Not necessarily.

Let us rephrase the original statement to read, "If one is good with math, then one is bad with people." This can be represented more generally as "if p then q".

Since we know q to be true if p is true, we can also say that if q is not true, p cannot be true--this is the contrapositive, which is always true if the original statement is true. You have attempted to assert that if you are bad with people, you must be good with math--in effect, if q then p. This is the converse of the original statement and is not necessarily true (as you have astutely pointed out). One may, assuming the original statement to be true, in fact be both bad with people and bad at math, or good with math and bad with people, but not good with people and good at math. (We shall ignore for the moment that the original statement is not really true because the article doesn't mention social confidence. Rather, it mentions confidence in mathematical ability and enjoyment of math, not people. But I digress.)

Of course, you'd know this already if you were good at math.