Thames Valley Marathon
Targeting 3:21:00 on 14 December 2026 · 18% probability · 75 sessions analysed
Iteros is writing your personalised sections. The full analysis is already below — the written commentary will appear here in a few seconds.
No finding is read from an archive yet, so nothing leads yet.
This report was made before the radar existed, so it has no shape to draw.
Each best against what today’s fitness says. The softer it is, the more is there to take.
Found in your archive: the runs your files mark as races, and any race result you entered.
Set a race and the next block is built around this.
Thames Valley Marathon, Monday 14 December, is already set.Fitness snapshot
Where you are now against what a 3:21:00 Marathon requires. What the goal demands comes from Daniels' equivalent-performance tables; your current fitness is measured from your own data.
Your threshold pace is 4:16/km. A 3:21:00 Marathon requires sustaining approximately 4:46/km, implying a threshold of about 4:27/km. You have a 10 second buffer — comfortable but not excessive.
How it was worked out: Measured from your fastest sustained 21-minute effort — 4:16/km at 185bpm.
A weekly 20-30 minute continuous effort at threshold pace moves this faster than any other single session.
How this has moved
The fastest sustained tempo or threshold block in each session — a direct measurement of what you can hold, rather than a value inverted from VO2max.
Only 2 sessions carry this metric, which is enough to report where you are but not enough to say how it has changed.
Against the population
77th percentile
Your measured threshold pace against the threshold pace implied by the race times of 500 men aged 35 - 54.
Economy scores 62/100 overall, 72/100 at easy pace (6:08/km). Each pace band is scored against your own range at that band, not against an absolute standard — so this tracks whether you're getting more efficient, rather than ranking you against other runners.
How it was worked out: Scored across 3 pace bands from heart-rate cost at pace, cadence, ground contact time, vertical oscillation.
Economy responds to cadence work, hill sprints and strength rather than to more aerobic volume.
Lactate thresholds
Two thresholds, not one. LT1 is where easy stops being easy; LT2 is the pace you could hold for about an hour. Almost every training decision is really a decision about which side of these two lines to run on.
LT1 · Aerobic threshold
low confidence159bpmat5:37/km
83% of max HR · Cardiac drift onset
The ceiling of true Zone 2. Below this, a steady pace costs a steady heart rate and you can run for hours; above it, heart rate climbs through the run even when pace doesn't.
Across 32 runs over an hour, heart-rate drift crossed 5% at 159bpm. Every one of those runs stayed on the same side of 5%, so the crossing is projected beyond your data — treat it as a direction, not a number.
LT2 · Anaerobic threshold
low confidence171bpmat4:16/km
89% of max HR · Threshold pace, modelled HR
The hardest effort you can hold for 40-60 minutes. Above it, lactate accumulates faster than you clear it and the clock starts.
Your sustained threshold blocks averaged 185bpm — 96% of max, which is race effort rather than threshold, so those blocks were harder sessions than their pace suggested. 171bpm is used instead, placed at 85% of your heart-rate reserve. Measured from your fastest sustained 21-minute effort — 4:16/km at 185bpm.
Your productive training zone: 159-171bpm (5:37-4:16/km)
Sessions in this zone drive the most adaptation. Below LT1 you are building base without much stimulus; above LT2 you are accumulating fatigue faster than fitness.
How LT1 was located
Each point is one run over an hour: the heart rate it was held at, against how much your efficiency fell between its first and second halves. Runs below the 5% line held together; runs above it drifted. LT1 is where the two behaviours meet.
Running economy by pace band
| Band | Score | Pace | HR cost | GCT | Cadence |
|---|---|---|---|---|---|
| Easy (>5:00/km) | 72 | 6:08/km | 146bpm | 236ms | 178spm |
| Moderate (4:30-5:00/km) | 64 | 4:38/km | 180bpm | 215ms | 183spm |
| Marathon pace (4:10-4:30/km) | 50 | 4:16/km | 192bpm | 200ms | 191spm |
HR cost is the heart rate this band costs after correcting for heat and climbing. Each band is scored against your own range within it, not against other runners — a hard session naturally costs more heartbeats than an easy one, so one shared scale would say nothing. The overall economy figure above is the average across the bands with data.
Badges earned from this data
3 badges from 75 sessions. Every one is checked against metrics already calculated above — tap any badge to see exactly what earned it.
Closest to earning
Thousand Club
153km to go.
Century
25 more sessions.
Elastic
Needs 3+ sessions at marathon pace with ground contact recorded; you have 2.
No GOAT tier reached yet.
All six axes must reach 70, and Terrain could not be scored from this upload.
GOAT tiers require every one of the six profile axes to clear the threshold at the same time, which is why they are rare — a single weak axis holds back the whole tier.
Full collection
3 of 25 earned
Locked badges show what they need. Rarity figures are current estimates and will be measured from real data as more runners generate reports.
Earned
Your form holds up when it matters most. Past 25km, your mechanics barely change.
- Durability score
- 95
- Session date
- 2026-08-06
- Distance km
- 28.0
You finish faster than you start. Smart pacing and the fitness to back it up.
- Negative split rate
- 0.9
- Qualifying sessions
- 7
Every runner starts somewhere. This is where your data begins.
- Sessions
- 75
Closest 3
A thousand kilometres on the clock.
153km to go.
85% of the way there
One hundred sessions analysed. The picture of you as a runner is properly detailed now.
25 more sessions.
75% of the way there
Springs for tendons. Your ground contact time at race pace is what elite coaches dream of.
Needs 3+ sessions at marathon pace with ground contact recorded; you have 2.
67% of the way there
Training analysis
75 sessions across 55 weeks, 2 October 2025 to 19 October 2026.
Total distance
847km
75 sessions
Recent volume
31km
per week, last 8 weeks
Sessions per week
2.1
1.4 across the full period
Longest run
42.2km
single session
- You averaged 31km/week across 2.1 sessions over the last 8 weeks. Your longest run was 42.2km.
- 75 sessions and 847km in total, from 11 September 2025 to 28 September 2026.
- Peak week: 54km beginning 31 August.
- 6 of 75 sessions (8%) were tempo or interval work. That's low — most successful builds sit around 15-20% quality.
Weekly volume
Session mix by distance
| Type | Sessions | Distance | Share |
|---|---|---|---|
| Easy | 49 | 438km | 52% |
| Tempo | 2 | 24km | 3% |
| Interval | 4 | 44km | 5% |
| Long run | 17 | 289km | 34% |
| Race | 3 | 52km | 6% |
Long run progression
Kilometres by heart-rate zone
Measured per second of running, not per session — so a session that alternates hard reps with easy recovery contributes to both, rather than averaging into the middle. Zones are shares of heart-rate reserve. Session labels are a judgement; kilometres are not.
| Zone | Distance | Time | Share |
|---|---|---|---|
| Z1 Recovery | 0.0km | 0:00 | 0% |
| Z2 Base | 717.2km | 74:07:39 | 85% |
| Z3 Aerobic | 7.6km | 48:12 | 1% |
| Z4 Threshold | 42.2km | 3:36:18 | 5% |
| Z5 Above | 79.9km | 6:04:14 | 9% |
Training load — fitness, fatigue and form
Fitness (CTL) is a 42-day weighted average of training load; fatigue (ATL) a 7-day one. Form (TSB) is the difference — positive means fresher than your recent training, which is what a taper is for.
Your progress
What actually changed across 13 months of training. Starting and current values are averages of the first and last quarter of the block, not single sessions — one Tuesday is not a baseline.
What actually worked
For each metric that moved, the training in the fortnight before its biggest gains, measured against your own typical fortnight. This is the part of the report that is about you specifically rather than about running.
Easy-pace HR improved most after fortnights with 2.0 long runs against your usual 0.9.
| Gain seen | Runs | Distance | Quality | Long runs | Climb |
|---|---|---|---|---|---|
| 6 Aug | 8 | 110km | 2 | 2 | 187m |
| 20 Aug | 8 | 111km | 1 | 2 | 191m |
| your usual | 6.3 | 74km | 0.7 | 0.9 | 126m |
Based on the 2 largest improvements in this metric — an association across a single training block, not a controlled comparison. Fitness rises through a block whatever the runner does, so anything that trended upward alongside it will appear here.
Late-run contact-time drift improved most after fortnights with 1.0 long runs against your usual 0.7.
| Gain seen | Runs | Distance | Quality | Long runs | Climb |
|---|---|---|---|---|---|
| 6 Aug | 7 | 82km | 2 | 1 | 138m |
| 3 Oct | 7 | 82km | 2 | 1 | 138m |
| 29 Aug | 4 | 58km | 1 | 1 | 99m |
| your usual | 5.4 | 63km | 1.3 | 0.7 | 108m |
Based on the 3 largest improvements in this metric — an association across a single training block, not a controlled comparison. Fitness rises through a block whatever the runner does, so anything that trended upward alongside it will appear here.
Ground contact time improved most after fortnights containing 1.3 quality sessions (interval, tempo, threshold or race) against your usual 0.9.
| Gain seen | Runs | Distance | Quality | Long runs | Climb |
|---|---|---|---|---|---|
| 4 Aug | 7 | 82km | 2 | 1 | 138m |
| 22 Aug | 8 | 94km | 2 | 1 | 161m |
| 21 Jan | 5 | 47km | 0 | 0 | 79m |
| your usual | 6.1 | 68km | 0.9 | 0.6 | 115m |
Based on the 3 largest improvements in this metric — an association across a single training block, not a controlled comparison. Fitness rises through a block whatever the runner does, so anything that trended upward alongside it will appear here.
Cadence improved most after fortnights containing 2.0 quality sessions (interval, tempo, threshold or race) against your usual 0.9.
| Gain seen | Runs | Distance | Quality | Long runs | Climb |
|---|---|---|---|---|---|
| 4 Aug | 7 | 82km | 2 | 1 | 138m |
| 22 Aug | 8 | 94km | 2 | 1 | 161m |
| your usual | 6.1 | 68km | 0.9 | 0.6 | 115m |
Based on the 2 largest improvements in this metric — an association across a single training block, not a controlled comparison. Fitness rises through a block whatever the runner does, so anything that trended upward alongside it will appear here.
Weekly volume improved most after fortnights containing 1.0 quality sessions (interval, tempo, threshold or race) against your usual 0.3.
| Gain seen | Runs | Distance | Quality | Long runs | Climb |
|---|---|---|---|---|---|
| 10 Aug | 7 | 98km | 1 | 2 | 167m |
| 12 Oct | 7 | 84km | 2 | 1 | 144m |
| 6 Jul | 0 | 0km | 0 | 0 | 0m |
| your usual | 2.7 | 30km | 0.3 | 0.3 | 50m |
Based on the 3 largest improvements in this metric — an association across a single training block, not a controlled comparison. Fitness rises through a block whatever the runner does, so anything that trended upward alongside it will appear here.
Every metric, start to now
The fastest sustained tempo or threshold block in each session — a direct measurement of what you can hold, rather than a value inverted from VO2max.
Only 2 sessions carry this metric, which is enough to report where you are but not enough to say how it has changed.
Against the population
77th percentile
Your measured threshold pace against the threshold pace implied by the race times of 500 men aged 35 - 54.
Heart rate on easy and long runs, corrected for heat and climbing. Falling at the same pace is the clearest single sign of an improving aerobic base.
Starting
143bpm
Current
142bpm
Change
no material change
Sessions that drove this
- 23 Jul 2026 · 8.0km easy — 6:10/km at 140bpm−14bpm
- 27 Jul 2026 · 10.0km easy — 6:10/km at 140bpm−12bpm
- 20 Aug 2026 · 9.0km easy — 6:10/km at 140bpm−9bpm
Measured against the four sessions before each one, so a strong session after a poor one isn't credited with the poor one's recovery.
Against the population
The population dataset records training volume, pace and consistency — no heart rate, no running dynamics and no elevation — so there is no like-for-like comparison for this metric.
How far your ground contact time rises above its fresh cost once form starts giving way, per long run. This is the measurement the Durability axis scores — the axis turns it into a 0-100 position, and this is the raw quantity moving underneath it.
Starting
0.7%
Current
0.7%
Change
no material change
Sessions that drove this
- 3 Oct 2026 · 28.0km — +0.6% contact-time drift after the trajectory turned−0.1%
Measured against the four sessions before each one, so a strong session after a poor one isn't credited with the poor one's recovery.
Against the population
The population dataset records training volume, pace and consistency — no heart rate, no running dynamics and no elevation — so there is no like-for-like comparison for this metric.
The heart-rate cost of your easiest sustained effort in each session, corrected for heat and climbing — how much of your range easy running takes. This is a different measurement from the economy figure in the fitness snapshot, which is how much of your aerobic ceiling converts into race performance and is computed once from a race result.
Starting
54/100
Current
38/100
Change
−15/100
Sessions that drove this
- 5 Oct 2026 · 8.0km easy — 6:10/km+14/100
- 19 Sept 2026 · 8.0km easy — 6:10/km+12/100
- 8 Aug 2026 · 8.0km easy — 6:10/km+12/100
Measured against the four sessions before each one, so a strong session after a poor one isn't credited with the poor one's recovery.
Against the population
The population dataset records training volume, pace and consistency — no heart rate, no running dynamics and no elevation — so there is no like-for-like comparison for this metric.
Milliseconds each foot spends on the ground. Shorter usually means a more elastic, less braking stride.
Starting
236ms
Current
227ms
Change
−9ms (−4.0%)
Sessions that drove this
- 14 Aug 2026 · 5.0km race — 4:16/km−33ms
- 11 Oct 2026 · 5.0km race — 4:16/km−33ms
- 2 Aug 2026 · 11.0km interval — 5:42/km−30ms
Measured against the four sessions before each one, so a strong session after a poor one isn't credited with the poor one's recovery.
Against the population
The population dataset records training volume, pace and consistency — no heart rate, no running dynamics and no elevation — so there is no like-for-like comparison for this metric.
Steps per minute, excluding walk-pace sections. Higher cadence at the same pace usually means less overstriding.
Starting
178spm
Current
181spm
Change
+3spm (+1.8%)
Sessions that drove this
- 14 Aug 2026 · 5.0km race — 4:16/km+12spm
- 11 Oct 2026 · 5.0km race — 4:16/km+12spm
- 2 Aug 2026 · 11.0km interval — 5:42/km+10spm
Measured against the four sessions before each one, so a strong session after a poor one isn't credited with the poor one's recovery.
Against the population
The population dataset records training volume, pace and consistency — no heart rate, no running dynamics and no elevation — so there is no like-for-like comparison for this metric.
Kilometres per week. More volume is the single strongest predictor of endurance performance, up to what you can absorb.
Starting
21.7km
Current
32.2km
Change
+10.4km (+48.1%)
Sessions that drove this
- 27 Jul 2026 · Week of 6 Jul — 4 sessions, longest 24.0km+48.0km
- 21 Sept 2026 · Week of 31 Aug — 4 sessions, longest 24.0km+41.4km
- 28 Sept 2026 · Week of 7 Sept — 3 sessions, longest 28.0km+32.5km
Measured against the four sessions before each one, so a strong session after a poor one isn't credited with the poor one's recovery.
Against the population
19th percentile
Your current weekly volume against the average weekly volume of 500 men aged 35 - 54.
What you can run
Your best efforts at every distance, what each level of effort buys you right now, and whether those bests still describe you.
Personal records
| Distance | Time | Pace | Date | Where it came from |
|---|---|---|---|---|
| 1km | 4:15 | 4:15/km | 14 Aug 26 | segment inside a longer run (from km 1.0) |
| 5km | 22:21 | 4:28/km | 29 Aug 26 | segment inside a longer run (from km 0.1) |
| 10km | 44:43 | 4:28/km | 29 Aug 26 | segment inside a longer run (from km 9.3) |
| Half marathon | 1:34:20 | 4:28/km | 29 Aug 26 | segment inside a longer run (from km 0.0) |
| Marathon | — | — | — | never covered this distance |
Best continuous effort at each distance, found anywhere inside a session — a hard 5km buried in a long run counts, which is why some of these are segments rather than races. Distances you have never covered are left blank rather than estimated. The population line is the median for your cohort, converted between distances through measured race results rather than a formula.
Pace at every effort
Measured on the first ten minutes of every session, so this is what each effort buys you FRESH. The shaded band is the productive zone between your two thresholds. The pale markers are the intensities each race distance is typically run at — where they cross your curve is the pace that distance should cost you.
How it all fits together
Heart rate
Ground contact
At 6:00/km you run 138bpm (72% of max) with 236ms of ground contact.
Hover any chart to move the marker on all of them. Both curves are measured on fresh running, so this is what each pace costs you before fatigue, not averaged across it. No gradient band cleared its sample minimum, so there is no climb figure here.
Were these well executed?
not assessable
Set when your VO2max was 44.1; it is now 48.4, so this time is probably 23s soft.
under-executed
Held at 90% of max heart rate, your fresh curve says that effort buys 8s/km faster than you ran. That gap is pacing, conditions or fatigue on the day rather than a fitness ceiling.
under-executed
Held at 93% of max heart rate, your fresh curve says that effort buys 14s/km faster than you ran. That gap is pacing, conditions or fatigue on the day rather than a fitness ceiling.
under-executed
Held at 92% of max heart rate, your fresh curve says that effort buys 12s/km faster than you ran. That gap is pacing, conditions or fatigue on the day rather than a fitness ceiling.
Which way you are heading
No fitness trend available, so the prediction range is left symmetric.
Your pace decays 1.0% per doubling of distance, against 2.3% for the population over the same range.
47km. 3 runs.
September.
August: 184 kilometres.
No new bests in September. Your 1K best, 4:15, is about 23 seconds soft against your current fitness.
The highest load of the month.
6 runs, among them 1 interval session, 1 tempo or threshold run and 1 long run, and runs up to 24km.
2025.
Not yet: the type you ended the year as is not drawn: a type is read for today, and the weekly replay keeps each axis's score, not a type.Thames Valley Marathon.
56 days. Everything here points at one number.
Load down 81% from peak. Freshness falling.
Your race is in 56 days.
The exact figure
18 in 100 of your finish range is under 3:21.
Eight in ten finish between 3:16 and 3:52, most likely about 3:34. Failure is pushing on past 11km once your form goes.
This forecast hasn't yet been checked against your own races.
To break 3:21, your first half must be 1:40:30.
Good luck. The race review arrives when your file does.
Strengths
What your data says you're genuinely good at.
Threshold pace comfortably ahead of the goal
Your threshold pace is 4:16/km. The goal implies a threshold of about 4:27/km, so you're carrying a 10 second per kilometre buffer. That margin is what absorbs a warm day, a headwind, or a bad patch mid-race.
Long run built patiently
Your long run grew from 24.0km on 8 July to 42.2km on 8 August across 7 long runs — progressive overload done at a rate that builds tissue rather than breaking it.
Improvement areas
Ranked by how much each one actually costs you against this specific goal, at this specific distance.
Structural load tolerance is still building
Accumulated lifetime load is the slowest-moving input in the model — tendon, bone and connective tissue adapt over years, not training blocks, and this is the one limiter you cannot rush.
The fix
Don't try to fix this before the race. Hold volume steady, keep the easy running genuinely easy at 5:23/km or slower, and treat consistency over the next months as the intervention.
VO2max is short of what the goal needs
The goal requires a VO2max of about 47.0, and your current estimate sits below that. VO2max responds faster than any other input on this list, which is what makes it worth attacking directly.
The fix
You've done 2 interval sessions in the last 6 weeks. Add one a week: 5-6 × 3 minutes at 4:02/km with 2-3 minutes jog recovery. Expect a measurable shift in 4-6 weeks.
Threshold pace is the binding constraint
Threshold work is the highest-leverage input for this distance and it's underweighted in your recent training.
The fix
You've done 1 threshold or tempo session in the last 6 weeks. Add one a week: 20-25 minutes continuous at 4:27/km, or 3 × 8 minutes at that pace with 2 minutes easy between. This is the single change that most directly closes the gap.
Race day
Kilometre by kilometre through Thames Valley Marathon, using your durability profile, cardiac drift, and the course elevation.
Predicted finish
3:33:43
3:15:37 – 3:51:48 range
Confidence
High
anchored to a real result at this distance
Versus your goal
+12:43
outside your target
Cost of the course
+0:01
80m climb, 80m descent
Cost of fatigue
+0:12
fade begins around km 32
Estimated from your races and your running
3:30:06
4:59/km
high confidenceRange 3:15:36 – 3:44:42, from your race and training projections.
What it rests on
Your fastest race and your fastest sustained running, each carried to this distance. A race stands unless your training says faster, and then the race still carries five-eighths of the figure.
| Source | Suggests | Weight | Why it carries that weight |
|---|---|---|---|
| Your training datahigh | 3:44:42 | 30% | The pace you hold at 83% of heart-rate reserve, carried out to race duration. |
| Your race historyhigh | 3:15:36 | 50% | Your Marathon in 3:15:36, converted using ratios measured from population athletes who raced both distances. |
| Economy-adjusted ceilinglow | 3:44:29 | 20% | VO2max 48.4 multiplied by an assumed 85% economy, since yours cannot be measured without a race result, giving an effective 41.1. The conservative bound. |
How it was carried to the distance
| Step | Cost | What it means |
|---|---|---|
| Pace you hold at 83% of heart-rate reserve | — | 5:01/km, measured across 135 minutes of your own sustained efforts either side of this intensity. That is 3:31:21 for 42.20km before anything else. |
| Holding it for race duration | +13:20 | Those efforts lasted about 27 minutes; this race asks for 225. Pace at a fixed heart rate decays about 1.9% per hour for you, from 6.5bpm/hour of measured cardiac drift, adjusted by your durability of 64/100. |
Checked against your race history
Training data predicts 3:44:42. Your Marathon in 3:15:36 suggests 3:15:36 — faster than your current training supports. Racing brings out more than training does, but a gap this way round can also mean the training block has been easier than the one that produced the race.
Your pace at each intensity
Every sustained effort in your training, grouped by how hard it was as a share of your heart-rate reserve. Shown beside the estimate rather than inside it: the estimate rests on what you have run, and this curve on what each heart rate costs you.
Where the hard parts are
No route geometry for this course. Upload the race GPX to see the course on a map — the elevation profile below is modelled from the total climb you entered.
Course elevation
Hovering the map highlights the same point on the charts, and vice versa. Elevation and pace share a kilometre axis rather than being stacked on two y-scales, which would distort both. The pace axis is inverted so a falling line means slowing down.
The five hardest kilometres, and what to do in them
km 35 · 5:16/km (+19s vs your opening km)
accumulated fatigue — ground contact rising to 229ms.
Hold cadence above 165spm through this section. Form goes before fitness does, and cadence is the part you can consciously keep.
km 36 · 5:20/km (+23s vs your opening km)
accumulated fatigue — ground contact rising to 230ms.
Hold cadence above 165spm through this section. Form goes before fitness does, and cadence is the part you can consciously keep.
km 37 · 5:21/km (+23s vs your opening km)
accumulated fatigue — ground contact rising to 230ms.
Hold cadence above 165spm through this section. Form goes before fitness does, and cadence is the part you can consciously keep.
km 38 · 5:18/km (+21s vs your opening km)
accumulated fatigue — ground contact rising to 229ms.
Hold cadence above 165spm through this section. Form goes before fitness does, and cadence is the part you can consciously keep.
km 42 · 5:17/km (+20s vs your opening km)
accumulated fatigue — ground contact rising to 227ms.
Hold cadence above 165spm through this section. Form goes before fitness does, and cadence is the part you can consciously keep.
Predicted splits
Tap any kilometre to see why it's predicted at that pace.
| Km | Pace | HR | GCT | Elevation | Elapsed | Note |
|---|---|---|---|---|---|---|
| 1 | 4:57 | 156 | 232ms | +1m | 4:57 | |
| 2 | 5:00 | 157 | 233ms | +3m | 9:57 | |
| 3 | 5:01 | 158 | 233ms | +4m | 14:58 | |
| 4 | 4:58 | 158 | 231ms | +2m | 19:57 | |
| 5 | 4:54 | 158 | 229ms | −1m | 24:51 | |
| 6 | 4:51 | 159 | 227ms | −3m | 29:41 | |
| 7 | 4:50 | 159 | 226ms | −4m | 34:31 | |
| 8 | 4:52 | 160 | 227ms | −2m | 39:23 | |
| 9 | 4:56 | 161 | 229ms | +0m | 44:20 | |
| 10 | 5:01 | 161 | 230ms | +3m | 49:20 | |
| 11 | 5:03 | 162 | 231ms | +4m | 54:23 | |
| 12 | 5:02 | 163 | 230ms | +3m | 59:25 | |
| 13 | 4:58 | 163 | 228ms | +0m | 1:04:23 | |
| 14 | 4:54 | 163 | 226ms | −2m | 1:09:17 | |
| 15 | 4:52 | 164 | 225ms | −4m | 1:14:09 | |
| 16 | 4:54 | 164 | 225ms | −3m | 1:19:03 | |
| 17 | 4:57 | 165 | 226ms | −1m | 1:24:00 | |
| 18 | 5:02 | 166 | 228ms | +2m | 1:29:03 | |
| 19 | 5:06 | 166 | 230ms | +3m | 1:34:09 | |
| 20 | 5:06 | 167 | 229ms | +3m | 1:39:15 | |
| 21 | 5:04 | 167 | 228ms | +2m | 1:44:19 | |
| 22 | 5:00 | 168 | 226ms | −1m | 1:49:19 | |
| 23 | 4:57 | 168 | 224ms | −3m | 1:54:17 | |
| 24 | 4:57 | 169 | 224ms | −4m | 1:59:14 | |
| 25 | 5:01 | 169 | 225ms | −2m | 2:04:15 | |
| 26 | 5:06 | 170 | 227ms | +0m | 2:09:20 | |
| 27 | 5:10 | 170 | 229ms | +3m | 2:14:31 | |
| 28 | 5:13 | 170 | 229ms | +4m | 2:19:43 | |
| 29 | 5:12 | 170 | 229ms | +3m | 2:24:55 | |
| 30 | 5:08 | 170 | 227ms | +0m | 2:30:03 | |
| 31 | 5:05 | 170 | 225ms | −2m | 2:35:08 | |
| 32 | 5:04 | 170 | 224ms | −4m | 2:40:11 | |
| 33 | 5:06 | 170 | 225ms | −3m | 2:45:17 | |
| 34 | 5:11 | 170 | 227ms | −1m | 2:50:28 | |
| 35 | 5:16 | 170 | 229ms | +2m | 2:55:44 | |
| 36 | 5:20 | 170 | 230ms | +4m | 3:01:04 | |
| 37 | 5:21 | 170 | 230ms | +3m | 3:06:24 | |
| 38 | 5:18 | 170 | 229ms | +1m | 3:11:42 | Fatigue starting to bite — hold form, not pace |
| 39 | 5:15 | 170 | 227ms | −1m | 3:16:57 | Fatigue starting to bite — hold form, not pace |
| 40 | 5:12 | 170 | 225ms | −3m | 3:22:10 | Fatigue starting to bite — hold form, not pace |
| 41 | 5:13 | 170 | 225ms | −4m | 3:27:23 | Fatigue starting to bite — hold form, not pace |
| 42 | 5:17 | 170 | 227ms | −2m | 3:32:40 | Fatigue starting to bite — hold form, not pace |
| final 195m | 5:20 | 170 | 228ms | −0m | 3:33:43 | Fatigue starting to bite — hold form, not pace |
What this model assumes
- · Baseline pace comes from your own pace-to-heart-rate curve — the pace you actually hold at race intensity — rather than from a table. For reference, inverting your VO2max through Daniels' tables would have predicted 3:15:36; the difference between the two is your running economy.
- · Fade modelled from your durability score of 64/100 across 7 long runs.
- · Race-day 18°C against your 13°C training average: about +5bpm at the same effort, costing roughly 1.6% in pace.
- · Heart rate column uses your measured cardiac drift of 6.5bpm per hour at steady effort.
- · Heart rate is held at 170bpm from km 27 — 88% of your 193bpm max, about the highest anyone sustains for this distance. Drift beyond that point shows up as 0s of lost time rather than a higher number on your watch.
- · High confidence — you have raced this distance before, so the prediction is anchored to a real result rather than extrapolated.
- · No GPX supplied, so the 80m of climbing is spread across the course using a typical "rolling" shape. Total elevation is yours; where it falls is modelled.
- · Goal of 12060s sits at the 18% point of this distribution.
Your pacing plan
Even split — targeting 3:21:00.
First half
1:40:30
Second half
1:40:30
Split differential
0:00
even split
Target splits
| Split | Target pace | HR ceiling | Elevation | How it should feel |
|---|---|---|---|---|
| 0-5km | 4:46/km | 156 | +9m | Settle straight onto target pace — no time to bank anything. |
| 5-10km | 4:46/km | 159 | −6m | Settle straight onto target pace — no time to bank anything. |
| 10-15km | 4:46/km | 163 | +1m | Locked on target pace. Check in on breathing rather than the watch. |
| 15-20km | 4:46/km | 166 | +4m | Locked on target pace. Check in on breathing rather than the watch. |
| 20-25km | 4:46/km | 169 | −8m | Locked on target pace. Check in on breathing rather than the watch. |
| 25-30km | 4:46/km | 170 | +10m | Locked on target pace. Check in on breathing rather than the watch. |
| 30-35km | 4:46/km | 170 | −8m | Locked on target pace. Check in on breathing rather than the watch. |
| 35-40km | 4:46/km | 170 | +3m | This is where the race is decided. Hold form and cadence; the pace follows. |
| 40-42.2km | 4:46/km | 170 | −5m | This is where the race is decided. Hold form and cadence; the pace follows. |
Heart rate
- · Stay below 156bpm through the first 15km. Going over this early is the single most common way to lose time later.
- · By the closing kilometres, 170bpm is acceptable — heart rate rises at the same effort as the race goes on, so a rising number is expected, not a warning.
- · 170bpm is the ceiling, and it does not move — that is roughly 88% of your 193bpm max, about the highest anyone holds for this distance. If you are there and still climbing, the answer is to ease the pace, not to accept the heart rate: past this point a higher number buys you nothing and costs you the finish.
- · Your measured cardiac drift is 6.5bpm per hour on long runs. That is what these rising ceilings are built from.
When to push
- · Do not push before km 32. That's where the model expects fatigue to start showing in your pace, and anything spent earlier comes out of that section.
- · Final commitment point: km 38. From there it's 4.2km — close enough that going into the red can't cost you a blow-up.
Fuelling
| When | What |
|---|---|
| 3 days before | Begin carb loading: 8-10g carbohydrate per kg bodyweight daily while training volume drops. |
| 3 hours before | 2-3g carbohydrate per kg bodyweight. Practised breakfast only. |
| 45-60 min before | 3mg/kg caffeine. |
| From km 8, every 25-30 min | Gel with water — roughly 8 across the race, 60-90g carbohydrate per hour. |
| km 30-32 | Take a gel here even if you don't feel like it. This is where under-fuelled marathons end. |
| Throughout | Water to thirst; electrolyte if it's warm or you're a heavy salt sweater. |
The marathon is the only one of these distances where fuelling genuinely decides the outcome. Every gel timing above should be rehearsed in a long run before race day, not tried for the first time.
Comparable runners
200 male runners aged 35 - 54 with training like yours who raced a marathon, matched on weekly volume, longest run, session frequency and average pace.
Their median finish
3:57
Reached your goal
31 of 200
16% of similar runners
Achievers' average
3:07
among those who hit it
What the ones who hit the goal did differently
Runners with your profile who beat this goal trained 5km/week more than you — 41km/week against your 36km/week.
Their longest run averaged 36km against your 42km.
They ran 3.0 times a week against your 2.6 — the gap is frequency more than it is any single hard session.
Runners who hit this goal averaged 0.1 interruptions to their build against 0.2 for those who missed it. Consistency separated them more than volume did.
The closest 8 runners to you
| Marathon time | Weekly km | Longest run | Sessions/wk | Disruptions |
|---|---|---|---|---|
| 3:58 | 36 | 43 | 2.8 | 0 |
| 3:49 | 38 | 42 | 2.8 | 0 |
| 4:59 | 40 | 43 | 2.5 | 0 |
| 3:57 | 39 | 43 | 2.6 | 0 |
| 4:28 | 35 | 43 | 2.2 | 0 |
| 4:21 | 38 | 42 | 3.0 | 0 |
| 4:10 | 35 | 36 | 2.7 | 1 |
| 3:40 | 39 | 42 | 2.8 | 1 |
Green rows reached the equivalent of your goal.
Training load and injury risk
Your acute:chronic workload ratio against the observed disruption rates of 18,000+ marathon builds.
Current ACWR
1.59
1.5-2.0 (danger)
Disruption rate in this zone
32%
of 419 runners
Week-on-week change
-77%
most recent week's volume
Volume bracket risk
8%
at 40-50km/week — your last 4 complete weeks average 46km, which is the acute window this ratio is built on
Your ACWR is 1.59 — the 1.5-2.0 (danger) zone. Across 419 runners in the population dataset, 32% of those training in this zone experienced a disruption to their build.
Reduce volume by about 10% next week. At this ratio the disruption rate is materially above baseline, and a week lost to a niggle costs far more than a week of slightly reduced training.
Acute:chronic workload ratio over time
Green band is the productive range. Amber begins at 1.3, where the observed disruption rate starts to climb; red at 1.5. The first six weeks are omitted — the chronic side of the ratio needs that long before it means anything. The lowest observed disruption rate in the dataset is 12% in the <0.8 (under) zone.
Race readiness timeline
56 days until Thames Valley Marathon on Monday 23 November — that's 8.0 weeks.
This week
Your weakest input for this goal is structural load tolerance (17% of what the goal needs). With 8 weeks to go, hold volume steady rather than increasing it — accumulated load builds over months, and pushing now buys injury risk, not fitness.
Base and build
42-Infinity days out- · Grow weekly volume by no more than 10% a week, with a down week every fourth week.
- · Two quality sessions a week maximum: one threshold, one VO2max or hills.
- · Extend the long run steadily — this is the only window where that's still safe.
- · Specific to you: hold volume steady rather than increasing it — accumulated load builds over months, and pushing now buys injury risk, not fitness.
Race-specific block
14-42 days out- · Shift quality work towards Marathon pace specifically.
- · Practise race-pace running inside long runs, and rehearse race-day fuelling exactly as you'll do it.
- · Peak weekly volume lands in this window, roughly 3-4 weeks before the race.
- · Do a dress rehearsal: race kit, race shoes, race breakfast, race-morning timings.
Taper
7-14 days out- · Cut volume by about 25% this week while keeping intensity — the sessions stay sharp, they just get shorter.
- · Keep session frequency identical. Dropping runs entirely feels restful and leaves you flat.
- · Expect to feel sluggish. That is the taper working, not fitness disappearing.
Final 7 days
3-7 days out- · Volume down to roughly 50% of peak. One short session at race pace (about 15-20 minutes' worth) around 4-5 days out.
- · Last genuinely hard effort: 4 days before. Nothing after that improves fitness.
- · Sleep is the highest-value thing you can do this week — prioritise the nights from 5 days out, not just the last one.
- · Confirm logistics: travel, start time, bag drop, where you'll warm up.
3 days out
1-3 days out- · Start carb loading now: 8-10g carbohydrate per kg bodyweight daily, with training volume near zero. This is worth 2-3 minutes over the marathon distance.
- · Easy 20-30 minutes with a few strides, or complete rest. Both are fine.
- · Hydrate steadily across the day rather than drinking heavily the night before.
- · Lay out everything the night before: kit, number, gels, shoes, charged watch.
Race morning
-1-1 days out- · Wake 3 hours before the gun.
- · Breakfast 2.5-3 hours out: 1.5-2g carbohydrate per kg bodyweight, all of it familiar.
- · Caffeine 45-60 minutes before the start, around 3mg/kg.
- · Short warm-up only: 10 minutes easy and a few strides. The first 3km is the rest of your warm-up.
- · Be at the start pen 20 minutes early — rushing costs more than the warm-up gains.
Goal assessment
3:21:00 for Marathon at Thames Valley Marathon.
Probability of hitting the goal
Uncalibrated: no prediction this app has made has yet been paired with a result, so nothing says how often a figure like this one is right.
For the marathon, structural readiness, durability and long-run coverage together outweigh raw VO2max — past about 30km the limiter is rarely the engine.
Predicted finish distribution
3:15:37
Best case (p10)
3:33:43
Median (p50)
3:51:48
Worst case (p90)
On this course
- Your flat-course estimate
- 3:30:06
- Climbs and descents
- +1s
- Surfaces
- ±0s
- The course's cost
- +1s
- The day and the distance: the late fade and the temperature
- +3:35
- Course-adjusted finish
- 3:33:43
The course-adjusted figure is your flat estimate plus the course's cost, and what the day and the distance add. The course's cost is priced from your own running where it is measured: each climb and descent half from the physics of running on that slope and half from your own training pace on that gradient (physics alone where your training does not measure it), each surface at your own measured cost (the published figure where you have not run on it), and ground steeper than you walk from at your own walking pace.
What's helping and what's hurting
Terrain readiness is unscored. No measured elevation profile for this course. Its weight has been spread across the components that could be measured rather than counted against you.
Structural readiness is the component most likely to decide this race: a goal needs every input that matters for the distance to be sufficient at the same time, not just a good average across them. The percentage above is read off the modelled finish-time distribution — it is the share of simulated races that come in under your target — so this is the input to work on rather than a penalty already applied to that number.
Against the population
- · Your 36km/week over the last 12 weeks puts you in the 27th percentile for training volume among 500 male runners aged 35 - 54.
- · Against 500 male runners aged 35 - 54, this goal would place you in the top 30%.
Recommendations
The specific changes worth making, ranked by impact.
Written recommendations haven't been generated for this report yet.
Structural load tolerance is still building
Don't try to fix this before the race. Hold volume steady, keep the easy running genuinely easy at 5:23/km or slower, and treat consistency over the next months as the intervention.
VO2max is short of what the goal needs
You've done 2 interval sessions in the last 6 weeks. Add one a week: 5-6 × 3 minutes at 4:02/km with 2-3 minutes jog recovery. Expect a measurable shift in 4-6 weeks.
Threshold pace is the binding constraint
You've done 1 threshold or tempo session in the last 6 weeks. Add one a week: 20-25 minutes continuous at 4:27/km, or 3 × 8 minutes at that pace with 2 minutes easy between. This is the single change that most directly closes the gap.
Surface and route data © OpenStreetMap contributors, licensed under the ODbL. Elevation data © Environment Agency, Open Government Licence v3.