The last third of your run is either defense or attack, depending on how you spent the first, and the progression run trains exactly that choice.
- Anyone who ends up defending in the last third of a run has almost always spent too much in the first few miles, when fresh legs hand you a pace that feels free.
- A progression run splits the session into blocks: you start below your usual pace and step up slightly at every change, with the last stretch the fastest.
- You set the opening pace by your breathing and confirm it with the watch afterward, never before.
- Four weeks of ready-made sessions, plus two cues you can feel on the run to tell whether you’re holding to the plan.
The Mile Where the Run Turns Into Defense
You head out meaning to keep it easy. The first miles roll by without your thinking about them, the legs are fresh, the watch shows a better number than the one you had in mind, and the fatigue hasn’t arrived yet. Then, somewhere in the last third, pace stops being a choice and becomes something you defend: shoulders creep up, breathing shortens, and you run the closing miles only to be done with them.
What gives way in that stretch is the effort distribution you settled in the first five minutes, more than your condition on the day. The difference changes what you do about it: condition takes months to build, distribution trains in four weeks.
Starting below the pace that would come naturally leaves more sugar available in the muscles for the closing stretch and keeps core temperature lower through the early part of the run. It also slows cardiac drift, that gradual rise in heart rate that shows up after a while even at a steady speed. These are effects that research on marathon pacing describes together, because they all act on the same point of the race.
A restrained start is a technical decision, then, and it deserves the discipline you’d give an interval. The demanding part of a progression run sits in the first ten minutes, when fresh legs offer you a pace you could hold and you decide not to take it. For anyone who fades late, this session pays off faster than a block of intervals: intervals train the engine, progression runs train the foot on the accelerator.
How to Build a Progression Run
Split the session into three blocks of equal length and run each block one step faster than the one before it. The warm-up stays outside the count: ten minutes of very slow running before the first block starts.
The number of blocks matters less than the rule holding them together, which is that no block can be slower than the one preceding it. If the pace drops at any point, the progression ended there, and what you run from that moment on is an easy run.
The Opening Pace and the Size of the Steps
Set the opening pace by your breathing and use the watch only to confirm it afterward. At the end of the first block you should be able to speak a full sentence without breaking it to catch your breath: if that takes effort, you’ve already gone out too hard. It’s the same cue that governs zone 2 work, with one difference — there the slow pace is the ceiling of the session, here it’s the floor you climb from.
The step from one block to the next should stay small, on the order of fifteen seconds per mile, twenty-five at the most. Go wider and the last block becomes a time trial, and the session stops training effort distribution and starts training fatigue tolerance, which is a different goal with its own tools. There’s an opposite risk too: stepping up too early produces a run spent almost entirely at an in-between pace, the moderately quick junk miles that overload you without building anything.
On the run you have two cues worth more than any number on the display.
Breathing in steps.
If the progression is right, your breath climbs one step at every block change and then settles: you notice it about a minute after you accelerate, not the instant you do. When breathing instead takes a continuous ramp and never finds a stable level, you’re stepping up too much or you started too fast.
The margin.
At the end of every block except the last, you should feel you could still go up. The sensation is specific: the legs answer when you raise your cadence, with no need to lengthen the stride to move faster. If that margin is missing at the end of the second-to-last block, the last block won’t happen.
The Four Weeks, Session by Session
One progression session a week, always on the same day, with at least two days between it and any other hard work. The other runs stay easy and don’t change. Before the blocks, ten minutes of very slow running; after the last one, five minutes of cool-down.
The three pace levels in the table are relative to your own habitual pace:
- Below: slower than your everyday easy run, a pace where talking costs you nothing.
- Easy: your easy running pace, the one you’d use on a recovery day.
- Above: one step beyond easy, the pace you could hold for about an hour if you had to.
| Week | Blocks | Pace progression | What it trains |
|---|---|---|---|
| 1 | 3 x 10 minutes | below / easy / just above | starting slower than you want to |
| 2 | 3 x 12 minutes | below / easy / above | holding the fast block longer |
| 3 | 4 x 10 minutes | below / easy / just above / above | managing three changes instead of two |
| 4 | 20 + 15 + 10 minutes | below / easy / above | closing with the fatigue already in you |
Week four flips the durations, the blocks shorten as the pace rises, and the fast stretch arrives after thirty-five minutes of running already done. It’s the structure closest to a race, and it’s also how progression later enters the weekly long run, where the final twenty or thirty minutes go faster than everything before them. If the three blocks feel manageable after four weeks, the way forward is to lengthen the closing block rather than raise its pace.
What to Do If the Second Block Is Already Too Fast
Noticing it mid-session happens often in the first two weeks and doesn’t call for stopping. The correction that would come naturally, slowing down to get back to the planned pace, is the worst one: it breaks the rule that holds the session together and trains you to fix an opening error by braking, which in a race is exactly the mechanism you want to get away from.
Hold the pace you’ve taken through to the end of the block instead, without adding a single second, then halve the length of the last block and run it just barely faster, even by ten seconds a mile. The progression stays intact and you pay for the error in volume rather than in structure.
If it happens three sessions in a row, the problem is in your starting cue. Cover the display for the first ten minutes and let your breathing decide: the pace that comes out will be slower than the one you’d have chosen looking at the watch, and it’s the right one.