How much does weight loss effect running speed (ie, seconds saved per mile/km per pound/kg lost)?

That calculation above translates to about 2.6 seconds per pound lost per mile.

I lost 14 pounds and dropped 1:27 off my 5K time from a couple of years ago. That translates to exactly 2 seconds per pound lost per mile.

I believe Runner's World published a number about 2 years ago which put the number at about 2 sec/mile/lb. This is, however a very crude number, and probably applies better to those who are out of shape better than those in shape. It must also be recognized that there is an optimum weight for each body and each distance. In general, the shorter your race, the more muscle mass can be tolerated. If your weight drops below the optimum, you'll likely be losing muscle mass and slowing down. (Don't blame an eating disorder on me, you've been warned.)

Here's an interesting thread about this...

http://www.letsrun.com/forum/flat_read.php?board=1&id=1701081&thread=1701040

Its not this simple but here are some interesting facts:

Dec 2006: weight 154 - marathon time 3:19

Feb 2007: weight 145 - marathon time 2:53

Many other factors but I believe weight/BMI is significant.

I think it greatly effects how efficient you can be and in the marathon that is huge.

chunk wrote:

Anyone?

I'll tell you, but first you have to do the truffle shuffle.

2 seconds per pound per mile.

Good ballpark figure. I actually find it more useful for feeling better about the times I run with 5 extra pounds of clothing on a cold day.

September 2006: I weighed about 180 lbs. Raced a 4 mile XC race and ran 24:31.

February 2007: I'm down to about 163 lbs. Recently did a threshold workout of 4 miles in 24:18 on the roads. Controlled effort, definitely had a lot left in the tank.

Obviously you can't compare courses, and I've put in a lot of training over the last 5 months, but 17 lbs. is nothing to sneeze at. What was once all-out is now workout pace.

Would it be a direct linear formula? Wouldn't there be a sort of sweet spot where you are at good running weight range at which a pound could mean more time, and then when you are way too heavy or even a little too light (yes, it can happen) your gain/loss would be less per pound?

Seriously now fellas! wrote:

Would it be a direct linear formula? Wouldn't there be a sort of sweet spot where you are at good running weight range at which a pound could mean more time, and then when you are way too heavy or even a little too light (yes, it can happen) your gain/loss would be less per pound?

Yes.

Also, if you are too light, you could be susceptible to poor recovery and thus poor training.

It would be more like a bell curve (maybe not exactly, but you get the idea I hope).

The 2 seconds per pound per mile number came from Tom Osler.

When I first heard this, I thought "He obviously doesn't know anything about numbers -- it can't be that simple". Then I found out he's got a Ph.D. in math.

for "perfect machines", from a physics point of view, dropping weight with no change in "power output"

new time = old time * ( new weight / old weight ) ^ 1/2

This all must premised on someone being overweight. If I am 6' tall and weigh 120 pounds, I am not going to be 40 seconds per mile faster if I drop 20 pounds and weigh 100 pounds.

Probably more important than how much you weigh is the process of weight loss. A calorie deficit sets off all kinds of warning systems in your body designed to respond to starvation situations. People who are dangerously underweight arent really all that unhealthy by virtue of weight alone. It's the fact that they're still maintaining calorie deficits and the body has nothing left to metabolise.

jsquire wrote:

The 2 seconds per pound per mile number came from Tom Osler.

When I first heard this, I thought "He obviously doesn't know anything about numbers -- it can't be that simple". Then I found out he's got a Ph.D. in math.

A PhD in math does not preclude one from not having common sense.

Would someone who runs 10:00 miles lose 2 secs/lb/mile as well as someone who runs 5:00 miles? Would a 200 pound person lose time as much as a 110 pound person who lost the same amount of weight?

It is NOT that simple. Besides, there are too many unknowns and unaccounted-fors.

Take myself as an example. At 155 pounds and with training very rigorously (70+ mpw) I could run about 4:20. At 190 pounds currently and averaging less than 10 mpw over the past several months I can still break 5:00.

Would losing those pounds (assuming I still ran minimal mileage) mean I could suddenly run a 3:50 mile? I think not...

And another consideration is what distance are we talking about? I ran 53.6 in the open 400 last spring at 185 pounds off of almost NO running at all in months. When I was running all the time and at 155 pounds, my best times were high 55 (relay). Explain that if it's just "that simple."

5 lbs lost = 1:00 of time off one's 10K time

extrapolate from that

2 sec per mile per lb is pretty good for elite runners. For slower runners it is more like 2.5 - 2.7 sec per lb per mile

Chunk + Truffle Shuffle!!!!

Maybe because it's 2 am, but that comment was so funny!

I'm going to be useless tomorrow.... Sigh...

but losing 20 pounds from 120 is almost certain to reduce your power output

for small losses in weight, it's a reasonable estimate

look at the different runners; sprinters are thick and muscular, distance runners are lean; it's all about the strength/mass ratio and endurance.

and it's also a factor of training - fast twitch and slow twitch. no surprise that you were slower in the 400 when you were doing a lot of distance running

the 2 sec/mile/lb is a rule-of-thumb, not etched in stone. and remember, it's the removal of un-needed weight

and it's an average over miles, i can't imagine it would be applicable to one's fastest mile