TV shows are judged by their Neilsen rating, measuring the proportion of TV viewers who watched a show. Certain popular episodes of shows attain much higher ratings than other episodes. For instance, “The Contest” episode of Seinfeld, the wiki page says, “is considered to be one of the best Seinfeld episodes, due to it winning several awards, achieving high ratings and postitive reviews by critics.”
How would anyone know an episode was going to be good until after it had aired? Wouldn’t the highest rated episodes be those for which the advertising beforehand had been best received? I’m not seeing the correlation between quality and high viewership.
ETA: Are reruns during the same week of initial broadcast counted into the ratings?
If you’re talking about the first time a regular episode of a TV show is aired, why do the episodes thought of as the “best”, quality wise, seem to have higher ratings, I think that’s a faulty premise. Usually its final episodes (Fugitive, MASH, Cheers), episodes where its well known in advance what’s going to happen that night, and the episode is intriguing (Lucy’s baby, Who Shot JR), but otherwise I don’t think it holds true- the Seinfeld episode may be an anamoly.
Two Beverly Hillbillies episodes for example to get some of the highest ever ratings were not known as the best epsiodes (“The Giant Jackrabbit” and “The Girl From Home”).
Ratings never include syndication. Only recently have they included viewings on DVR after the original run, but even those are only counted if less than one week from the first showing.
Ratings are for advertisers, not viewers or programmers. They are a measure of how many eyeballs were available to view the ads. That’s why internet viewing doesn’t count, why reruns are counted separately, and why syndication is counted separately. Each has a different set of ads.
Program popularity is just a gauge to help networks figure out how much they can charge advertisers. How many people watch is important, but so is the age range of the viewers. Shows with large percentages of older viewers can’t charge as much for ads as those with lower overall viewers but more young viewers that advertisers prefer. That’s the only metric anyone really cares about. Everything else is a red herring for public consumption.
I think it’s #1, since as has been pointed out, #2 doesn’t hold true. “This Monday, on a special episode of Medium, Paris Hilton guest-stars as a nuclear physicist with a deadly secret in her past.” If the advertising is heavy enough, and the premise appealing enough, more folks will tune in.
Back pre-Internet, everyone got TV Guide, which highlighted certain episodes of shows with a quarter page or so, with picture and extended write-up. I have no proof, but I expect these improved ratings. It was not a guarantee of quality - I suspect they had guidelines about highlighting all shows at some time or another, but in my recollection they were often better than average.
A lot of shows would also get special advertising and publicity, like the last episode of MASH or Who Shot JR? Things were a lot easier to publicize back when there were only three networks.
Audience retention – there is a tendency for better episodes to have better ratings not because people somehow know to tune to them, but because once someone begins watching, that viewer is less likely to change the channel or turn off the set.
Just a point about the example in the OP, it mentions being considered one of the best because it won awards. For awards like the Emmys the show has to be submitted for the award by the producers (or network I’m not sure which). The audience or the academy does not pick which episodes get put up for the awards. The members of the academy are not required to watch every show of every series to come up with the nominees.
Also if you read further in the article it says that the first airing of the episode had OK ratings. The repeat of the episode had the highest rating for the show so far. So in this case people watched because of word of mouth. It had already aired.