There are two issues you hint at but don’t state clearly.
- I don’t see people claiming that VP is not an effective training tool. In fact, I see plenty of people (myself included) state that VP can and does work well as intended.
- When used with consistent setup of the bike and trainer, you can get repeatable data for setting training zones and executing workouts.
- This leads to performing workouts with consistent info and can lead to effective training (when done as part of a properly executed training plan).
- So, there is no need to “justify” VP because most of us recognize how useful of a tool it really is. Ignore any dissenters because they likely misunderstand the reality of the tool and it’s proper use.
Accuracy (compared to “true power values”):
- VP is an estimate of power. It is based on testing performed at the TR facility, on a specific bike, with a specific tire, at a specific tire pressure, and a specific trainer (with it’s associated roller pressure). They measure wheel speed and power output from multiple power devices (power meter pedals, cranks and/or hubs). They correlate all that data into a “Power Curve” associated with that trainer. Your measured wheel speed is then “converted” to a Virtual Power value (an estimate of “real power”).
- Why state all of that? Because if you change any one of those variables, you may experience different results in the data reported. A single difference in tire pressure, roller pressure, actual tire used, etc will all affect the VP value.
- As such, when you survey VP users who get a power meter, you see reports of examples that show VP as higher, lower, or roughly equal to the real power data. In many ways, it’s much ado about nothing. The desire to have VP be “accurate” is largely secondary to the desire for it to be consistent (the real need for training purposes).
- The issues arise when people try to make or expect VP to be “accurate”. It may well be so, but there is no guarantee without verification using a proper power meter.
- Your case is great and interesting that it might have matched (I question your method and conclusion). But we can find many examples where that is not the case. No single example proves that VP is “accurate”, because we can see many more where that is not the case.
In your case, you are comparing info from different points in time and that could well lead to variation. I wouldn’t take that as any validation. Even if you do back to back tests, you introduce the before/after timing and other variables that are likely to skew results.
The only good way to do a VP to Smart Trainer comparison is to use a 3rd device (power meter installed on the bike) and run through a range of power outputs. Correlate the data from each trainer to the power meter (as the “measuring stick”) and see how they compare across a range of power outputs. They may align at one point or many, or they may be offset. No one knows and it is pure guessing without proper testing.
Be happy that you got good training from your VP use and move on to the smart trainer use. There is not much benefit in trying to back validate your VP training. It was what it was, and the numbers are largely irrelevant now.