An easy way to get sidetracked on the way to a solution is to assume that something is true, without realizing how that assumption constrains what you can do. One way to avoid this is to spend some time early on thinking about what assumptions you might be making, and whether you can relax them,

A: I was always taught that pi is equal to 3.14
B: me too, but this is a different world, right?
B: so would pi have to have the same value here?
C: I learned that pi is defined as circ/diam
B: so does pi define those, or do those define pi?

or whether you'd like to make extra assumptions, as a way of simplifying the problem,

K: I think if we assume that BCD is a diameter...
K: then we can use a theorem that says BPD is a right angle
L: that would let us use the Pythagorean theorem
M: but what if it's not true?
K: if we can get from there to the solution...
K: we can come back later and try to prove what we assumed