Ride data is only useful when you know what to do with it. A file full of watts, beats per minute, and speed numbers is not analysis — it is just noise. This guide walks through the core metrics, how they relate to each other, and how to build a repeatable habit of post-ride review that actually improves your training.
Start With the Purpose of the Ride
Before opening any file, recall what the session was supposed to achieve. A recovery spin, a threshold interval block, and a long endurance ride each produce data that looks completely different — and should. Comparing them without that context leads to bad conclusions.
Ask two questions before you look at a single graph:
- What was the intent of the session?
- Did execution match intent?
If the intent was a two-hour aerobic ride and your heart rate averaged well above your aerobic threshold, something went wrong regardless of how the power trace looks. Intent is the lens through which all other data gets interpreted.
Power: The Foundation Metric
For riders with a power meter, power is the starting point for almost every analysis. Unlike speed, power is not affected by wind, gradient, or drafting. It tells you exactly how much work your muscles produced.
Average vs. Normalised Power
Average power is the mean of every recorded second. Normalised power (NP) accounts for the physiological cost of variable effort — hard surges cost more than steady riding at the same average. On a flat time trial, average and NP will be close. On a punchy criterium or a ride with repeated climbs, NP will be meaningfully higher than average.
When NP is significantly higher than average power, the ride was metabolically more taxing than the average suggests. This matters when estimating recovery needs.
See the glossary entry for Normalised Power for the full calculation.
Intensity Factor and Training Stress Score
Intensity Factor (IF) is normalised power divided by your FTP. An IF of 1.0 means you averaged your threshold for the entire ride. Most endurance rides sit below 0.75. A hard race might reach 0.90 or above.
Training Stress Score (TSS) combines IF and duration into a single load number. The formula penalises sustained high intensity more than duration alone. A three-hour moderate ride and a ninety-minute threshold effort can produce similar TSS values for very different reasons.
Use the FTP estimator if your FTP has not been tested recently — an accurate FTP is essential for IF and TSS to mean anything.
Power Distribution and Time-in-Zone
The power histogram shows how many seconds you spent at each watt level. For a long endurance ride, the distribution should be a clear hump sitting in your aerobic zones. For an interval session, expect a bimodal shape — time at low power during recoveries and a cluster near or above threshold during efforts.
If an endurance ride shows a wide, flat distribution with lots of time in zones three and four, the ride was likely too uncontrolled. If an interval session shows most time in zone two with only brief spikes at the target zone, the intervals were too short or not hard enough.
Heart Rate: Context and Lag
Heart rate reflects cardiovascular demand, not mechanical output. It responds to heat, fatigue, caffeine, sleep quality, and emotional stress in ways that power does not. That makes it a poor real-time pacing tool but a valuable post-ride signal.
Cardiac Drift
On long aerobic rides, heart rate tends to rise over time even when power stays constant. This is cardiac drift — largely driven by dehydration and rising core temperature. A drift of a few beats per hour during a long ride is normal. Significant drift early in a session can indicate poor recovery, illness onset, or inadequate fuelling.
Plot heart rate and power on the same timeline after a long steady ride. If power holds flat and heart rate climbs steeply, flag it. If both hold steady, the aerobic system was working efficiently.
Efficiency Factor
Efficiency Factor (EF) is normalised power divided by average heart rate. It is a rough proxy for aerobic fitness: a fitter rider produces more watts per beat. Track EF over several months on comparable sessions — same route, similar conditions, similar duration. A rising EF trend is a reliable sign that aerobic fitness is improving.
Day-to-day EF variation is normal and expected. Trend over weeks matters; single-ride readings do not.
Speed, Cadence, and What They Tell You
Speed is the least useful metric for training analysis because it depends on too many external variables. Use it for route comparison over identical segments, not for effort assessment.
Cadence is more interesting. Most road riders settle into a self-selected cadence that reflects both neuromuscular preference and fatigue state. A rider who normally pedals at ninety rpm dropping to seventy-five on the back half of a long ride may be accumulating muscular fatigue faster than cardiovascular fatigue — a sign of insufficient endurance base or inadequate fuelling.
For climbing, compare cadence at a given power on repeated ascents of the same grade. A consistent cadence across similar efforts suggests stable pacing. Erratic cadence may indicate poor pacing judgment or fatigue.
Elevation and Gradient Analysis
The elevation profile contextualises everything else. Spikes in power that look alarming on a flat section are expected on a steep ramp. Before judging any power or heart rate anomaly, check whether it coincides with a gradient change.
For structured climbing work, use per-climb breakdowns. Compare average power, average gradient, duration, and VAM (vertical ascent rate) across repeated efforts on the same climb. If VAM drops across three repeats on Alpe d'Huez while power holds constant, muscular fatigue is accumulating faster than aerobic recovery allows.
Building a Post-Ride Review Habit
Consistency in how you review data matters more than depth on any single ride. A five-minute structured review after every session is more valuable than an occasional deep dive.
The Five-Minute Review Checklist
- Check intent vs. execution. Did power and heart rate reflect the goal zones?
- Scan the timeline for anomalies. Unexpected drops, spikes, or gaps often have explanations worth noting.
- Record subjective feel. Rate effort and how legs felt on a simple one-to-ten scale in your training log. Data without perceived exertion context is incomplete.
- Note one takeaway. One actionable observation per ride. Not ten. One.
- Flag recovery needs. High TSS, large cardiac drift, or a hard effort two days before a key session all warrant adjusting the next day's plan.
Weekly Load Review
Once per week, zoom out. Look at:
- Cumulative TSS for the week vs. your planned load
- Acute vs. chronic load ratio (often called Form or TSB in training platforms) — this signals whether you are building toward fatigue or recovering toward freshness
- Consistency of zone distribution across the week — are you actually spending most time in endurance zones, or has creep pushed everything harder than planned?
If you are using a structured training plan, weekly load review is where you catch whether the plan is working or whether life has pushed you off track.
Common Interpretation Mistakes
Chasing a single big number. A peak five-second power or a maximum heart rate is rarely useful in isolation. Context — where in the ride it occurred, what came before, what came after — is everything.
Comparing rides without controlling variables. A faster time on a segment after a week of rest versus after a hard training block tells you nothing about fitness change. Control for fatigue state before drawing conclusions.
Trusting TSS as the whole story. TSS captures load but not specificity. A hundred TSS from a flat endurance ride and a hundred TSS from climbing repeats have different training effects and different recovery demands.
Ignoring subjective data. If the numbers look fine but you felt terrible, the subjective signal matters. Data files do not capture every variable acting on your body.
Over-indexing on single sessions. One anomalous ride — high heart rate, low power, poor efficiency — is a data point, not a trend. Wait for a pattern before adjusting training.
Connecting Data to Decisions
Data analysis earns its value only when it changes behaviour. The chain is: data → interpretation → decision → outcome.
If reviewing your rides is not regularly leading to adjusted training decisions, the process is incomplete. That does not mean every review triggers a change. Sometimes the decision is to stay the course because the data confirms the plan is working. But there should always be a decision, even a passive one.
Common decisions that data should inform:
- Whether to execute the planned session tomorrow or substitute recovery
- Whether the current training block is building load appropriately or stagnating
- Whether pacing strategy on a target climb or segment needs adjustment
- Whether an equipment change (different gearing, bike fit tweak) has produced a measurable efficiency gain
Tools and Data Sources
Most training platforms ingest data from cycling computers automatically. What varies is how well they surface the metrics that matter for your specific goals. Look for platforms that give you:
- Clean interval detection and per-interval breakdowns
- Long-term EF and TSS trend views
- Customisable zone definitions tied to your current FTP
- Easy side-by-side comparison of similar sessions over time
Regardless of platform, keep your FTP updated. An outdated FTP corrupts IF, TSS, and zone calculations across every ride. Retest every six to eight weeks during active training, or after any significant fitness event — illness, a training camp, a peak race.
Ride data rewards patience and consistency. The riders who improve most from data-driven training are not those who analyse the most deeply on any given day — they are the ones who review with discipline, connect numbers to decisions, and adjust without ego.