Odds of Two Radio Stations Playing the Same Song at the Same Time?

Let’s say that in the Anytown, USA radio market there are two radio stations playing all-Christmas-all-the-time for the month of December. Assume that each station has exactly the same playlist of 1,000 songs. Assume further that each song lasts exactly 3 minutes; and that for each hour of each programming day, 15 minutes are given over to commercials, weather, news, inane banter from the DJ’s, etc. Assume that I listen to one station or the other for eight hours per day, and when one station plays an aggravating song, I switch to the other; I stay with it until they play an aggravating song, whereupon I switch back to the original station until they play an aggravating song, ad infinitum.

With these assumptions, what are the odds that, at any given time, I could switch stations and find the exact same song being played on the other station? (Assume that they may start at different times: so for example I may hear the melodius strains of Jose Feliciano’s Feliz Navidad come on, head to the radio, switch stations, and find the same song playing there (albeit at a different point in the song).)

I may be wrong here but I think you’ve over complicated the matter.

If you hear annoying a song on one station and change then the chances of the other station playing the same song is 1 in 1000 if they’re playing music. Since a quarter of each hour is given over to non-music you’ve got a 3 in 4 change to catch any music.

I think that means you’ve got around a 1 in 1333 chance of hitting the same song, but I could be wrong in that last step.

How do stations decide the next song to play? You need a model for that in order to answer this sort of question, and it’s not going to be simple.

I don’t know where you can get the playlists for satellite radio, but I can confirm that on more than one occasion I have heard the same song playing on different channels on Sirius. Whether this is intentional or just more “evidence” of Sirius’ narrow programming, I don’t know.

I think he’s saying that it’s random.

Using that logic, you’d get 1/1000 * 3/4, or 3/4000 chance.

If I understand it right, Sirius does this on purpose. I’ve often seen, “Now Playing On Channel 1 Hitz” scroll across the screen during a song.

Lucky you having radio stations with 1000 songs on their playlists. Here the average playlist is about 25 songs (and the more popular ones get more weight so it’s more like 5 songs they play constantly and 20 songs they play occasionally, or so it seems). Over and over. All day long. Many times have I switched from one station to another to hear the exact same song.

That’s definitely a false model. If it were true, radio stations would occasionally play the same song twice in a row, and (barring a moronic DJ’s intervention) that never happens.

I once tuned into my radio, and WWEG in Hagerstown, WZBA in Baltimore, and WBIG in Washington were all playing the same song, “Gypsy” by Fleetwood Mac.

Which is the same as the 1/1333 that I got to in my first post.

Yeah, in reality these things are not random. Certain songs are played at certain times, ad breaks are matched up (making it more likely you’ll change from music to music) and I believe that a day’s playlist isn’t anywhere near 1000 songs. I mean that’s 66 hours of music.

Most stations just have enough songs to vary the content for two hours or so, around 30 three minute songs. That’s about a one in forty chance to hit the same song. If there was one song you hate you’d be changing around 4 times a day you should notice the same song playing about once every ten days. Half that number if you hate two of the songs and so on.

On preview, dammit, Rigamarole said some of what I said …

They could still use random song selection. On my Winamp playlist, for instance, I randomize my songs, but Winamp will not play the same song twice until it has cycled through every other song once.

Other have given you the odds of them playing the same song. You seem to want to add a complexity of knowing how often you’ll catch this happening.

To do this, we need a probability of an aggravating song, or how many time an hour you’ll be aggravated.

No, it’s not. [/pedantic]

True, I guess it’s not exactly the same, I should have been more precise.

This past Friday, I was switching stations, and two different ones had the same song, same artist, same version, but out of sync by about 20 seconds. It was eerie listening to one and switching to the other. Sorta kinda like being in a time machine.

A few years ago, I heard not only the same same, same artist, same version, but at the same time as well. Flipping back and forth I tried to discern if one might have been even a second ahead of the other, but they seemed to be exactly in sync. I started to wonder if both buttons were on the same station, so I kept going back and forth until after the song ended on both, and different DJs started talking about different things. And the following songs were different, so it was not a case of simulcasting, which would have used a single DJ anyway. Very weird.

I’ve heard that quite a few times, generally on the two rap stations or rock stations near me, but only once in my life on two different Radio Stations from different genres. The song was Gnarls Barkley’s Crazy, and it was playing on the rock station AND the rap station at the exact same time. I don’t really know many other songs that could even work that way.

A few years ago, the song Butterfly Kisses by Bob Carlisle was getting heavy airplay on the Christian stations, the country stations, the Top 40 stations, and the Light Hits stations.

It is true that for any one instance there is about 1/1000 chance of an overlap. However if you listen for 8 hours then there is a much larger chance of an overlap somewhere during that day (which is closer to the OP question? ) The main determinate is how many times you switch (as about 50% of the songs on my station is annoying say you switch 5 x an hour or 40X a day. With about .001 chance of catching a match each time then there is about a 4% chance of a match during the day. However using a more realistic 100 song playlist (esp round Christmas) then you should catch a match 33% of the time during that day (e.g. one every 3 days) using the binomial theorem.

You are also neglecting the influence of requests or other outside forces altering the playlist order.

Two examples: on a classic rock station on Tuesday, I can occasionally be nearly 100% certain what the next song will be, given the following outside forces: (1) Tuesday is “two-fer” Tuesday, and (2) “Living Loving Maid” will always follow “Heartbreaker”, and “We Are the Champions” will always follow “We Will Rock You.” I listened to Arlo Guthrie’s “Alice’s Restaurant” in its entirety on Thanksgiving morning to inoculate myself, but it was no use: that song – seventeen-plus-minutes long! – was on two different radio stations during a 1.5-hour drive up to Philly later that day. Coincidence? Hardly.

An office worker in your metropolitan area who loves Feliz Navidad might conceivably request it on every station playing carols just before getting in the car and heading home for the day, on the off chance that they might get to hear the song once (or twice!).

A few years ago there were three Country music stations in my area (now only 2). I was driving in my truck and the new (at the time) Toby Keith song came on. I didn’t care for it, so I changed stations. The same song, about 20 or so words ahead of the other station. Changed it again. In around 50 seconds, the same song came on. All three had thw same song on for a short time.