Marathon Pacing Strategies: Four Plans, Real Split Math

The marathon pacing strategies that hold up through the last 10 km tend to share one shape. You start a few seconds per kilometre slower than goal pace, hold even effort from about 5 km to 30 km, and race only the final 10 km. Even splits and small negative splits both fit that shape. Banking time with a fast first half doesn't, and the arithmetic shows why: every minute you bank buys less than 3 seconds per kilometre of slack later.

Four pacing plans, one 4:00:00 finish

Most marathon pacing advice comes down to four plans. The difference between them is how the two halves compare:

• Even split: both halves take the same time. For a four-hour finish that's 2:00:00 + 2:00:00.
• Negative split: the second half is faster than the first. The version below is a 1% negative split, with the second half 72 seconds quicker.
• Controlled start: the first 5 km is deliberately slow, then a steady pace absorbs the deficit.
• Banked time (positive split): a fast first half builds a cushion for the expected fade.

All four finish at exactly 4:00:00 on paper. Here are the cumulative clock times at the checkpoints most races mark:

The paces behind each column:

• Even: 5:41.3/km the whole way.
• Negative split: 5:43.0/km to halfway, then 5:39.6/km.
• Controlled start: 5:50/km for 5 km, then about 5:40/km.
• Banked time: 5:30/km to halfway, then 5:52.6/km.

What the table tells you at 10 km

At 10 km the four plans are 2 minutes 30 seconds apart. At 40 km they're 29 seconds apart.

So your strategy shows up early and blurs late. Say you're on a 4:00 plan and pass 10 km in 55:00. You're not "a bit ahead." You're running a different plan from the one on your wrist, and it's the plan with the least room for error.

Here's a rule you can use. Choose the plan before race day and write its 10 km and halfway clock times on your band. If you're more than about 30 seconds off at 10 km, take that as a reason to adjust, not as a bonus. A marathon pace band calculator will print cumulative checkpoints for any of the four columns.

Why a controlled start barely differs from even pacing

Running the first 5 km at 5:50 instead of 5:41 costs 44 seconds. Spread over the remaining 37.2 km, that's about 1.2 s/km, roughly the gap between 5:41 and 5:40 on your watch. The slow start feels like a sacrifice in the corral. On the clock, it hardly registers.

Banking time: a loan at 2.8 s/km per minute

Banking time sounds like insurance. The numbers make it look more like a bad loan.

A marathon's second half is 21.0975 km. Each minute you bank in the first half gives you 60 ÷ 21.0975 = 2.84 s/km of slack in the second. In the banked-time column, the runner holds 5:30/km to halfway, 11.3 s/km faster than even pace, and arrives with a 3:58 cushion. That cushion allows a 5:52.6/km second half with the finish still at 4:00:00.

So the whole bet comes down to one question. After running 21 km at 11 s/km faster than your fitness suggests, will your second half slow by less than 11 s/km? Few training plans let you test that honestly before race day.

The small version of the mistake is harder to notice. Say you get excited in the corral and run 10 s/km too fast for the first 10 km. You've banked 100 seconds. Over the remaining 32.2 km, those 100 seconds cover a slowdown of just 3.1 s/km. On a warm, windy or hilly day, that's gone before 30 km.

What high-level runners actually do

A 2024 systematic review of marathon pacing strategies by Sha and colleagues describes the race strategy among high-level runners as holding a steady speed through the first part of the race, then slowing slightly in the last phase. That's neither a big negative split nor banked time. It's even pacing with a small, late fade, close to the Even and Controlled start columns above.

The competing advice is closer than the headlines suggest. Trenara's pacing article argues that negative splits win races. Even so, its first phase is labelled "Start Controlled (0–10K)". Both the evidence and the negative-split camp reject the fast start. Where they differ is how much faster the second half should be, and that gap is smaller than it sounds.

When a big negative split means you set the goal too low

Take a 3-minute negative split on a 4:00 marathon. That means 5:45.5/km to halfway and 5:37.0/km home. If you can run 21.1 km at 5:37/km at the end of a marathon, the time you lost was in the first half. A large negative split tells you the goal was conservative. It doesn't tell you the strategy was clever.

How slow is a "slow start"? Two rules, 61% apart

Two widely read pacing articles give slow-start advice that looks identical. RunnersConnect says to run 5 to 10 seconds per mile slower than goal pace for the first three or four miles. Trenara says 5 to 10 seconds per kilometre slower. Because the units differ, one rule is 1.61 times the other.

What matters is the payback. You repay a 40-second deficit over the remaining 36 km at about 1.1 s/km, which is too small to notice. You repay a 100-second deficit over 32.2 km at 3.1 s/km.

That last number should look familiar. It's the same 3.1 s/km that banking 100 seconds would let you lose later. A slow start and a fast start are the same arithmetic with the sign flipped. The difference is timing. A deficit asks for something at the end, when you know how the day is going. A cushion commits you at the start, when you don't.

If you're unsure about your fitness, use the per-mile version. Stretch the per-km version across a full 10 km only if you've planned a real negative split and have run goal pace or faster at the end of long runs.

The 10-10-10 strategy, phase by phase

The 10-10-10 plan splits the marathon into 10 miles, 10 miles and 10 km: 16.09 + 16.09 + 10.01 km = 42.195 km. Each phase gets one job, and together they're meant to stop the fast start. The usual version holds back in the first 10 miles, locks onto goal pace for the next 10, and races whatever is left over the last 10 km.

Here it is on a 4:00:00 goal, holding back 3 s/km in phase one:

Holding back only 3 s/km costs 48 seconds over 10 miles. Getting them back in the final 10 km means running 4.8 s/km faster than goal pace while at your most tired.

Here's how to decide. Use the full 10-10-10 structure only if you've run a strong final 10 km at the end of your long runs. If you haven't, run phase one at goal pace and treat phase three as "hold on." That turns it into an even-split plan with a named checkpoint at 20 miles, which still helps.

Pacing a 4-hour and a 5-hour marathon

4:00:00 at 5:41/km or 9:09/mile

Even pace is 5:41.3/km, or 9:09.2/mile. Your checkpoints are 56:53 at 10 km, 2:00:00 at halfway and 3:03:05 at 20 miles. Use the Even or Controlled start column from the first table as your band.

The most common mistake at this goal involves pace groups. If you follow a 4:00 pacer, find out whether the group targets gun time or chip time. Suppose it targets gun time and starts four minutes after the gun. The pacer then has to run a 3:56 chip time, which is 5:35.6/km. Stay with that group and you're running 5.7 s/km faster than your plan. That's about the size of a typical slow-start discount, pointed the wrong way.

5:00:00 at 7:07/km or 11:26/mile

Even pace is 7:06.6/km, or 11:26.5/mile.

Run-walk: how fast the running segments have to be

A lot of 5-hour runners use run-walk intervals. The common error is running at 7:07/km between walks, so each walk break adds to the deficit. Your running segments need to be faster than your average pace. How much faster depends on the ratio. The figures below assume a brisk 10:00/km walk:

Going from 9:1 to 2:1 moves the required running pace by 40 s/km. Test that running pace on your last long runs. If it feels hard at 25 km, the ratio is wrong, not the goal.

Common marathon pacing mistakes and how to spot them by 10 km

Most pacing mistakes show up before halfway, as long as you know which number to check.

The watch-distance problem gets bigger as the race goes on. Certified road courses are measured along the shortest possible route a runner could take without being disqualified, and the IAAF/AIMS measurement manual notes that any particular runner is most unlikely to follow it. Say your watch reads 42.6 km at the finish. That's 405 m your average pace never accounted for, or 2:18 of running at 4:00 pace. Cumulative clock times at the course markers avoid the problem. Your watch's "current pace" figure doesn't.

A lost minute leads to another common mistake. You stop at a toilet at 25 km and lose 60 seconds. Win them back over the next 2 km and you're running 30 s/km faster than goal. Spread them over the remaining 17.2 km and it's 3.5 s/km. The first choice turns a one-minute problem into a race-ending one.

Adjusting your plan for an October or November marathon

If you're racing this autumn, three things matter more than which strategy you pick.

Cold corrals can make the surge more likely. After twenty minutes of standing around in cool air, goal pace can feel easy for the first few kilometres. Stick to the 5 km clock time on your band, whatever that pace feels like.

Check the forecast for your 30 km split, not just for the start time. A 4:00 runner whose wave starts at 09:00 reaches 30 km at about 11:51. The forecast for that hour is what shapes the last 12 km. TrainingFlow's race guides list typical race-day ranges such as 4–13 °C for Philadelphia and 6–17 °C for Twin Cities. That spread is big enough that 10:00 and 12:00 can feel like different races.

Treat wind as a pacing input. On exposed stretches into a headwind, tuck into a group and let your pace per kilometre drift. Holding goal pace alone into the wind can cost more than the seconds it saves.

A pace band only works if the rest of race morning goes to plan. The race day checklist for marathons covers gear, fuel and logistics. If you want checkpoints built from your VDOT (Jack Daniels' score that converts a recent race result into equivalent paces), the course elevation and the race's climate baseline, the free marathon pacing preview shows what TrainingFlow produces before you decide whether to use it.

FAQ

What is the 10-10-10 marathon pacing strategy and what is its goal?

It divides the marathon into 10 miles, 10 miles and 10 km. You run the first phase with restraint, the second at goal pace, and race the last 10 km with whatever is left. The goal is to prevent a fast start by giving each segment one job. On a 4:00 plan, holding back 3 s/km for the first 10 miles means running the final 10 km at about 5:36/km.

What are common marathon pacing mistakes?

The costly ones are surging in the first 5 km, pacing from watch distance instead of course markers, and following a pace group that targets gun time when you care about chip time. Holding flat pace on hills and trying to win back a lost minute within two kilometres are also common. Most of these show up as a 5 km or 10 km split more than about 30 seconds off plan.

What is a good pacing strategy for a 4-hour marathon?

Run even or near-even at 5:41/km (9:09/mile), with the first 5 km a few seconds per kilometre slower. Your checkpoints are 56:53 at 10 km, 2:00:00 at halfway and 3:03:05 at 20 miles. If you pass 10 km faster than about 56:20, you're running a banked-time plan whether you meant to or not.

What is a good pacing strategy for a 5-hour marathon?

Average 7:07/km (11:26/mile) and reach halfway at 2:30:00. With run-walk intervals, the running segments have to be faster: about 6:53/km on a 9:1 ratio and 6:38/km on 4:1, assuming a 10:00/km walk. Check that you can hold that running pace on your long runs before choosing a ratio.