Training for my first marathon while backpacking South America (2009)

In 2009 I made a decision that made sense at the time and less sense on race day: I signed up for my first marathon, in Chicago that October, and then did most of the training while backpacking through South America. The trip ran for 92 days, from 3 May to 3 August, through Venezuela, Colombia, Ecuador and the Galápagos, Peru, Bolivia, Chile, Argentina and Paraguay.

This is not a race report; that is Chicago, split by split. It is about the build-up: what training on the move looked like, what my diaries from the road recorded, and what I would do differently if I tried it again.

The timeline

3 May 2009: left London for Caracas.
Late June: the Machu Picchu trek, where I first wrote of the marathon "I have diabolically chosen to run in October".
Around the end of June: by my own count afterwards, training began "some 3.5 months" before the race.
July: Bolivia, including the cycle down the Yungas road and a stint at the La Senda Verde animal refuge.
3 August: back in London, where the rest of the block happened.
11 October: race day in Chicago.

The honest summary is that much of the meaningful running happened in the last two months, in London, and the South American weeks were more about staying active than following a plan.

What training on the road actually involved

When the race was over, I wrote a one-paragraph summary of the whole build. It is the best record I have:

"Having trained across the world for this event, with train tracks, rabid dogs, construction sites, battered plimsoles, excrement ridden parks, a dodgy knee, swimming for my life and going out to the States for my first marathon – its been a wonderful journey."

Every phrase in that list is a constraint. There was no measured route, no running club and no reliable surface. I ran where I could: along railway lines, past building sites and through city parks. The shoes were not proper running shoes. There was a knee niggle. And there was swimming, some of which, by my description, felt less like cross-training than survival.

Altitude and terrain

A good part of the route was high: Cusco, Lake Titicaca, La Paz, Potosí. I did not record specific runs at altitude, so I will not invent any. What the diaries do record is the general physical load of the trip: multi-day treks with eight-hour walking days on the way to Machu Picchu, long bike descents, and plenty of time on my feet. In my Machu Picchu diary I described a long, uncomfortable walking day as "like the marathon... a good test of mental strength". I was already thinking of the trip as preparation, even when it was not structured preparation.

Everything else a backpacker deals with

Stomach bugs, late nights, irregular food and overnight buses all appear in the diaries. None of them helps a training block. I would not dwell on them except that they are the real reason improvised training underdelivers: the runs themselves are only part of it.

What it cost on race day

The Chicago splits show a runner on course for about 4:10 at halfway who slowed by several minutes per 5K after 30K and finished in 4:25:55. My own race diary said I set off too fast. I would add, looking back, that a stop-start training block on the road probably left me with enough endurance to finish but not enough to hold pace late on. I cannot prove the second part from the data; it is my judgement.

Five months later, after a winter of training at home, I ran Barcelona in about 3:45.

If you are training while travelling

These are lessons from my own trip, not medical or coaching advice.

Bring proper shoes. "Battered plimsoles" are in my list for a reason.
Count time on your feet. Treks, walking days and cycling all add load, which can help or hurt. Note it down.
Be realistic about the last two months. If most of your structured running will happen after you get home, plan the race for that.
Respect local conditions. Stray dogs, traffic and altitude are real hazards. Run in daylight and ask locals where it is sensible to go.
Treat niggles early. A "dodgy knee" on the road is harder to manage than at home.

More from that trip

The same 92 days produced some of my favourite travel memories, retold with today's context: cycling Bolivia's Death Road, volunteering at La Senda Verde, riding and hiking to Sierra Negra in the Galápagos and the bike-and-jungle route to Machu Picchu. They are collected on the adventure travel page.

Questions readers ask

How long did you train for your first marathon?

About three and a half months, by my own count in October 2009, with much of the early part improvised while travelling.

Can you train for a marathon while backpacking?

You can stay fit and build some endurance, but my experience was that the structured running that mattered happened once I was home.

July 2009 field diary: how training actually happened across Bolivia

Archive addition · 17 October 2026. The existing retrospective below this section is preserved; these newly rechecked primary notes clarify its earlier approximations.

My Marathon Training Begins entry, published 6 September 2009 gives a specific starting point: 1 July 2009, in Bolivia. I called it roughly 103 days before Chicago and left the preparation section blank because there was no plan. The photograph shows the battered footwear I was using. This supersedes the earlier “around the end of June” estimate in this retrospective without changing the original source.

The Bolivia running diary published the same day does record individual runs at altitude, contrary to the earlier summary here. I wrote of a slow 33-minute La Paz run on 7 July; three runs, each no longer than 40 minutes, during a week in Sucre; one 45-minute run in Potosí on 15 July; and a 35-minute morning run in Uyuni on 17 July.

In Potosí I described worn shoes and no pedometer. I chose an out-and-back line to avoid getting lost, then turned around when the route became dark. In Uyuni I wrote about running before the salt-flats departure and turning back at the sight of dogs. The old title says “rabid”, but my diary does not diagnose any animal; it records frightening encounters, not confirmed rabies.

These are recorded durations, not measured distances or a reconstructed weekly training load. They give the journey detail without proving an altitude benefit or explaining Chicago’s slowdown medically. The official net finish was 4:25:55; my race-day account admitted a too-fast start. The connection between those short, improvised runs and that race is autobiographical, not a controlled training comparison.

There is also a return-date distinction: the South America route ends on 3 August, while the training-begins note says I returned from further backpacking via Bremen on 17 August and counted 54 days at home. The earlier London timeline below is therefore an approximation, not a verified complete return itinerary.

Sources and notes

No individual training runs are reconstructed here; only what my contemporaneous posts recorded.

Written in October 2026 from my own contemporaneous records.