This is true! This reasoning was explored in an excellent paper by Blaes et al. in 2002 [1]. (Incidentally, one of the authors on this paper has the wonderful

This is true! This reasoning was explored in an excellent paper by Blaes et al. in 2002 [1]. (Incidentally, one of the authors on this paper has the wonderful name Aristotle Socrates.) Due to a very interesting dynamical phenomenon called the Kozai-Lidov mechanism, if the orbit of the third body is at high inclination relative to the inner two, it can drive the inner orbit to very high eccentricities and cause them to merge much more quickly. So even though you would form triples fairly frequently, the third body would cause the other to rapidly merge, so you would typically expect only to see binaries. On a more shameless note, I also explored this in one of my first papers [2]. [1]: http://adsabs.harvard.edu/abs/2002ApJ...578..775B http://adsabs.harvard.edu/abs/2002ApJ...578..775B [2]: http://adsabs.harvard.edu/abs/2014MNRAS.439.1079A http://adsabs.harvard.edu/abs/2014MNRAS.439.1079A