[symmetricsaurus]

The result is only significant to 3.39 sigma. Pretty good chance this is just a random fluctuation that will go away if you look at more galaxies.

[BlackFly]

This was also my thought, since selection bias seemed like a good explanation as well. The survey only covers 220 square arcminutes out of a total of about 148 million. Interesting data point though, definitely seems like something that one should measure!