Ok tried to only count the "porn" word occurences in home of the sites (in any place, not just in meta), and some newspapers got more P words than porn tubes:
43 xvideos.com porn: 52
60 reddit.com porn: 12 OK after all
57 xhamster.com porn: 13
84 pornhub.com porn: 236
100 dailymail.co.uk porn: 9 ** newspaper
104 livejasmin.com porn: 5
106 xnxx.com porn: 35
108 aol.com porn: 2 ******* AOL
113 redtube.com porn: 47
120 youporn.com porn: 211
209 tube8.com porn: 22
257 hurriyet.com.tr porn: 3 ** newspaper
326 tmz.com porn: 15 ** well...
336 hardsextube.com porn: 45
345 chaturbate.com porn: 2
358 motherless.com porn: 12
359 beeg.com porn: 11
352 youjizz.com porn: 69
382 repubblica.it porn: 11 ** newspaper
423 ixxx.com porn: 527
419 nydailynews.com porn: 7 ** newspaper
439 gazeta.pl porn: 4 ** newspaper
505 chip.de porn: 2 ** newspaper
534 slate.com porn: 3 ** newspaper
...
With sex this is gets too many false positives.. sex word can be useful perhaps if ' sex ' (with space before and after) and only from >3 found.
So let me run a script which checks RTA and counts porn and sex words:
43 xvideos.com +RTA +porn: 54
57 xhamster.com +RTA +porn: 17 +sex: 8
60 reddit.com +porn: 14 ??????
84 pornhub.com +porn: 238 +sex: 10
106 xnxx.com +porn: 33 +sex: 63
100 dailymail.co.uk +porn: 9 +sex: 9 *******
113 redtube.com +porn: 47 +sex: 5
104 livejasmin.com +porn: 5 +sex: 53
108 aol.com +porn: 2 *******
120 youporn.com +RTA +porn: 209 +sex: 4
209 tube8.com +porn: 22 +sex: 14
257 hurriyet.com.tr +porn: 3 *******
326 tmz.com +porn: 14 +sex: 10 ??????
336 hardsextube.com +porn: 49 +sex: 23
358 motherless.com +RTA +porn: 6
359 beeg.com +porn: 11 +sex: 8
382 repubblica.it +porn: 11 *******
352 youjizz.com +RTA +porn: 69 +sex: 8
345 chaturbate.com +RTA +porn: 2 +sex: 4
419 nydailynews.com +porn: 7 +sex: 9 *******
423 ixxx.com +RTA +porn: 527
439 gazeta.pl +porn: 4 *******
505 chip.de +porn: 2 *******
511 adultfriendfinder.com +RTA
533 myfreecams.com +RTA
534 slate.com +porn: 3 *******
587 drtuber.com +RTA +porn: 72 +sex: 11
625 bongacams.com +RTA +porn: 14
667 mynet.com +porn: 2 *******
670 focus.de +porn: 26 +sex: 6 *******
688 independent.co.uk +sex: 5 *******
715 news.com.au +porn: 5 +sex: 5 *******
719 porn.com +porn: 132 +sex: 5
723 sfgate.com +sex: 4 *******
725 sex.com +porn: 183 +sex: 27
735 haberturk.com +porn: 10 ******* (topless perhaps)
776 privatehomeclips.com +RTA +porn: 17 +sex: 8
784 thechive.com +porn: 5 ******* (bikini most)
804 nuvid.com +RTA +porn: 95 +sex: 37
803 bloglovin.com +porn: 24 *******
812 spankwire.com +RTA +porn: 27 +sex: 6
830 smh.com.au +porn: 7 *******
864 xtube.com +porn: 14 +sex: 6
866 cam4.com +RTA +porn: 9 +sex: 4
947 xhamstercams.com +RTA +porn: 13 +sex: 104
963 keezmovies.com +RTA +porn: 262 +sex: 13
962 bravotube.net +RTA +porn: 38 +sex: 55
977 anysex.com +porn: 26 +sex: 52
1022 video-one.com +porn: 239 +sex: 15
1036 pornmd.com +RTA +porn: 56
1039 nypost.com +porn: 7 *******
1064 yourlust.com +porn: 47 +sex: 13
...
It requires some manual checks. Any ideas how to improve easily (I use other system I get % of male vs female but this is not free/public data, I wanted to just use this free csv).