It's been done pretty successfully many, many, times. My fastest marathon came with nothing more than 15-16 mile runs in the year or so preceding it. There was a Brit called Chris Stewart who was third at New York in successive years in the late 70s each time in 2:13. He told me that in one of those years his longest run was 8 miles. There's this guy:
https://www.runnersworld.com/masters/day-after-day-with-yoshihisa-hosaka
.
Here's a two time Olympic marathon medallist:
Karel Lismont (Olympic brons (1972) and silver (1976) Marathon)
Lismont worked full time (8 hours a day) at the Ministry of Finance (Tax registration). Despite the fact that he was a marathon runner, he hardly did any real long runs, though he did run a lot. He’s training regime was very hard and featured lots of interval training. Most of his easy runs were done at 4’/km (roughly 6’30â€/mile)
The following training week is very accurate and he did these kind of training weeks year round – I know Lismont personally and he assured me this was how he trained.
Monday:
Morning: 6 miles
Noon: 6 miles + 10 x 100m
Evening: Interval training (see below for examples) (Total incl. warmup en cooldown: 5 miles)
Tuesday, Wednesday, Thursday and Friday: Same as Monday
Saturday:
AM: 9-10 miles
PM: 9-10 miles
Sunday:
easy day; not specified in the book
Week Total: +- 110 miles
Interval training examples:
- 10 – 12 x 200m
- 10 x 300m
- 6-8 x 600m
- 6 x 800m
Lismont preferred short but fast intervals. He never ran intervals longer than 800m.
He didn’t like running longer than 1 hour because he found it to be boring and by running no more than 6 miles at a time Monday to Friday, he felt he was recovering better while being able to combine the running more easily with working full time.
Side note – Lismont was in peak shape and ready to win at the Montreal olympics in 1976. But a few weeks before the race he went through a rough period personally (divorce) and developed problems with his stomach. He was already at racing weight, but lost about 10 pounds due to this condition and his blood values were terrible, despite this he still ended up with the bronze medal just 1’20†after gold medal Cierpinski and 27†after Frank Shorter.
Read more:
http://www.letsrun.com/forum/flat_read.php?board=1&id=8160693&thread=8160693#8160693#ixzz4rGyEbPDH
Buy your shoes from LetsRun and save 20% everday
http://www.letsrun.com/save
I used to know a guy called Jack Mahurin who was a low 2:20s marathoner in the 70s and also had a Ph.D in exercise physiology. Jack would never do any runs longer than an hour outside of races because he said in his research he'd found there is an enzyme released into your muscles that damages them His version of a long run was three 10 milers on Sundays all in fifty seven to sixty minutes.
I think overall volume is much more important. Stewart was doing 120-130 miles a week when he did that 2:13. Mahurin was a 100 or more a week guy. Lismont was doing about 110. Hosaka is well over 100. I was doing 80-90 but was pretty much beyond the time when I was really serious about marathons.
I'm not saying long runs don't matter and I think that to some extent, but only some, you can compensate for lower mileage with really long runs. But other things being equal, if I have to bet on who'll win in a marathon between a 70 a week guy doing long runs and a 100 a week guy not doing them my money is on the second guy.