> pbinom(13,17,.9)   # 
[1] 0.08264062
> pbinom(13,17,.95)  # 
[1] 0.008800605
> pbinom(13,17,.925) # 
[1] 0.03415986
> pbinom(13,17,.93)
[1] 0.02734
> pbinom(13,17,.932)
[1] 0.02487068
> pbinom(13,17,.9319) # Pretty much = .025.
[1] 0.0249907
> 
> p = seq(.01,.99,.01)
> L = pbinom(12,17,p,lower.tail=F)
> U = pbinom(13,17,p)
> plot(p,L,ylim=c(0,1),ylab="Tail Probability",type="l",lty=3,col="red")
> lines(p,U,lty=2,col="blue")
> abline(h=.025)
> legend(0.1,0.8,c("P(Y<=13 | p)","P(Y>12 | p)"),lty=c(2,3),col=c("blue","red"))
> 
> locator(2)
$x
[1] 0.9324363 0.5083268

$y
[1] 0.02274088 0.02645869

> 13/17
[1] 0.7647059
> binom.test(13,17,conf.level=.95)

	Exact binomial test

data:  13 and 17 
number of successes = 13, number of trials = 17, p-value
= 0.04904
alternative hypothesis: true probability of success is not equal to 0.5 
95 percent confidence interval:
 0.5010067 0.9318923 
sample estimates:
probability of success 
             0.7647059 

>