The primary concern of this contribution is that accreting millisecond pulsars (AMXPs) show a much larger amount of information than is commonly believed. The three questions to be addressed are: 1. Is the apparent spin torque observed in AMXPs real ? 2. Why do we see correlations and anti-correlations between fractional amplitudes and timing residuals in some AMXPs ? 3. Why the timing residuals, the lightcurve and the 1Hz QPO in SAX J1808.4\(-\)3658 are related ?