What are the differences between these two methods? They both get the standard deviation and 95% confidence interval but they do it differently, what's the difference between them?

Here's the set up for the problems:
>wins = 30
>losses = 19
>total = 49

>win rate = 0.61
>lose rate = 0.39
>mean = 29.89

---

Now for the 1st pair of equations:

Standard deviation:
>sqrt(0.61 * 0.39/49) = 0.07

95% confidence interval:
>1.96 * sqrt(0.61 * 0.39/49) = 0.14

Standard deviation is 7% and ninety five confidence interval is 14%. Also the numbers (0.47, 0.75) were gathered from the 95% confidence interval but I'm not sure how.

---

Now for the 2nd pair of equations:

Standard deviation:
>sqrt(49 * 0.61 * 0.39) = 3.41

95% confidence interval:
>29.89 - (1.96 * 3.41) = 23.21
and
>29.89 + (1.96 * 3.41) = 36.57

Which is (23.21, 36.57) but they have nothing to do with the previous numbers of (0.47, 0.75)

Which one is the correct way?