Strip the Play by Play file so we are only left with the play records. Players On Ice not handled.

stripPlaybyplay.py

# takes 2 seconds to run per infile
# source for infile: http://www.nhl.com/scores/htmlreports/20152016/PL020666.HTM

from bs4 import BeautifulSoup

print("Parse start")

first_game_id = 1  #+665
last_game_id = 668  #-2

subseason_id = '02'
datafile_id = 'PL'

replace_str1 = "<br/>"

header_row = ('<table border="1" cellpaddin="4" cellspacing="0">'
            '<tr bgcolor="#CCCCCC">'
            '<td>EVENT_ID</td>'
            '<td>PERIOD_CD</td>'
            '<td>STRENGTH_CD</td>'
            '<td>PLAY_TM</td>'
            '<td>dummy1</td>'
            '<td>REMAIN_TM</td>'
            '<td>EVENT_CD</td>'
            '<td>PLAY_TX</td>'
            '<td>dummy2</td>'
            '<td>ASSISTS_TX</td>'
            '<td>dummy3</td>'
            '</tr>')

for int_game_id in range(first_game_id, last_game_id+1):
    game_id = str(int_game_id).zfill(4)
    print(game_id)

    sourcefile = "C:\\Users\\TOM\\PycharmProjects\\downloadNHL\\datafiles\\{d}{ss}{g}.HTM".format(d=datafile_id, ss=subseason_id, g=game_id)
    targetfile = "C:\\Users\\TOM\\PycharmProjects\\downloadNHL\\datafiles\\stripped_{d}{ss}{g}.HTM".format(d=datafile_id, ss=subseason_id, g=game_id)

    with open(sourcefile,'r') as infile, open(targetfile,'w') as outfile:
        soup = BeautifulSoup(infile, "lxml")
        tableRow = soup.findAll('tr', {'class': ('evenColor' , 'oddColor')})
        outfile.write(header_row + '\n')
        for r in tableRow:
            tableRowData = r.findAll('td')
            onice_flag = False
            outfile.write("<tr>\n")
            for d in tableRowData:
                elements_in_td=len(d.contents) # the TD itself is made up of possibly more than one element
                if onice_flag is False:
                    for i in range(0, elements_in_td):
                        if str(d.contents[i]).find("table") >= 0:
                            onice_flag = True
                            j=0
                            break
                        else:
                            printline=str(d.contents[i]).strip().replace(replace_str1,'')
                            outfile.write('<td>' + printline + '</td>\n')
                else: # we are now in the OnIce Players section
                    for i in range(0, elements_in_td):
                        pass
            outfile.write('</tr>\n')
        outfile.write('</table>\n')

print("Parse end")

a suggestion, partially tested,

import re

ifile = somepath + 'PL020668.HTM'
s = reduce(lambda a, b: a+b, [l.strip() for l in open(ifile).readlines()], '')
all_paragraphs = s.split('evenColor')
for a in all_paragraphs:
 onice_list = re.findall('title=(.+?)>', a)