Simple hypothesis testing

Niels has a Magic $8$8-Ball, which is a toy used for fortune-telling or seeking advice. To consult the ball, you ask the ball a question and shake it. One of $5$5 different possible answers then appears at random in the ball. Niels sensed that the ball answers "Ask again later" too frequently. He used the ball $10$10 times and got "Ask again later" $6$6 times.

Let's test the hypothesis that each answer has an equal chance of $20\%$20, percent of appearing in the Magic $8$8-Ball versus the alternative that "Ask again later" has a greater probability.

The table below sums up the results of $1000$1000 simulations, each simulating $10$10 random answers with a $20\%$20, percent chance of getting "Ask again later".

Let's agree that if the observed outcome has a probability less than $1\%$1, percent under the tested hypothesis, we will reject the hypothesis.