::  Posts  ::  RSS  ::  ◂◂RSS  ::  Contact

Effect of Catapult

December 15th, 2016

Catapult is a contra dance weekend with the goal of making it easier for bands to start getting booked for dance weekends, festivals, etc. They get something like six bands and six callers together in Atlanta over memorial day weekend, and they encourage people who are involved in booking bands and callers elsewhere to come dance. This makes some sense: dance weekends are generally once a year, so they're generally pretty conservative in their booking because a bad band or caller could do a lot of damage. Instead you have one weekend that goes out on a limb with less proven performers, but they have a whole lot of different performers and dancers know what they're getting.

They're currently near the end of their application process, with applications due 12/29, and people were asking whether it was worth applying. Now, when I went with the Free Raisins in 2013 we had a blast, so I'd definitely recommend it as a fun weekend. But is it worth it from the perspective of it's goals? Will it get you booked for weekends and things?

This is a hard question to answer, because it could be that Catapult is just identifying people who would soon be getting booked for these things otherwise [1], as opposed to making the difference between getting booked and not. On the other hand we could at least look at bookings for bands and callers post-Catapult.

I have data [2] on how many weekend/festival gigs bands played in 2014, 2015, 2016, and part of 2017 . So if we go through the bands who played Catapult in 2012, 2013, and 2014, how many gigs do we see?


The bands that played Catapult in 2013 seem to have gotten booked for many more things than the ones who played in 2012 or 2014. What's going on? Is it just that the dances I have data from are when the 2013 cohort was at its best? Let's look at counts by "years-since-Catapult" to correct for this:


In the years where we have comparable data, the 2013 cohort was booked much more widely than the 2012 or 2014 one. It's not just one outlier band distorting the numbers either—if we remove the most booked band from the 2013 cohort we still have:


Or we can look at the number of bands in each cohort with at least one gig in that year (each year had six bands):


That's pretty strange! Is it just that Catapult 2013 had especially good musicians, or that in 2013 lots of bookers came to Catapult? One factor is that La Banane Enchantee (2012) lost a band member to the UK. They were the most popular band of their cohort, winning the audience vote the only year Catapult did voting, and if they had continued to be playing gigs in the US they might have played a lot of them. But that's not enough to explain the difference.

I don't have as good data for callers, just 2016, which isn't even a consistent distance out from the three cohorts, but lets look at it. If we see many more callers from the 2013 cohort getting gigs then that suggests (a) Catapult was helping and (b) more bookers were at the 2013 event, while if we don't then it's more likely that 2013 just happened to select good bands.

Interesting! Both the 2012 and 2014 cohorts had substantially more bookings for 2016 (four and two years out) then the 2013 cohort did (three years out). So it's probably not that there were more dance bookers at Catapult 2013.

Another thing we could look at is whether the same weekends tend to book Catapult talent. If there's a lot of overlap, that's more reason to think catapult had an effect. Here are dances by their frequency in the two lists above:

(Keep in mind that I only have caller data for 2016, my 2017 band data is incomplete, and I don't have any 2013 data. With those included there would probably be more overlap. This is also why the number of slots is kind of strange.)

Weekends by how many catapult performers were booked in the sample:

number of catapult bands and callers booked number of weekends

This is 73 slots out of approximately 1,088. [3] With ~6% of slots going to Catapult alums, it doesn't sound like Adirondack, FolkMADness, etc being nearly half Catapult alums is a coincidence. Except that we still can't tell the difference between those weekends selecting people the same way Catapult does, and Catapult helping them get selected.

[1] Some people say this is what top colleges do.

[2] For this post I'm using the raw data in Dance Weekends, Festivals, and Long Dances, and I'm not counting things like the Brattleboro Dawn Dance or other gigs where bands play much less than you do at a typical dance weekend.

[3] 1,088 comes from totalling:

Comment via: google plus, facebook

More Posts:

Older Post:

Newer Post:

  ::  Posts  ::  RSS  ::  ◂◂RSS  ::  Contact