Determine the NULL and Alternate Hypothesis The following is a question related to Hypothesis testing from an introductory level stats book:

Every year on Groundhog Day (February 2), the
  famous groundhog Punxsutawney Phil tries to predict
  whether there will be 6 more weeks of winter. The article
  “Groundhog Has Been Off Target” (USA Today, Feb. 1, 2011)
  states that “based on weather data, there is no predictive
  skill for the groundhog.” Suppose that you plan to take
  a random sample of 20 years and use weather data to
  determine the proportion of these years the groundhog’s
  prediction was correct.
a) Describe the shape, center, and spread of the sampling distribution
  of pˆ for samples of size 20 if the groundhog has only a 50–50 chance
  of making a correct prediction.
b) Based on your answer to Part (a), what sample proportion values
  would convince you that the groundhog’s predictions have a better than
  50–50 chance of being correct?

I'm having a hard time figuring out the null and alternate hypothesis here?
 A: Typically the null hypothesis is the one in which you assume that no effect is present. In this case "no effect" would indicate that the groundhog's ground truth accuracy, which is unknown to you, is in fact 50% (chance, i.e. the groundhog's predictions have no true meaning). The alternate hypothesis is the one in which you assume that an effect is present. In this case the alternate hypothesis is that the groundhog's true accuracy is in fact greater than 50%. Since the groundhog's true accuracy can not be known, we sample his predictions and try to determine whether our observations are more consistent with the null hypothesis or the alternate hypothesis.
Due to binomial sampling statistics binomial distribution, if the groundhog's true accuracy is 50% and he makes 20 predictions, there is a range of possible 
fraction correct rates. From the binomial distribution (with p = 0.5 and n = 20), you can calculate the probability (p-value) of observing a given or more extreme value for fraction correct. In frequentist statistics, if the observed fraction correct has a p-value lower than a threshold (often p = 0.05), we say that we reject the null hypothesis in favor of the alternate hypothesis. 
