Thanks, Michael, for the interesting example. But there are some things it doesn't take into account:
During every hour of brokered content your listeners are forced to go elsewhere. You lose continuity. When the next show starts the audience has to be built from scratch. Let's face it, not every show is "appointment radio" and lead-ins are important.
Then there's the element of "discovery." Listeners looking for something of interest to listen to on the weekend may get hooked on your station if there's real content there -- not someone selling snake oil. Discovery is important in the tech world, as evidenced by Microsoft's recent decision to give away it's operating system to manufacturers of <9" tablets IF they set the default search to Bing and the home page to MSN. That's something the end user can easily change but it's the first step toward forming a user habit that pays off big in the long run.
Also, the weekends are valuable for grooming new talent, and they're an important way to promote the weekday lineup. If the time has value, as evidenced by the direct marketers being able to make money from it, then every hour of genuine talk programming on the weekends is in effect an ad for the station.
I suppose call letter recognition was more important in the diary days than with PPM -- the value of being a "favorite station," reinforced by listeners staying tuned all weekend -- would cause participants to list the station(s) that were foremost in their minds. That might have been an incentive for stations to not throw away the weekends for a quick buck. That's a shame because I think the "favorite station" benefit still exists -- it just can't be seen directly in the minute-by-minute weekday figures.