We don't often get a good look at the C3 numbers, otherwise known as the "real" ratings in terms of setting advertising rates. They measure viewing of only the commercials both live and on DVR for three days beyond the airdate. So I feel obliged to mention even this little nugget. Yesterday, MediaPost's Wayne Friedman looked at the premiere week C3 ratings on the big four.
The ultimate conclusion is the same as it usually is when we've seen the C3 numbers in past years: Live + Same Day is still a better indicator than the networks would like to have you think.
Friedman notes that two networks (Fox and ABC) did a tenth of a point better in the C3 numbers for the entirety of premiere week than in the Live+SD. The other two networks (CBS and NBC) did a tenth of a point worse. Add those together and the combined big four rating is exactly the same in Live+SD and in C3 during premiere week. And the combined big four year-to-year trend that week was exactly the same both in C3 and in live program ratings: -17%. So while DVRing beyond the airdate is on the rise, it's not really making the ratings any more favorable monetarily speaking than the live numbers suggest. Maybe -17% is not the reality of the shows' "true audiences" across all platforms, but it's pretty much still the reality of the shows' TV advertising audiences.
He also looks at some of the shows whose gaudy DVR numbers have been bandied about in recent weeks. Revolution is perhaps the most visible example. And, indeed, that extreme DVRing has led to a lift when comparing Live+SD with the "real" numbers: from 3.4 -> 3.8. That's an increase of... 12%. So that 68% gain for Revolution from Same Day to +7 is only about 12% in terms of the actual advertising rate-setter. Elementary gains 45% in L+7 but just 10% (3.1 -> 3.4) in the "real" numbers, and Last Resort gains 45% in L+7 but just 9% (2.2 -> 2.4) in C3.
What does this mean for the relative landscape? As others have noted about DVR numbers, it usually just means the rich get richer and the poor get poorer. Does it help illuminate marginal cases? Possibly, but it's also not nearly as big a deal as the big +7 lifts would have you think. For a marginal show, even one that appears very heavily DVRed, you probably can't count on a "real" lift of more than about 10%. In today's landscape, a 10% lift for a marginal show usually only means a tenth or two tenths, and that's rarely if ever a decisive game-changer. Maybe you could say it was what put something like season three of DVR-happy Fringe ahead of other bubblers like Human Target and The Chicago Code. But even that's not such a clear difference-maker that we can be sure.
Bottom line: for all the DVR numbers that are thrown about, there are only two things that define the comparison between our usual Live+SD currency and the "money" numbers: commercial skipping within the same day window (dropped from L+SD -> C3) and commercial viewing beyond same day but within the three-day window (added from L+SD -> C3). Both of those things increase as DVRing increases, and it appears they're still increasing at about the same rate. Shows that are massively viewed via DVR beyond the first day may be exceptions to that rule, but even they are not significant exceptions; they'll probably only go up by about 10%. And 10% is rarely the difference between cancel and renew, much less between hit and flop.