Big D - you are describing apples to oranges and not a 'downside' from a purely correct technical standpoint. YM to CL is a good, solid statistical correlator, and I have no doubt that you do indeed successfully trade one based upon the relative value price action of the other (the spread). I have clients who hire me just to uncover those kinds of exploitable relationships for directional plays. (there are better intra-market correlators for each of those, by the way)
There have been times in the past when the Yen and the third-month Eurodollar were golden, as was the CD vs. CL and the MX vs. CL.
But from a statistical fitness standpoint the cointegration traits for that particular combination (the beta plot study on your Bloomberg) suck - no way I'd trade spread trade that combo with simultaneous execution of both legs both ways. There are much better spread combinations (some not so obvious) to risk capital on from a pure spread trading standpoint (simultaneous execution of all legs).
So, there is a tangible difference between correlation trading and pure spread trading.
There have been times in the past when the Yen and the third-month Eurodollar were golden, as was the CD vs. CL and the MX vs. CL.
But from a statistical fitness standpoint the cointegration traits for that particular combination (the beta plot study on your Bloomberg) suck - no way I'd trade spread trade that combo with simultaneous execution of both legs both ways. There are much better spread combinations (some not so obvious) to risk capital on from a pure spread trading standpoint (simultaneous execution of all legs).
So, there is a tangible difference between correlation trading and pure spread trading.
