Precision describes how reproducible a measurement is, regardless of whether the observed value matches the true value. Measurements can be very precise even if they are not perfectly accurate. For example, if you weighed yourself on a scale that was offset by 5 kg, you would always be off by the same amount, but you would not be measuring your real weight. So despite not being accurate, the value you obtain could still be very precise. This will generally be true in situations involving a systematic error that is simply a constant offset from the actual value.