Ramp back in after a December break.

Start from the weeks you actually trained

A December break leaves a gap between the volume you remember and the volume your last few weeks show. A ramp starts from the second number. This guide takes a January return one decision at a time, then works two eight-week examples line by line: one in minutes and one in kilometres. Every figure can be reproduced with a calculator, and the same arithmetic runs in the January training ramp planner.

The method has five parts: a baseline, a cap, a length, a deload rhythm and a split across your days. None of them has a single correct setting. The point of writing them down is that you choose them before the table starts to look persuasive.

Step 1: choose a baseline

The baseline is the weekly volume the first step is measured from. There are three sensible choices, and the planner offers all of them.

• Your most recent entered week. Use this when last week is a fair picture of where you are now, for example after a few weeks of easy movement.
• The average of the weeks you entered. Use this when the last four weeks bounce around and no single one tells the story.
• A starting figure you type. Use this when you know what week 1 should be built on, whatever the log says.

Missing weeks matter. A blank week is a week with no record, and it is skipped. A zero is a week you took off, and it counts. If you log 100, blank, blank and 140, the most recent entered week is 140 and the average is (100 + 140) ÷ 2 = 120. The two baselines differ by 20, and that difference carries straight into every later week.

If there is nothing to go on at all, the maths has no percentage to start from. The planner then steps up in equal amounts from zero, shows no percentage for week 1, and measures every later step against your cap.

Step 2: set a cap that is yours

The cap is the largest week-to-week increase you want to see in the table. No universal figure exists. People pick different numbers depending on their sport, their history, the time of year and how a given week feels, and the planner shows 10% only as an editable starting value, not as a recommendation.

What the cap does is narrow. For every week the planner works out the change, (this week − reference week) ÷ reference week, rounds it to one decimal and compares it with your number. Steps above it are marked. The table does not move the plan to fit the cap; it reports where each step sits so you can decide what to do about it.

Write the cap down before you look at the result. A number chosen after the table is easy to bend toward the plan you already like.

Step 3: choose the number of weeks

The planner accepts 4 to 16 weeks. The length decides how big each step has to be, because a fixed distance covered in fewer weeks needs larger steps. There are two ways to spread the distance.

Equal percentages: step = (target ÷ baseline)^(1 ÷ build weeks) − 1

Equal amounts: step = (target − baseline) ÷ build weeks

Equal percentages multiply each build week by the same factor, so the absolute jumps grow as the weeks go on while the percentage stays level. Equal amounts add the same figure each week, so the percentage is largest at the start and shrinks. Neither is better in itself. If your cap is a percentage, equal percentages keep every step the same distance from it; equal amounts may put the early weeks above it.

Build weeks are the weeks that remain once deloads are removed. An eight-week plan with one deload has seven build weeks, and the exponent in the formula is 1 ÷ 7.

Step 4: place the deload weeks

A deload rhythm of every 3rd, 4th or 5th week puts a lighter week at those positions. The last week is never a deload, so the plan finishes on your target. A deload week is a share of the last build week; the planner starts at 80% and lets you change it.

The week after a deload raises one question: which week is it compared with? Measured against the deload week, the rebound looks like a large step. Measured against the last build week, it looks like the normal step it is. The planner defaults to the last build week and has a setting for the other view. In the minutes example below, week 5 is 110.1: against the deload week 4 (81.3) that is +35.4%, and against build week 3 (101.7) it is +8.3%.

Step 5: split each week across your days

Pick the days you can train and the share of the week that goes into the long session. The remainder is divided evenly across the other days. The planner works in tenths of a unit, so the sessions add up to exactly the week shown, with any spare tenth going to the earliest day.

One thing to watch: with two days, a long-session share under 50% makes the long session the shorter one. The planner says so when that happens.

Worked example: 8 weeks in minutes

Last four weeks, oldest first: 150, 120, 60 and 80 minutes. The break shows in weeks 3 and 4. Baseline: the most recent entered week, 80. Target: 140 minutes a week. Eight weeks starting 4 January 2027, cap 10%, deload every 4th week at 80%, equal percentages. Days: Tue, Thu, Sat and Sun, with a 35% long session on Sunday.

Week 4 is the only deload, so there are 7 build weeks and the step is (140 ÷ 80)^(1 ÷ 7) − 1 = 8.32%. That sits under the 10% cap, so no week is marked.

Reading the table: week 1 is 80 × (1 + 8.32%) = 86.7. Week 4 is the deload, 101.7 × 80% = 81.3, and week 5 resumes from week 3's 101.7. Week 8 lands on the 140 target. The eight weeks add up to 862.2 minutes.

Splitting week 1 (86.7 minutes) across four days with a 35% long session: the Sunday session is 35% of 86.7, 30.3 minutes. The other 56.4 minutes go to three days at 18.8 each. The total is 86.7, matching the week.

Try compressing it. With the same inputs and 6 weeks, the deload still falls in week 4, leaving 5 build weeks, and the step rises to 11.84%. Against a 10% cap, 5 weeks are marked. You can lengthen the plan, lower the target or raise the cap, or edit individual weeks. All three are your call.

Worked example: 8 weeks in kilometres

Last four weeks, oldest first: 28.5, 24, 12.5 and 18 km. This time the baseline is the average of the four entered weeks: (28.5 + 24 + 12.5 + 18) ÷ 4 = 20.75. Target: 32 km a week. Eight weeks starting 28 December 2026, so the labels cross into 2027 at week 2. Cap 9%, deload every 3rd week at 80%, equal percentages.

Deloads fall in weeks 3 and 6, leaving 6 build weeks. The step is (32 ÷ 20.75)^(1 ÷ 6) − 1 = 7.49%, under the 9% cap.

Notice what the baseline choice does. Had the baseline been last week alone, 18 km, the same target in the same eight weeks needs a step of 10.06% and 6 weeks sit above the 9% cap. Nothing is wrong with either baseline; they answer different questions, and the table makes the difference visible. Stretching the plan to 10 weeks with a last-week baseline brings the step to 8.57% with 0 weeks marked.

Week 1 begins on 28 December 2026 and week 2 on 4 January 2027. The planner prints the year on every date and tags the first week that falls in 2027 in the table, and the CSV export carries ISO dates such as 2026-12-28 and 2027-01-04, so a spreadsheet sorts them correctly.

When the table marks a week

A marked week is information, not a verdict. It means that step is larger than the cap you set. The usual levers are the same ones each time: more weeks, a lower target, a higher cap, or a different baseline. You can also edit a single week. An edited week keeps the value you typed, the other weeks keep their planned values, and the change column updates for the edited week and the one after it, so you can see at once what your edit did to the neighbours.

Targets below where you are now

If the target is under your baseline, the table steps down evenly instead. Deload weeks are left out because the whole path is already reducing, and nothing is compared with the cap, which only looks at increases. Taper-style weeks work the same way: the arithmetic is symmetrical, with a negative step.

Switching units

Enter a typical pace in minutes per km and the planner can show the same plan in either unit. It converts from the figure you typed, not from a rounded copy, so going from minutes to kilometres and back returns exactly what you started with. Percentage changes are identical in both units because the pace cancels out.

Take it to paper

The planner exports a CSV with one row per week: week number, start date, type, volume, change, whether the step is over your cap, whether you edited it, and a column for each training day. It also prints a one-page ramp sheet on A4 or Letter with a done box and a write-in column for what you actually did, plus any notes you type.

If a twelve-week block is what you need, the winter training plan lays out base, build and specific weeks. When the weeks are set and race day is near, the split planner turns a target finish time into splits, and the pace band calculator prints a wrist band.

More from the course