The power reading on a Flow isn’t precise within a session
Mine’s 11 years old and I find that. However, the way it fluctuates is reasonably predictable, and hasn’t changed over the years, so if you factor that in it’s not too bad.
But, if you do a ‘coast down test’ at the start of the session on your current trainer, set it to be the same every time, and use speed as proxy for power, you get the same result. You have to do a ‘calibration’ every time you use a Flow, which is just a coast down test where it counts the seconds for you…
I doubt it is accurate enough for a pro
I’d bet that powertaps etc are nothing like as accurate as folk think. There’s a difference between resolution and accuracy that fools people too often. I’ve done 30+ years in Test Equipment and it still amazes me how people believe what they read on a screen without thinking about just how that number was generated.