I've tried a few iterations, but can't quite get this. I've passed all of the self-checking tests, even the additional tests suggested by Gabbek in the other discussion post for this challenge. This will even pass the first big test (Big/1). but it fails on Big/2. (It says the correct answer is 8; my code finds 9.) Could someone help me figure out where the flaw in my logic is??

from math import ceil

difficulties = lambda q: [ ceil((i[-1] - i[0]) / (len(i) - 1)) for i in q ]

def workout(sessions, additional) -> int:

    # Transform from list [x, y, z, ...], turn to [(x,y), (y,z), ...]
    sessions = [[i, j] for i, j in zip(sessions[:-1], sessions[1:])]

    while additional > 0:

        # difficulties() [at top] will take index[-1] - index[0] of each list in sessions,
        # and will divide by len() - 1 to give us the difference in number of pushups 
        # capable over those workouts, including with any additional workouts we've added.
        gaps = difficulties(sessions)

        # we want index of the highest difficulty gap, to add workouts to break it up.
        # Then change that difficulty to '0' in this so we can evaluate other difficulties.
        m = gaps.index( max(gaps) )
        gaps[m] = 0

        # This will break the difficulty in 2, if we only add 1 workout.
        # (Continues to divide by 3, or 4, etc., depending on how many workouts we'll add)
        breaks = 2

        # While the broken-up difficulty is still higher than next max difficulty,
        # add another workout to break it up.
        while (sessions[m][-1] - sessions[m][0]) // breaks >= max(gaps) and breaks <= additional:
            breaks += 1

        # When I know how many breaks to add, add that number of '+'
        for i in range(1, breaks):
            sessions[m].insert(i, '+')

        # Number of workouts to add is the # of breaks - 1 
        # (e.g., halving the difficulty means adding 1 workout)
        additional -= (breaks - 1)

    return max(difficulties(sessions))