For example, this chart displays the minute by minute “normal” switching average of all stations in the top 40 US metros across a statistically-typical weekday.
This chart shows a sliding four minute window (similar to the opportunities to switch during any song). Media Monitors then tallied all the switchers to another radio station (switch off is not counted) as a proportion of the average audience during these four minutes. The process was done and aggregated for all music stations in all these markets.
That is the black graph on the chart with its scale on the left axis.
Next to that, you will see the percentage of stations in these 40 markets that are in a commercial break at each minute, as one of the obvious correlation factors and for reference only.
This is plotted as a green shade on the chart with its scale on the right.
It's noteworthy that not as many people switch stations for commercials during the workday as do during the drivetime periods.
Since this 1% to 8% of the audience is leaving one station for another, when considered broadly, the well-programmed station often gains as much as it loses if not even more by careful placement of programming based on the behavior of their shared-cume competition and listeners, but the average is about 4% in and 4% out in the majority of minutes calculated.
The highest points in this switching, 7-8 am and 4-5 pm, likely occur because of these audience members changed listening location.
If you discovered your listeners doing that, what could you do to change that behavior?