Strava Fitness going down while you're still training hard: the 42-day decay math

How Strava Fitness decays: the 42-day EWMA math

Fitness is a 42-day exponentially-weighted moving average of daily training load — CTL in Coggan and Allen's terms [Allen et al. 2019]. Each day it blends in today's load and lets the rest decay by a fixed fraction. The headline number that explains a falling line: with zero training, you lose about 15% of your Fitness in a week and roughly half in 29 days.

The mechanism is one line of arithmetic. A 42-day EWMA carries forward about 41/42 of yesterday's value and adds 1/42 of today's load [Allen et al. 2019]. The window length traces to Eric Banister's 1970s fitness-fatigue model, whose decay constants were curve-fit against athlete performance and have held up across decades of validation [Hellard et al. 2007]. Because the average is exponential, recent rides count more than old ones, and the line is always drifting toward whatever your last few weeks of load actually were.

Run the decay forward and the paradox dissolves. Ride nothing and Fitness multiplies by (41/42) each day: about 0.85 after a week, so roughly 15% gone in seven days, and about 0.36 after 42 days — you keep barely a third. The half-life is near 29 days. The point isn't that you should panic; it's that the line falls fast relative to how it climbed. CTL rises slowly because you can only add a little load per day, but it sheds quickly the moment your input drops below the running average.

That asymmetry is the whole story. To hold a Fitness number steady, today's load has to roughly match your 42-day average load — not your peak week, the average. A rider who built CTL on 10 hours a week and drops to 6 will watch Fitness fall even if every one of those 6 hours is harder than before, because total daily load fell below the baseline the average is still anchored to. The chart is reporting load, not effort, and the two are not the same.

The real reasons your Fitness is dropping

Four causes account for almost every falling Fitness line in a rider who feels like they're working: volume dropped even as intensity rose, a few rides went missing, a sensor or input gap under-scored real work, or you're tapering on purpose. The first three are fixable misreads; the fourth is the chart working correctly.

Cause one, and the most common: volume fell while intensity climbed. Training load is roughly intensity multiplied by duration, so a 45-minute threshold session can score less total load than a 2-hour endurance ride even though it hurt far more [Allen et al. 2019]. Trade three long base rides for three short hard ones and your weekly load drops, dragging the 42-day average down with it. The legs feel trashed; the Fitness line feels insulting. Both are telling the truth about different things.

Cause two: missed or shortened rides you've half-forgotten. A skipped weekend long ride is the single biggest one-day hit to CTL most riders have, because it was the largest load input in the week. Two or three of those across a month — travel, weather, a sick kid — quietly reset the baseline the average chases. Cause three is a data problem, not a fitness one: if you train indoors with power and outdoors on heart rate only, or your max HR is mis-set, the load Strava records can badly under-count what you actually did. Our companion piece on mixing power and heart rate in the Strava chart walks through how that sensor switch puts a discontinuity in the curve that looks like lost fitness but isn't.

Cause four is deliberate: a taper. Cut volume before a goal event and Fitness will fall by design while freshness rises — that's the trade you want. Bosquet and colleagues' meta-analysis found that reducing training volume by roughly 41 to 60% over about two weeks maximizes performance gains [Bosquet et al. 2007], and a drop that size visibly lowers the Fitness line. Friel's framework treats the same dip as the cost of arriving fresh [Friel 2018]. If your Fitness is falling in the last two weeks before an event, that is the plan working, not failing.

When a falling Fitness line is exactly right

A dropping Fitness line is correct, not alarming, in three situations: a planned taper into a goal event, a recovery week after a hard block, and the first days of any genuine rest you actually needed. In all three, the freshness you gain is the reason the fitness number gives a little back.

Form is fitness minus freshness, and you cannot raise one without spending some of the other. During a taper, cutting volume drops Fitness by a few points but lets fatigue clear so Form climbs into the positive range research and Friel associate with peak performance [Friel 2018]. Losing 3 to 5 CTL points over a two-week taper while Form swings from negative to +10 or +20 is not detraining — it's the mechanism by which you show up rested. Reading the Fitness drop as a problem is how riders sabotage their own peak.

A recovery week works the same way at smaller scale. Pull volume back for seven days after a long build and Fitness slips a few points; the adaptation you banked in the preceding weeks is still there. Short-term reductions in training don't erase fitness so much as pause its accumulation — Mujika and Padilla's detraining review defines short-term as under four weeks and notes the early losses are dominated by blood-volume and cardiovascular factors that rebound quickly once you train again [Mujika & Padilla 2000]. A one-week dip in the line is noise against that backdrop.

The trap is conflating a correct dip with a failure and adding load to fight it. If the chart falls during a planned recovery week and you respond by riding hard to push it back up, you've converted a recovery week into a monotony problem and undone its purpose. The Fitness line going down is not a command to train more. In a taper or a recovery week, it's confirmation the plan is doing what it's supposed to.

Trend versus noise: how long a drop has to last to mean something

A single ride moves Fitness by 1 to 3 points; daily wiggles are noise. A genuine downward trend needs roughly two to four weeks of load consistently below your 42-day baseline before it means your fitness is actually receding rather than just settling. Read the slope over weeks, never the daily number.

Because Fitness is a 42-day average, one big ride barely moves it and one missed ride barely dents it — typically 1 to 3 points either way [Allen et al. 2019]. That smoothing is the feature: it filters the day-to-day chaos of real life so you can see the multi-week signal. The cost is lag. The line keeps falling for days after you've resumed normal training, because the average is still digesting the low-load stretch behind it. Judging the chart day by day is reading the lag as if it were news.

The honest threshold is two to four weeks. A drop that persists that long, with your weekly load genuinely below the baseline the whole time, is a real downtrend worth acting on. A drop that reverses inside a week or two was a blip — a missed weekend, a sensor gap, a deliberate easy week. This sits inside the broader discipline of using Strava as a training tool rather than a journal: the data layer every cyclist already has only helps if you read its trend, not its daily mood. Strava syncs the number every few minutes; whether the trend means anything is a judgment the chart can't make for you.

Quick answers

Sources cited in this guide

1. 01
   Allen et al. 2019. Training and Racing with a Power Meter (3rd ed.). VeloPress.
2. 02
   Hellard et al. 2007. Assessing the limitations of the Banister model in monitoring training. Journal of Sports Sciences.
3. 03
   Bosquet et al. 2007. Effects of tapering on performance: a meta-analysis. Medicine & Science in Sports & Exercise.
4. 04
   Mujika & Padilla 2000. Detraining: loss of training-induced physiological and performance adaptations. Part I: short term insufficient training stimulus. Sports Medicine.
5. 05
   Friel 2018. The Cyclist's Training Bible (5th ed.). VeloPress.

More inside Training with Strava

Other articles in this series

1. 01

   What your Strava Fitness number means (and if yours is good)

   Strava Fitness is CTL — a 42-day weighted load average. What the number means, why it is personal, and the decisions to make from it.

   →
2. 02

   Apps that connect to Strava: read vs display

   How to tell which training apps actually read your Strava data and adapt versus the ones that only display your rides.

   →
3. 03

   Is Strava Premium worth it for a self-coached cyclist?

   What Strava Premium gives a training-focused rider, what it doesn't (no coaching), and when the free Intervals.icu chart beats paying.

   →
4. 04

   Relative Effort vs TSS: which to trust by workout

   A per-workout-type rule for when to trust Strava Relative Effort vs power-based TSS, and why they are not the same units.

   →
5. 05

   How to set up Strava for training: one-time configuration

   Configure Strava once for training: real FTP and max HR, honest zones, one sensor stack, and feed privacy that protects your plan.

   →
6. 06

   Strava segments as fitness tests: map efforts to tests

   Use Strava segments as scheduled benchmark tests: which profiles map to which test, how to schedule every 4-6 weeks, and controlling variables.

   →
7. 07

   What to look at on Strava after a ride: 4 metrics

   A 30-second post-ride routine: the four Strava metrics that matter after every ride, the ones to ignore, and why.

   →
8. 08

   Strava indoor power vs outdoor HR: Fitness chart jumps

   Mixing power-based TSS and HR-based Relative Effort splices incompatible units into your Strava Fitness chart. Why it jumps and how to fix it.

   →
9. 09

   Heart rate drift on long rides at same power: what it means

   Why heart rate climbs at flat power on long rides — cardiovascular drift, aerobic decoupling (Pw:HR) as a durability signal, and what to do.

   →
10. 10

    Strava heart rate zones wrong: the whole-dashboard cascade

    How a wrong max HR in Strava cascades into bad zones, Relative Effort, and Fitness — and how to set zones from real data.

    →
11. 11

    Why your Strava Relative Effort is high on easy rides

    Relative Effort can spike on a genuinely easy ride — usually a mis-set max HR, not lost fitness. What inflates it on Strava and what to do.

    →