def simulate_under_null(num_chances_to_change):
    """Simulates some number changing several times, with an equal
    chance to increase or decrease.  Returns the value of our
    test statistic for these simulated changes.

    num_chances_to_change is the number of times the number changes.
    """
    uniform = Table().with_columns(
        "Change", make_array('Increase', 'Decrease'),
        "Chance", make_array(0.5,        0.5))
    test_statistic = uniform.sample_from_distribution('Chance', num_chances_to_change)
    return abs(test_statistic.column("Chance sample").item(0) - test_statistic.column("Chance sample").item(1))

uniform_samples = make_array()
for i in np.arange(5000):
    uniform_samples = np.append(uniform_samples, simulate_under_null(num_changes))

Table().with_column('Test statistic under null', uniform_samples).hist(0, bins=np.arange(-100, 400+25, 25))