{4} \times \binom{13}{2}.$

Therefore, the overall enumeration here is

$$12 \times \binom{13}{4} \times \binom{13}{2}. \tag3 $$


$\underline{\text{Case 4: Pattern = 4-1-1}}$

There are $(4)$ ways of choosing which suit to select $(4)$ cards from. Once this is done, there are then $(3)$ ways of selecting which suit will be unused.

So, the suit-factor is $(4 \times 3 = 12).$

Then, the combination-factor is $~\displaystyle \binom{13}{4} \times \binom{13}{1} \times \binom{13}{1}.$

Therefore, the overall enumeration here is

$$12 \times \binom{13}{4} \times \binom{13}{1} \times \binom{13}{1}. \tag4 $$


$\underline{\text{Case 5: Pattern = 3-3-0}}$

The suit-factor is $~\displaystyle \binom{4}{2} = 6.$

The combination-factor is $~\displaystyle \binom{13}{3} \times \binom{13}{3}.$

Therefore, the overall enumeration here is

$$6 \times \binom{13}{3} \times \binom{13}{3}. \tag5 $$


$\underline{\text{Case 6: Pattern = 3-2-1}}$

The suit-factor is $~\displaystyle 4 \times 3 \times 2 = 24.$

The combination-factor is $~\displaystyle \binom{13}{3} \times \binom{13}{2} \times \binom{13}{1}.$

Therefore, the overall enumeration here is

$$24 \times \binom{13}{3} \times \binom{13}{2} \times \binom{13}{1}. \tag6 $$


$\underline{\text{Case 7: Pattern = 2-2-2}}$

There are $(4)$ ways of selecting the suit to be unused. Therefore, the suit-factor is $4.$

The combination-factor is $~\displaystyle \binom{13}{2} \times \binom{13}{2} \times \binom{13}{2}.$

Therefore, the overall enumeration here is

$$4 \times \binom{13}{2} \times \binom{13}{2} \times \binom{13}{2}. \tag7 $$


$\underline{\text{Final Computation}}$

To get the overall enumeration, all that you have to do is use a hand calculator against the computations in each of
(1), (2), (3), (4), (5), (6), and (7), above.

Then, take the overall sum of these seven computations.

$\endgroup$
5
  • 2
    $\begingroup$ Although your answer is not incorrect, I do find it irresponsible to be promoting an approach by casework to this level. Especially when the problem has an elementary approach that avoids it entirely. $\endgroup$
    – JMoravitz
    Commented Sep 21, 2022 at 23:01
  • $\begingroup$ @JMoravitz Different opinions make a horse race. The point of my answer is not to show the best way of attacking the problem. The original poster showed, in his posting, confusion over basic enumeration concepts. As I remarked at the start of my answer, I specifically detailed the more difficult way, precisely because it is more educational. The original poster is (in effect) drilled in the concepts of mutually exclusive partition patterns, suit-factors, and combination-factors. If I was writing a textbook, this is how I would educate the student. ...see next comment $\endgroup$ Commented Sep 21, 2022 at 23:04
  • $\begingroup$ @JMoravitz For the math student new to Combinatorics, detailing a more elegant approach is, in my opinion, cart before the horse. It is better to start with a deep understanding of what is going on, before striving for elegance. $\endgroup$ Commented Sep 21, 2022 at 23:06
  • 1
    $\begingroup$ Thanks for answering first, but I don't think this much level of details is involved in this problem. $\endgroup$
    – MeTAZOIS
    Commented Sep 21, 2022 at 23:40
  • $\begingroup$ @MeTAZOIS Did you find my answer educational? $\endgroup$ Commented Sep 21, 2022 at 23:41

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .