Sounds like they used a Bruce Protocol for your test – typical for cardiology/CPET testing, basically useless for performance testing. You need more of a ramp protocol (e.g., increasing 3% every 3-5 minutes), with more samples to really get meaningful data. A large workload increase, like that in the Bruce, misses all the important physiological transitions, and without a longer step length/smaller work increment, it is difficult to match the changes temporally with heart rate.
