Here’s only one well-known LTCM technique: The algorithm recognized conditions the place the unfold had widened between the yield on probably the most not too long ago issued 30-year U.S. Treasury bond and that of an “off-the-run” bond, reminiscent of one issued a 12 months earlier (basically, a 29-year bond). Historically, buyers pay a modest liquidity premium for “on-the-run” bonds (which implies their yields are barely decrease than the off-the-runs) however, on condition that each investments represented the identical “risk-free” publicity to the U.S. authorities, the thought was that the unfold between the 2 shouldn’t deviate too far for too lengthy. So, at instances when the liquidity premium rose and the yield unfold widened, the LTCM algorithm would place a brief place (promote) towards the on-the-run bond and an extended place (purchase) on the off-the-run. Historical imply reversion would do the remainder.