Read on Twitter
so how do the merger rates for BHBH, NSNS and BHNS from #GWTC2 by @LIGO and @ego_virgo ( https://arxiv.org/abs/2010.14533 ) compare with theoretical predictions? I show that here in a plot from my paper on this in prep.: 
the shaded colored areas in each of the panels shows the observed 90% confidence intervals of the BHBH, BHNS and NSNS merger rates from https://arxiv.org/abs/2010.14533 by @LIGO and @ego_virgo
I plotted the old rates from #GWTC1 with gray horizontal lines in the background as well (from https://arxiv.org/abs/1811.12907 ). You can see that the new results by #GWTC2 have already amazingly increased the merger rate estimates! Incredible work from @LIGO and @ego_virgo!! 
Only the BHNS rate estimate has stayed the same since there has not been a confident detection for BHNS in #GWTC2 or #GWTC1 yet... This is also why the rate for BHNS is given as an upper limit! 
Now, what else does this figure show:
The points in each of the panels show the predicted merger rates of BHBH (top panel), BHNS (middle panel) and NSNS (bottom panel) based on simulations I ran with the binary stellar evolution code #COMPAS
The points in each of the panels show the predicted merger rates of BHBH (top panel), BHNS (middle panel) and NSNS (bottom panel) based on simulations I ran with the binary stellar evolution code #COMPAS
On the x-axis I am showing the predicted rates for 12 different population synthesis models. Where I vary some of the key uncertain assumptions in the stellar evolution modeling such as the common-envelope phase, the supernovae (and supernova kick) and mass transfer.
but that's not all! since these mergers might originate from binaries formed at cosmological distances, we need to model how many binary systems form over the age of our Universe and what metallicity they are born with. Something known as "star formation history" 
This star formation history is also uncertain, and so I have 28 different models for it. These are shown in the figure by different lines. In combination with the 12 COMPAS models this gives me 336 (12x28) total models !! Or many TB of BHBH, BHNS and NSNS data. 
