Arbitrage: I completely agree. I'm embarrassed to admit that until three or four weeks ago, I only cared about average correlation. I didn't even care how it did per-era. And I really believe I just got lucky in my model design. And this is what I teach, right? Have a good approach prior to even looking at the data. Set up in your head how you're going to go about this, and then at the end, if it's good, run with it. I think I just got lucky, and I'm not afraid to admit that. I'll take my six in the top 50 staked and run with that for now. We'll see how these new iterations come about. So here's a fun one: if you could have a billboard with anything on it, what would it say and why?