Race Result Analysis in N+One: Tagging Outliers Without Distorting Your Trend
5 min read
Learn how to tag race outliers in N+One without deleting data or distorting your performance trend, then make one clear training decision.
On this page
Photo by Rille Camera Strap on Unsplash.
Start with rules, not eyeballing. Tag race outliers, keep the raw result, and protect your trend before you adjust training.
Race results are noisy because the day is not just a fitness test. Course demands, tactics, weather, gear issues, illness, and role can all change the result without proving that your form changed. In N+One terms: the training system around a rider drifts; outlier tagging keeps your form signal clean.
Photo by Markus Spiske on Unsplash.
An outlier is not just a race you disliked. It is a result that breaks your expected pattern and has a clear reason outside normal training change.
Set the rule before you look at the result. For example, use a value more than 3 standard deviations from a rolling 30–90 day mean, or a sudden shift tied to a documented non-training event.
Your baseline matters more than one finish place. Compare the result with your recent pattern, then check the workout detail in single-session ride analysis.
If the result looks extreme but the day also had a crash, severe weather, or a mechanical, the tag is more likely to stand. If the only reason is disappointment, keep the point and write a plain note.
The aim is not to hide bad data. The aim is to keep longer trend reads in N+One from being bent by one odd day.
Set the rule before opening the race file.
Compare against a rolling 30–90 day mean.
Flag values more than 3 standard deviations from that mean.
Add the non-training event that explains the tag.
Keep unproven cases as normal results with notes.
This keeps the trend clean without rewriting the race.
In N+One terms: an outlier is a result that breaks your system’s expected response without a plausible training or health driver.
Photo by CLEMENT Ader on Unsplash.
Do not delete the file. Raw results are part of the record, even when they should not steer the short-term trend.
Add a metadata tag such as outlier=probable, then add a short reason code. Good reason codes include mechanical, illness, severe weather, crash, pacing, or tactical role.
Keep the tagged race visible in logs, but run the planning trend without it. This mirrors the basic habit behind clean cyclist data workflows.
You can also compare the race with a nearby workout. A side-by-side view helps separate one-day noise from a repeat pattern across similar efforts.
If the tagged event was an A-race, keep the race role clear. The tag should not erase the build, the taper, or the planning work behind race tagging in N+One.
Keep the raw activity file intact.
Add outlier=probable or outlier=confirmed.
Write one short reason code.
Exclude the tag from short-term trend fits.
Keep the race visible in the full log.
One tactical email with training ideas and product updates. No spam — unsubscribe anytime.
Define outlier rules before inspecting results to avoid bias (e.g., magnitude relative to recent trend or documented non-training events).
Over-tagging is the quiet risk. If you remove every awkward race, the chart starts to show your hopes instead of your history.
Require two independent reasons before you tag. A slow result plus documented illness is stronger than a slow result alone.
Do not tag a race only because power, pace, or placing was lower than expected. First check whether the same pattern appears in two-workout comparisons.
Use full-data and tag-excluded trend fits side by side. The full-data fit protects the record, while the excluded fit gives the cleaner short-term planning line.
When you are unsure, keep the point and annotate it. A noisy point is less harmful than erasing a real shift in form.
Require at least two independent reasons.
Never tag based on placing alone.
Compare full-data and tag-excluded trends.
Review tags after seven days.
Keep uncertain results with notes.
The rule protects your trend without making the record too neat.
In N+One terms: prefer a noisy point over a false tag that hides real change.
Keep reading
- N+One Power Curve vs Strava Power Curve: Why the Numbers Differ — Why N+One and Strava power curves can differ, what processing choices change the numbers, and a 7-day protocol to compare matched ride files.
- Understanding a workout or session — How to read a session page in N+One: header metrics, intervals, streams, maps, and calendar context so you can interpret every ride.
- The N+One Race Plan Module: How the App Builds a Pacing Strategy for Your A‑Race — See how the N+One Race Plan Module turns training history, race details, and recent form into one pacing strategy for your A-race.
A tag is useful only when it changes the next decision. After a probable outlier, use the cleaner trend for the next block, but keep the raw result in view.
If the excluded trend is stable, do not chase the race result. Hold the plan and let the next few sessions give better signal.
If the same pattern repeats, treat it as information rather than noise. Then review load, recovery notes, and form metrics without the jargon.
Breakthrough days need the same guardrails as poor days. Before you raise targets, check whether N+One sees a repeatable shift in new peak-effort detection.
Your next move should be clear: adopt the tag-excluded trend for planning, then review the tag after seven days.
Use the tag-excluded trend for short-term planning.
Do not chase one odd race.
Review the next few sessions for repeat patterns.
Keep breakthrough tags under the same rules.
Reassess the tag after seven days.
Step 1 — Tag within 48 hours: annotate the race file with outlier=probable if two context flags are present. Keep the raw data and add one reason code.
Step 2 — Hold for 7 days: keep intensity, cut volume by 20% for seven days, then let recovery and context metrics settle before changing the block.
Step 3 — Reassess after 7 days: re-check heart rate, HRV, sleep, and any repeated poor performances. If the wider signal supports the tag, exclude the race from short-term form calculations. If not, remove the tag.
Start with rules, not eyeballing. Tag race outliers rather than deleting them, run the trend with and without the tag, and use the tag-excluded line for the next training decision while the full record stays intact.
No. Keep the raw file, tag it, and add the reason code. Deleting the file weakens your record and makes later review harder.
Use the same rule. A great result can also bend the trend, so tag it only when it meets the criteria and review whether the gain repeats.
Keep the event visible in the log, but use the tag-excluded fit for short-term planning when the result meets your outlier rule.
Review the tag after seven days, then keep a count over time. Frequent tags may mean the rule, model, or race context needs a cleaner setup.