Adjust Your Market Risk Wisely With This Awesome Python and Google Sheets Rotati...

Methodology (in code):

(data is a DataFrame with ohlc prices.)

def get_sign(data):

    # Get sign of each bar (-1, 0, 1)
    data['sign'] = np.sign(data['close'] - data['open'])
    # Group together (including 0's at end of group)
    prev = data['sign'][0]
    g = 0
    for i, s in zip(data.index, data['sign']):
        if s == 1:
            if prev == 1:
                # continue up group
                data.loc[i, 'g'] = g
            else:
                # new up group
                g += 1
                data.loc[i, 'g'] = g
        if s == -1:
            if prev == -1:
                # continue down group
                data.loc[i, 'g'] = g
            else:
                # new down group
                g += 1
                data.loc[i, 'g'] = g
        if s == 0:
            # add to last group
            data.loc[i, 'g'] = g
        prev = s

    return data

def get_rotations(data):

    r = []
    for x in range(int(max(data['g']))):
        # if g == x or x+1, find min/max
        # if g sign is +, find the +/- rot (high)
        # if g sign is -, find -/+ rot (low)
        if data.loc[data['g'] == x, 'sign'].iloc[0] > 0:
            h = data.loc[(data['g'] == x) | (data['g'] == x+1), 'high']
            i = h.idxmax()
            h = h.max()
            r.append([i, h])
        if data.loc[data['g'] == x, 'sign'].iloc[0] < 0:
            l = data.loc[(data['g'] == x) | (data['g'] == x+1), 'low']
            i = l.idxmin()
            l = l.min()
            r.append([i, l])

    return pd.DataFrame(r)