In the absence of a true gold standard for non-invasive baroreflex sensitivity estimation, it is difficult to quantify the accuracy of the variety of techniques used. A popular family of methods, usually entitled ‘sequence methods’ involves the extraction of (apparently) correlated sequences from blood pressure and RR-interval data and the subsequent fitting of a regression line to the data. This paper discusses the accuracy of sequence methods from a system identification perspective, using both data generated from a known mathematical model and spontaneous baroreflex data. It is shown that sequence methods can introduce significant bias in the baroreflex sensitivity estimate, even when great care is taken in sequence selection.