Every version of this question online gets answered with an average. The average marathon time is around four and a half hours, so anything quicker is good — and that is true, useless, and slightly depressing, because it tells you nothing about you.
The useful question is different: what is a good marathon time for someone who runs what you run? That has a real answer, it is computable from a race you have already done, and it is the number worth aiming at.
Your own shorter races are the benchmark
Distance-running performances are strongly related to one another. Someone who runs a 22-minute 5K is not going to run a 3-hour marathon, and someone who runs a 20-minute 5K is not going to struggle to break five hours. The relationship is close enough to be modelled, which is what a race-time predictor does.
The table below is generated from the same prediction model that runs the Verus race time predictor, so you can check any row of it yourself. Find the shorter time you have actually run, and read across.
| 5K | 10K | Half marathon | Equivalent marathon |
|---|---|---|---|
| 18:46 | 39:08 | 1:26:20 | 3:00:00 |
| 21:54 | 45:39 | 1:40:43 | 3:30:00 |
| 25:01 | 52:10 | 1:55:07 | 4:00:00 |
| 28:09 | 58:42 | 2:09:30 | 4:30:00 |
| 31:17 | 1:05:13 | 2:23:53 | 5:00:00 |
How to read this honestly
These are equivalents, not promises. They assume marathon-appropriate training and a reasonable day. A runner with a fast 5K and a 25-mile-a-week habit will run slower than the table says; a patient, high-mileage runner will often beat it. The table tells you what is plausible, not what is owed to you.
The round numbers, and what they mean
Marathon goals cluster around round numbers, which is entirely irrational and completely universal. Here is what each of the common ones actually represents.
| Target | Pace per mile | Pace per km | What it usually represents |
|---|---|---|---|
| Sub-5:00 | 11:26 | 7:07 | A well-executed first marathon on consistent training. A genuine achievement and the most common realistic first goal. |
| Sub-4:30 | 10:18 | 6:24 | Comfortably above the field average, and reachable for most people who complete a full block properly. |
| Sub-4:00 | 9:09 | 5:41 | The most common serious target in the sport. Needs structure — not just mileage, but some of it run fast. |
| Sub-3:30 | 8:00 | 4:58 | Committed, structured training across a year rather than a block. A recognisable club-runner standard. |
| Sub-3:00 | 6:52 | 4:16 | Years of consistent, high-volume training for almost everybody who does it. |
Sub-4:00 gets the most attention, and deserves it — it is the point where a marathon stops being purely about completing the distance and starts requiring the distance to be raced. There is a separate guide on exactly how to get there.
What is a good first marathon time?
Finishing. That is not a consolation answer, it is the correct one, and here is why in concrete terms.
In a first marathon you have no idea what the last six miles feel like, because nothing in training tells you — the longest run in most plans is twenty miles and the distinctive part of a marathon is what happens after that. You are also, statistically, likely to run the first half too fast. A first-timer who targets a time typically compounds those two problems: they set an ambitious number, chase it early because it feels easy, and discover mile twenty-two.
- 1 Set a range rather than a time. "Between 4:15 and 4:45, aiming for the middle" is a plan; "4:15" is a trap.
- 2 Pace the first half deliberately slower than the equivalence table suggests. You are buying insurance against the part of the race you have never run.
- 3 Treat the finish time as data collection for the second one. Almost everybody's second marathon is significantly faster than their first, and it is not because they got fitter.
Age, and what a good time means over time
Endurance holds up better with age than speed does, which is why marathon fields skew older than 5K fields and why plenty of people run lifetime bests in their forties. If you started running at 45, your best marathon is almost certainly still ahead of you.
The more useful framing than age-graded tables, for most people, is direction of travel. A time that is faster than your last one, run on training you could sustain, in a race you paced properly, is a good marathon time. A time that beat someone else's benchmark but broke you for a month is not.
The same fitness produces different times
Two things move a marathon time by more than most people expect, and neither is fitness.
The first is the course. Among UK marathons the spread is real: Manchester is widely regarded as the flattest major in the country at around 121ft of elevation gain, Chester sits at around 249ft, London at around 419ft, and Brighton at around 477ft with a genuine climb in the first kilometre. Those are not equivalent afternoons.
The second is the weather, and it is the larger of the two. The coaching rule of thumb adds air temperature and dew point in Fahrenheit: below 100 there is no real penalty, 130-140 costs 2-3%, and 140-150 costs 3-4.5%. On a four-hour marathon, 3% is over seven minutes. A personal best attempt on a warm day is frequently a personal best attempt that was decided before the gun.
Setting a target you will actually hit
- 1 Start from a real result. A recent 10K or half marathon, not a training run and not a time you ran three years ago.
- 2 Convert it, then discount it if your weekly mileage is modest. Equivalence tables assume the training that supports the distance.
- 3 Sanity-check it against the conditions and the course you have actually entered.
- 4 Turn it into splits before race day, so the plan exists on paper rather than in your head at mile three.
- 5 Write it down somewhere other people can see. A target nobody knows about is easy to quietly revise at 6am in January.
On the last of those — a Verus profile exists to hold exactly that: the goal, the date, and the distance between where you started and where you are now, in public. It is a small mechanism and it works for the same reason telling a friend works, except the friend does not forget.