python said:because for hemmingson not to suspect or being aware of the measurement bias would be beyond believable for a professional involved in anti-doping. besides, clearly, it is a convenient position to take today. also, the astronomical early limit of 18.5 must have considered the bias.
Why do you assume that there's anything "sudden" about the knowledge of HemoCue's measurement errors? What do you mean by "it's a convenient position to take"? Ola Rønsen, Damsgaard and Hemmingson, all of whom have commented on the measurement errors involved in the test results from Lahti 1997, are not taking any positions regarding HemoCue, they are simply referring to established knowledge in the field. Hemmingsons's main role, besides being the co-author of the letter from Ola Rønsen, is being responsible for testing in Thunder Bay 95, member of the FIS antidoping committee as well as having been team doctor for Sweden.
Personally, I don't know exactly how much knowledge of HemoCues measurement errors there were at the time, in 1997, but it wasn't necessarily that vast. There's a reason it's not in use anymore, and there's a reason hemoglobin experts agree that it's not reliable enough to be used either in a doping context nor in medicine.