I don't want to discuss on the mathematical fact that 0 is an even number.
 You should include it although its effect is negligible here.
 It will not be negligible when the number of trials is small (say, n = 4) and P(T) is high (say 0.9).
 
 Am I right?