Every so often I pull the data and take an honest look at what the last block of training actually produced. Not what it felt like — what the numbers say. Six months of structured dual-sport training, from mid-September 2025 through mid-March 2026, gives me enough to see the patterns, the progress, and the setbacks.

Here's what 92 activities look like under the microscope.

The Numbers

  • 37 runs — 152.6km total
  • 51 virtual rides — 1,005km on the turbo trainer
  • 4 outdoor rides, 4 walks — and one memorable stroll along "where no fish live"
  • 13 parkruns across 10 different venues
  • 86 active days out of roughly 180
  • Longest streak: 6 consecutive days
  • Longest gap: 22 days over Christmas and New Year

That last number tells a story I'll get to.

The Bike: The Aerobic Engine Room

The turbo trainer continues to be where most of the aerobic work happens. TrainingPeaks Virtual through the autumn, then a switch to Zwift with TrainerRoad workouts from January onwards.

The usual structured fare — Pioneer, Vladeasa, Black, Starr, and the rest. Most sessions ran 45-75 minutes with average heart rate sitting around 110-125bpm. The harder efforts — Heiðarhorn, Abbey, Loreak — pushed into the 130s with spikes above 150.

Month Rides Distance Hours
Sep 5 86km 4.8
Oct 11 213km 11.3
Nov 9 165km 7.6
Dec 5 103km 5.0
Jan 6 105km 4.0
Feb 11 256km 9.8
Mar 4 77km 3.3

October and February were the standout months — over 200km each, fitting 2-3 sessions a week alongside the running. The switch from TrainingPeaks to Zwift/TrainerRoad in January was more about variety than anything fundamental. Zwift's virtual routes add something the plain workout player doesn't — pedalling through virtual New York and Scotland beats staring at a power graph, even if the intervals are the same.

The Running: Parkrun as My Weekly Benchmark

The 13 parkruns across this block became the clearest lens on where my fitness was sitting at any given point. A near-identical 5km effort every Saturday that exposes every fluctuation in form, fatigue, and aerobic capacity.

Date Venue Time Pace Avg HR
20 Sep Dean Castle Country Park 32:38 6:29/km 148
27 Sep Eglinton 32:02 6:21/km 147
11 Oct Palacerigg Country 32:08 6:20/km 152
18 Oct Ayr 34:25 6:33/km 146
1 Nov Troon 32:10 6:25/km 120
8 Nov Drumpellier Country 31:27 6:13/km 149
15 Nov Ayr 32:38 6:15/km 149
22 Nov Ayr #60 33:29 6:24/km 148
6 Dec Lister Park 33:42 6:37/km 145
31 Jan Troon 33:42 6:43/km 157
7 Feb Victoria Park 32:49 6:33/km 153
28 Feb Drumpellier 32:22 6:22/km 150
7 Mar Nostell 33:08 6:33/km 144

The trend tells a clear story. September and October saw steady progress — 32:38 down to 31:27 at Drumpellier in early November, the best of the period. Then the wheels came off.

That Troon parkrun at the end of January? 33:42 with an average heart rate of 157bpm. Working harder for slower times. The fitness I'd built through the autumn had evaporated over the Christmas break.

The Gap: What 22 Days Off Actually Costs

Between 26 December and 17 January, I did nothing. No runs, no rides, no training of any kind. Twenty-two days of complete shutdown that I didn't plan and couldn't quite justify beyond "the holidays happened."

Anyone who trains consistently knows what's coming. My parkrun pace went from a best of 6:13/km in early November to 6:43/km at the end of January — thirty seconds per kilometre slower. Worse, my heart rate was higher for the slower pace. The aerobic base had genuinely regressed.

It took until late February to claw back to somewhere near November form. The Drumpellier parkrun on 28 Feb (6:22/km, 150bpm) was the first time since the break that I felt like the engine was running properly again.

I've had enforced breaks before — injury, illness, life. But this one was just laziness, and the cost was three months of progress for three weeks of sofas. Lesson relearned.

The Midweek Sessions

Between parkruns, the running was the usual mix of prescribed sessions. Fartleks, resolution repeats, descending intervals, strides — Greg's Descending Intervals, Colby's Cutdown, Cool Float. Each targeting a different energy system in the broader plan.

The easy runs sat at 7.0-7.3 min/km with heart rates around 125-130bpm. The interval sessions pushed into the 130-140bpm range. Parkruns were the only time I properly raced, hitting 145-155bpm.

Looking at the data, there were days where my "easy" runs crept above 7.2 min/km pace but my heart rate was still 130+. Those were meant to be easier than they were. Keeping easy days genuinely easy remains an ongoing discipline, even after years of knowing better.

Ten Venues and Counting

One thing that's become a proper habit: parkrun tourism. Across six months I ran at Dean Castle Country Park, Eglinton, Palacerigg Country, Ayr (four times, including my 60th parkrun), Troon, Drumpellier Country (twice), Lister Park, Victoria Park, and Nostell.

Each venue has its own character. Drumpellier is flat and fast — my best time came there. Palacerigg has proper hills at 57m of elevation gain. Lister Park in Bradford was a holiday run — slower, hillier, unfamiliar. Nostell, the most recent, had 64m of climbing tucked into the 5k.

Comparing times across venues is meaningless without accounting for the course. A 33:42 at hilly Lister Park represents a better effort than the same time on flat Troon when I was carrying six weeks of detraining.

What the Heart Rate Data Tells Me

The most revealing metric across any training block isn't pace — it's what the heart rate does at a given pace over time.

Early November parkruns: 6:13-6:25/km at 148-149bpm. Late January: 6:43/km at 157bpm. Same effort ceiling, dramatically less speed for it.

By the end of February, I was back to 6:22/km at 150bpm. Not quite November levels, but close. The aerobic base was rebuilding.

The bike data shows a parallel pattern. My average ride heart rate dropped from around 115bpm in November to 108-110bpm in February for comparable workout durations. The structured riding was doing exactly what it's designed to do — quietly building cardiac efficiency underneath everything else.

What I'd Do Differently

Maintain through December. Even two rides and one run per week would have kept the engine ticking. A complete shutdown was unnecessary and expensive. I know this. I've always known this.

Race more. Thirteen parkruns in six months is decent, but I didn't enter a single event beyond parkrun. A 10k target would have given the training blocks more structure and purpose.

Track the feel, not just the data. I rely on Strava uploads and workout names to reconstruct what happened. A simple training log — even just a post-session "how did that feel?" — would make reviews like this richer. Numbers tell you what happened; they don't always tell you why.

What's Next

The numbers say I'm back to roughly where I was in October. February and March show a steady return to form. The bike is providing a solid aerobic foundation — 255km in February alone — and the parkrun times are trending back toward the low 32s.

The next six months need a target beyond "keep going." A sub-30 parkrun feels ambitious but not impossible from a current 31:27 PB. That probably means more midweek running volume, keeping the bike work consistent, and — critically — not letting December happen again.

Ninety-two activities. One thousand one hundred and fifty-eight kilometres. One lesson I already knew but needed reminding of.

I'm Alan, and I write about running, cycling, and training at WaddingtonWrites. You can find me on Strava.