Quantifying the 21st-century trend of film sequels, remakes, and franchise reboots over the top 10 grossing films each year since 2000.

sequelmania.py

#!/usr/bin/env python

# Quantifying the 21st-century trend of film sequels, remakes, and franchise
# reboots over the top 10 grossing films each year since 2000.

import numpy
from scipy import stats

y = range(2000, 2013)

# The 'installments' dict contains the number of the installment in a series
# for each of the top 10 grossing films for each year since 2000. Remakes are
# considered first in the series, and franchise reboots are counted from one.

# Special Notes
# - Star Wars Episodes II and III are counted as fifth and sixth, respectively.
# - The three Bond films are outliers, but fair is fair.

# Source: http://www.boxofficemojo.com/yearly

installments = {}
installments[2000] = [1, 1, 2, 1, 1, 1, 1, 1, 1, 1]
installments[2001] = [1, 1, 1, 1, 2, 2, 1, 1, 3, 1]
installments[2002] = [1, 2, 5, 2, 1, 1, 3, 2, 1, 1]
installments[2003] = [3, 1, 1, 2, 1, 2, 1, 3, 3, 1]
installments[2004] = [2, 2, 1, 2, 1, 3, 1, 2, 1, 1]
installments[2005] = [6, 1, 4, 1, 1, 1, 1, 1, 1, 1]
installments[2006] = [2, 1, 1, 3, 1, 1, 1, 2, 21, 1]
installments[2007] = [3, 3, 1, 3, 5, 1, 3, 2, 1, 1]
installments[2008] = [2, 1, 4, 1, 1, 1, 1, 2, 22, 1]
installments[2009] = [1, 2, 6, 2, 1, 1, 1, 1, 2, 1]
installments[2010] = [3, 1, 2, 3, 7, 1, 1, 4, 1, 1]
installments[2011] = [8, 3, 4, 2, 4, 5, 4, 2, 2, 1]
installments[2012] = [1, 3, 1, 23, 1, 5, 1, 1, 1, 3]

original_means = [numpy.mean(installments[year]) for year in y]
o_slope, o_intercept, o_r_value, o_p_value, o_std_err = \
    stats.linregress(y, original_means)
o = lambda year: o_slope * year + o_intercept

# If the film is a remake, the number of previous versions of that film and any
# sequels is added to its installment number. If the film is part of a franchise
# reboot, the total number of previous films is added.

# Special Notes
# - Neither of the other two 2005 War of the Worlds films are included.
# - The Batman film serials of 1943 and 1949 are not included.
# - Previous adaptations of Sherlock Holmes are all counted as one.
# - Previous adaptations of Alice in Wonderland are all counted as one.

installments[2000][0] += 1      # How the Grinch Stole Christmas! (1966)
installments[2001][1] += 1      # Ralph Bakshi's Lord of the Rings (1978)
installments[2001][7] += 1      # Ocean's 11 (1960)
installments[2001][9] += 5      # Planet of the Apes series (1968 - 1973)
installments[2002][0] += 1      # Spider-Man (1977)
installments[2002][1] += 1      # Ralph Bakshi's Lord of the Rings (1978)
installments[2003][0] += 1      # Ralph Bakshi's Lord of the Rings (1978)
installments[2003][9] += 1      # Cheaper by the Dozen (1950)
installments[2004][1] += 1      # Spider-Man (1977)
installments[2005][3] += 1      # The War of the Worlds (1953)
installments[2005][4] += 6      # various King Kong films (1933 - 1986)
installments[2005][6] += 1      # Willy Wonka & the Chocolate Factory (1971)
installments[2005][7] += 5      # various Batman films (1966 - 1997)
installments[2006][5] += 4      # Superman series (1978-1987)
installments[2007][0] += 1      # Spider-Man (1977)
installments[2007][2] += 1      # The Transformers: The Movie (1986)
installments[2008][0] += 5      # various Batman films (1966 - 1997)
installments[2009][1] += 1      # The Transformers: The Movie (1986)
installments[2009][6] += 10     # Star Trek series (1979 - 2002)
installments[2009][9] += 1      # various Sherlock Holmes adaptations
installments[2010][1] += 1      # various Alice in Wonderland adaptations
installments[2011][1] += 1      # The Transformers: The Movie (1986)
installments[2011][8] += 1      # various Sherlock Holmes adaptations
installments[2012][6] += 4      # Spider-Man (1977), Raimi series (2002 - 2007)

enhanced_means = [numpy.mean(installments[year]) for year in y]
e_slope, e_intercept, e_r_value, e_p_value, e_std_err = \
    stats.linregress(y, enhanced_means)
e = lambda year: e_slope * year + e_intercept

if __name__ == '__main__':
    import matplotlib.pyplot as plt

    plt.scatter(y, original_means, color='k')
    po, = plt.plot([y[0], y[-1]], [o(y[0]), o(y[-1])], color='k')

    plt.scatter(y, enhanced_means, color='b')
    pe, = plt.plot([y[0], y[-1]], [e(y[0]), e(y[-1])], color='b')

    plt.legend([po, pe], ['Direct', 'Indirect (Remakes & Reboots)'])
    plt.title('Average Series Installment of Top 10 Grossing Films by Year')
    plt.xlabel('Year')
    plt.ylabel('Average Installment')
    plt.axis((y[0], y[-1], 1, 5))
    plt.show()