RUNNING QUESTIONS
Why is my VO₂max dropping when I am running faster?
Your finish time is an observation. A watch or app VO₂max value is an estimate. When they move in different directions, review the evidence behind each before changing your training.

An estimate is not a laboratory measurement
VO₂max describes maximum oxygen uptake. A wearable estimates it from available measurements rather than measuring your breathing gases directly. Apple, for example, uses heart and motion sensors plus profile information for its cardio-fitness estimate. That explains why it is useful to know which service produced a displayed number.
Keep trends from different providers separate. A 50 in one app and a 52 in another are not automatically two tests on the same scale under the same conditions. Record which service produced the value, which activity it used and whether the number is a single-activity result or a longer-term summary. Do not average them into a more precise-looking score.
A faster 5K does not settle every question
Imagine running 5 km in 25:00 last month and 24:30 this month. Your time improved by 30 seconds, or 2%. Meanwhile, a displayed VO₂max estimate moved from 50 to 49, a 2% decline. These are illustrative numbers, not results from a Runome user. The percentages refer to different quantities and do not cancel each other out.
The faster time tells you what happened over that recorded distance. It does not establish that the route, timing, effort or conditions were identical. Likewise, the estimate alone cannot explain why you ran faster. Compare the route and elapsed time first; then look at the accompanying heart-rate record and how demanding each effort felt.
Check what changed around the estimate
Check profile values for mistakes rather than changing them to improve the score. Note sensor or device changes, missing samples and different activity types. Then compare route and conditions: Polar explicitly lists heat, altitude and terrain among factors affecting its Running Index. This is a reason to examine context, not to assume every provider handles those factors in the same way.
Look for a repeated pattern over comparable activities. A change confined to one unusually hot or poorly recorded outing answers a different question from a sustained change across similar runs. Keep both in the history with their context. Removing inconvenient observations just to make a trend rise makes the history less useful.
Bring the evidence together in Runome
Runome lets you inspect individual runs and follow available VO₂max, pace, efficiency and training-load trends. Use those views to ask whether performance and effort are changing together. Data availability depends on the recording. A missing estimate is not zero, and a trend in the app is not a substitute for a laboratory assessment.