You followed the plan. You did most of the long runs. You didn't skip too many sessions. Race day came, and something went wrong — you hit the wall at km 32, you finished 20 minutes slower than your goal, or you didn't finish at all.
The conclusion most runners reach: I wasn't fit enough. I needed to train harder.
That conclusion is usually wrong. The plan was the problem.
I've been through this twice. The first time, a heavy work month in weeks 13–15 meant five or six missed sessions and consistently poor sleep. I followed the taper as written and arrived at race day feeling flat rather than fresh. My TSB was still negative — I could see that later when I analysed the data. At the time, I just thought I'd trained poorly. I hadn't. I'd trained fine. The plan just had no way to know I needed a longer taper.
What a training plan actually assumes about you
Every static marathon training plan is built on a set of assumptions that are never written down:
- That you will complete every session as prescribed, every week, for 12–18 weeks
- That your life will be stable enough to allow this
- That you will sleep consistently, absorb training stress at a predictable rate, and recover on schedule
- That your current fitness exactly matches the plan's starting point
- That if you miss a session, the next week will simply continue as if nothing happened
None of these assumptions are true for any real runner over an 18-week training cycle.
Work gets heavier. Travel comes up. You get sick for five days. You miss a week, come back too fast, and accumulate fatigue that the plan never accounts for.
The plan keeps marching forward regardless. It doesn't know you were ill. It doesn't know your CTL dropped by 15 points while you were in bed. It just says: Tuesday, threshold intervals, 8 × 1000m at T pace.
The week that derails everything
Most runners can identify the week that broke their training. It's usually not dramatic — not a major injury or a crisis. It's a conference that meant three missed sessions. A cold that lasted ten days. A family situation that compressed sleep to five hours a night for two weeks.
After that week, the plan is already wrong.
If you were on week 9 of an 18-week plan and lost ten days, you no longer have the fitness base that week 10 assumes. The long run in week 10 is prescribed for a runner with the aerobic base that week 9 was supposed to build. You don't have that base. You have something lower.
What do most runners do? They either skip ahead and follow the plan as written, or they repeat a week and feel behind. Neither is correct, because neither responds to where your fitness actually is.
Why the long run doesn't mean what you think it means
The long run is the centerpiece of most marathon plans. Twenty miles, three weeks out. Everyone treats it as the key session.
The problem is that a long run is only the right stimulus when your aerobic base supports it. A 32 km run on a fatigued body doesn't build fitness — it depletes it. The physiological signal you get from a long run when your ATL (acute training load) is too high and your TSB (training stress balance) is deeply negative is not the same as the signal you get when you're appropriately fresh.
The plan doesn't know this. It just says: Week 14, Sunday, 32 km.
If you arrived at that session carrying three weeks of accumulated fatigue from a heavier-than-expected build phase, the long run becomes a net negative. You spend the following week recovering instead of absorbing the fitness gain.
The pacing problem nobody talks about
Most training plans give you a goal marathon pace and work backwards from it. If you want to run 3:45, your long runs are at a certain pace, your threshold sessions are at a certain pace, your easy runs are at a certain pace.
But this goal pace was set at the start of the plan, before you trained. It assumed a VDOT — your effective aerobic capacity — that you were supposed to reach by race day through the plan itself.
What actually happens:
- Some runners improve faster than the plan predicted → the prescribed paces are too easy, they're undertrained relative to their potential
- Some runners improve slower → the prescribed paces are too hard, they accumulate excessive fatigue
- Almost no runner improves at exactly the rate the plan assumed
When you're running threshold sessions at a pace that's 15 seconds/km too hard for your current fitness, you're not building your lactate threshold — you're just accumulating fatigue. The session feels productive because it's hard. It isn't.
The taper that doesn't work
Taper is supposed to shed accumulated fatigue (ATL) while preserving fitness (CTL), so that you arrive at race day with positive TSB — fresh legs, high fitness.
But if the preceding training block was miscalibrated — if you were running too hard, accumulating more fatigue than the plan intended — then the two-week taper isn't long enough to clear it. You arrive at race day with TSB still negative, legs heavy, and no explanation for why you feel terrible after tapering.
The answer is usually one of two things: the training load was too high for your fitness level, or the taper was too short given how fatigued you actually were. A good plan would have seen both of those problems coming.
What actually needs to happen
The solution is not to follow a better template. It's to use a plan that responds to what's actually happening in your training.
That means a plan that tracks your CTL (fitness), ATL (fatigue), and TSB (form) week by week. One that knows when you missed sessions and recalibrates. One that uses your actual VDOT — recalculated as your fitness changes — to set your paces, not the VDOT you were assumed to have at the start.
When you had a hard week and your ATL is elevated, the plan should respond by reducing intensity on the next session — not because you feel tired, but because the data shows you need recovery before the next training stimulus will be absorbed.
When you missed ten days due to illness, the plan should rebuild your base conservatively rather than pushing you into a peak week designed for a CTL you no longer have.
This is not about being easier. It's about being accurate. A plan that responds to your actual physiology will push you hard when the data shows you're ready for it — and those hard sessions will produce the adaptation they're designed to produce, because you arrived at them appropriately prepared.
The runner who follows the plan perfectly and still fails
There's a final scenario worth addressing: the runner who genuinely does everything right and still doesn't hit their goal.
Sometimes the plan was calibrated correctly from the start and the goal was simply wrong. The predicted marathon time was based on a VDOT that assumed a longer training cycle, a higher peak volume, or a more experienced runner.
Chasing a goal time that requires VDOT 52 when you currently have VDOT 46 is not a training problem. It's a goal-setting problem. A plan that knows your VDOT going in — and recalculates it as you train — will tell you early whether the goal is realistic. You can adjust the goal with enough time to prepare correctly, instead of discovering the gap at kilometer 32.
The science behind training load and fatigue modelling is well-established. Banister's original impulse-response model (1975) and subsequent work by Coggan on the Performance Manager Chart underpin how CTL, ATL, and TSB are calculated. For a deep dive, Stephen Seiler's research on polarised training and intensity distribution is also worth reading — it explains why most amateur runners spend too much time in the middle zones, exactly the pattern static plans create.
If you recognise any of this in your own training history, you're not unusual. Most intermediate runners have had at least one race where something like this happened. The plan wasn't built for the runner you actually are — it was built for an idealised version of you who never has bad weeks.
Zarkus is built for the runner who does. Join the waitlist below.