Using Riegel Formula Race Prediction to Adjust Marathon Training Pace Based on a Recent 10K Finish Time

Can a 50-Minute 10K Actually Tell You Whether to Run 8:30 or 9:15 Pace on Your Next 20-Miler?

You ran a 10K three weeks ago in 50 minutes flat. Now the training plan says "20 miles, steady" and you have no idea what number belongs on your watch. Two friends give you two different answers. One says 8:30 per mile is your marathon pace. The other insists 9:15. That is a 45-second-per-mile spread, and over 26.2 miles it is the difference between a 3:43 and a 4:03. Somebody has to be wrong.

Here is the thing: a 10K is a genuinely good predictor, but only if you push it through the right formula and then read the output correctly. The 50-minute 10K does not tell you 8:30 or 9:15. It tells you something specific and slightly inconvenient — and once you see how the arithmetic works, you can answer the question yourself in about a minute, for any 10K time.

The Formula, in One Line

The Riegel formula is the standard way to scale a race performance from one distance to another:

T2 = T1 × (D2 / D1) ^ 1.06

T1 is the time you actually ran, D1 is that race's distance, and T2 is the predicted time at the new distance D2. The exponent 1.06 is the whole trick: it encodes the fact that pace degrades as distance grows, and it does so at a consistent rate rather than an arbitrary one.

The running pace calculator on this site implements exactly this equation. It carries the four standard distances as constants — 5 km for a 5K, 10 km for a 10K, 21.0975 km for a half marathon, and 42.195 km for a full — so you can feed it a recent race distance and time and read a predicted finish for any of the others.

Why the exponent is 1.06 and not something rounder

Riegel derived it empirically from world-record performances across distances, and 1.06 has held up well enough that it is still the default in most race-time calculators. A larger exponent would predict a harsher slowdown; a smaller one would predict almost none. At 1.06, going from 10 km to 42.195 km multiplies your finishing time by 4.60 — noticeably more than the 4.22 you would get if pace held constant, which is the entire point. You do not run the marathon at 10K pace, and the exponent is what quantifies how much you slow down.

Running the Numbers on a 50-Minute 10K

Convert the 10K time to seconds first: 50 minutes is 3,000 seconds. Then apply the formula.

T2 = 3,000 × (42.195 / 10) ^ 1.06 = 3,000 × 4.6002 = 13,800.7 seconds

That is 230 minutes and 1 second, or 3:50:01. The predicted marathon is a 3:50.

Now convert that prediction into a pace, because a finish time is not something you can hold in your head on a long run. Divide the total time by the marathon distance: 13,800.7 ÷ 42.195 = 327.1 seconds per kilometer, which is 5:27 per kilometer, or 8:46 per mile.

So the answer to the opening question is neither friend. Not 8:30, not 9:15. Your goal marathon pace is 8:46 per mile. The first friend was optimistic by 16 seconds a mile; the second was pessimistic by 29. Over a marathon that is roughly 7 minutes and 13 minutes of difference respectively, which is the gap between a personal best and a bad day.

The "15 to 20 Seconds per Mile" Rule Belongs to a Different Pair of Races

If you have read anything about marathon pacing, you have met this claim: your marathon pace is about 15 to 20 seconds per mile slower than your 10K race pace. Run the arithmetic on the same 50-minute 10K and you get a 27-second-per-kilometer gap, which is 43.6 seconds per mile — more than twice the rule you were told.

The rule is not wrong, though. It is attached to the wrong pair of races.

Where the 15-to-20 number actually comes from

Feed the same 50-minute 10K into the formula for the half marathon distance instead. T2 = 3,000 × (21.0975 / 10) ^ 1.06 = 3,000 × 2.2064 = 6,619 seconds, or 1:50:19. That is a half marathon pace of 5:14 per kilometer.

Now compare the two Riegel predictions to each other: the half marathon at 5:14 per kilometer and the marathon at 5:27 per kilometer. The gap is 13.3 seconds per kilometer, or 21.4 seconds per mile. That is the 15-to-20-seconds rule — it describes the half marathon to marathon step, which is the final and smallest of the pacing jumps. People repeat it as if it applies to a 10K, and then wonder why their predicted marathon pace feels wildly slow.

The gap between 10K pace and marathon pace is not a fixed number at all. It grows as you slow down, because a slower runner is further from the aerobic base a marathon demands.

Turning the Predicted Time into Training Paces

A goal marathon pace is one number; a training week needs several. The standard approach is to anchor everything to the predicted marathon pace and step outward from it. For the 50-minute 10K runner, whose goal pace is 8:46 per mile:

Notice that the 10K time does double duty. It sets the marathon target through the Riegel prediction, and it independently sets your tempo pace, because lactate threshold sits close to 10K race pace for most runners. A stale 10K therefore corrupts both, which is why the next section matters.

Why the Prediction Usually Comes Out a Little Fast

The formula's one real assumption is that you are equally well trained for both distances. Runners rarely are. A 10K is a short, sharp effort that rewards speed and a high pain tolerance; a marathon rewards weekly volume and long-run durability. If you are a fast 10K runner with a thin long-run history, 3:50:01 is not a forecast, it is a ceiling you will not reach — and the known limits of Riegel's accuracy all lean the same way.

The practical fix is a buffer. Add 2 to 4 percent to the predicted time and train against the buffered number. For the 50-minute 10K runner that turns 3:50 into roughly 3:55 to 4:00, or a goal pace of 8:58 to 9:10 per mile. That is slower, and it is the version you should actually use for pacing your long runs, because a training pace that you cannot hold for 20 miles on a Tuesday is not a training pace — it is a race.

Then, if the long runs go well, you can race to the unbuffered prediction. Use the formula as a limiter on your ambition, not as a promise.

Which 10K You Should Feed the Formula

The prediction is only as good as the input, and two kinds of 10K ruin it.

The first is a stale one. Fitness moves in weeks, not months. A 10K from last season tells you about last season. For a marathon prediction, a race run four to six weeks before goal day is the sweet spot: close enough to reflect current fitness, far enough out that you have time to act on the number.

The second is a fake one. A 10K split inside a longer run, or a time trial squeezed into a heavy training week, is not a race result — it is a number produced under fatigue or without full effort. Feed the calculator something you actually raced: a real 10K on a flat course, in decent weather, at full commitment. If your only recent hard effort is a half marathon, use that instead. A half is a better predictor anyway, because it sits much closer to the marathon on the fatigue curve.

Reference Table: 10K Finish Times to Predicted Marathon

The table below applies the formula across a range of 10K results, using the same constants the calculator uses. The final column is the pacing gap you should expect between your 10K and your marathon — watch how it widens as the 10K gets slower.

Two things are worth noticing. First, the predicted marathon time is not a straight multiple of the 10K — a 40:00 10K does not predict twice the time of a 20:00 one, because the exponent applies to the distance ratio, not the time. Second, the pace gap in the last column is not constant. Anyone who quotes you a single number for "how much slower your marathon pace is than your 10K pace" has flattened a curve into a constant, and for slower runners that flattening is badly wrong.

Doing It in Ten Seconds with the Calculator

The arithmetic is not hard, but doing it four times over a training block is. The pace calculator collapses the whole workflow into a couple of entries.

1. Open the calculator and set the mode to Find Time, since you are predicting a finish time from a known distance and effort.
2. Under Recent race distance, enter your 10K — or pick 10K from the race distance presets, which fills in 10 km for you.
3. Enter the finish time in the h:m:s fields: 50 minutes, 0 seconds.
4. Read the predicted marathon time. It should come back as 3:50:01.
5. Switch to Find Pace, enter 42.195 as the distance and 3:50:01 as the time, and read the goal pace of 5:27 per kilometer.

The same tool covers the geometry you will need later in the block: a Find Distance mode for working out how far a given pace carries you in a set time, and preset buttons for 5K, 10K, half and marathon so you are not typing 21.0975 by hand. If you would rather see the shape of it first, the pace table further down the homepage lists finish times for 5K, 10K, half and marathon at common per-kilometer paces — a quick way to sanity-check any prediction that comes out of the calculator before you build a training block on top of it.

What to Carry Away

A recent 10K is a legitimate marathon predictor, but only through the Riegel formula, and only if you read the output as a pace rather than a wish. For a 50-minute 10K, the number is 3:50 and 8:46 per mile — not 8:30, not 9:15. The "15 to 20 seconds per mile" rule you have heard describes the half marathon to marathon step, and quoting it for a 10K doubles the error. Add a small buffer if your long runs are thin, race a real 10K four to six weeks out, and let the calculator turn it into a target you can actually run.