Pfitzinger/Douglas suggest positive splitting the marathon.

I ran a 1:18- 1:31. It was painful. If I were a woman I think I'd say it was more painful than childbirth.

Bad idea.

So far this thread has developed as usual on LR: no intelligent replies.

I vote for going out exactly on target pace. If you are in better shape than expected, do a negative split. If not, you haven't gone out way too fast. A big positive split feels like death even if you run a decent time.

My PR came from running almost exact splits. I was tired at the end but not completely dead.

The times where I've positive split I've felt dead at the end.

I've negative split once and felt I left some minutes out on the course.

My PR came with a 30 second negative split in Chicago and my second fastest time came with dead even splits on a hillier course. The only downside to negative splitting is that you will finish thinking that you may have left something on the table. Overall, shoot for even splits. I think planning to positive split is not a great idea.

Regardless of what you do, if you feel like slowing down (due to fatigue at least), first try speeding up very slightly for 5 minutes. You'll usually be surprised.

Negative split. It is by far the best way to go. Just the mental boost you receive from passing people over the last 8-10M is significant.

The best thing about the negative split is that what is the worst thing that happens? You get to mile 20 and feel too good?

In that case you pick it up and run faster.

The positive split is a hard, tough way to go. Maybe you feel like this is the way to squeeze ever last ounce out of your body, but honestly I think you can do that by running very hard over the last 10K.

well,, wrote:

So far this thread has developed as usual on LR: no intelligent replies.

You sweat out 4 lbs of water, your pace will be a few seconds faster. Was that taken into account?

McMillan calculator says if you are in 239 shape, but go out in 118 (557/mile), you will BONK (DONE) FINISHED at 20 miles.

Remind me of Scott Douglas's marathon PR again?

It's well known that banking time is a winning strategy.

You guys do realize that 90% of marathoners finish with a positive split ?

I’m a low mileage guy so the last 10 k always hurt big time.

And even though I fade just a little bit (+4 min pos split), I pass way more people than there are people passing me in the final miles.

Charlesvdw wrote:

You guys do realize that 90% of marathoners finish with a positive split ?

Yes. Why would you want to be like the 90%?

My best two marathons were a negative split any where from .90-.120. I'm not talking about jogging through 13M then running, but just backing off that slight bit for the first half can make the second feel so much better.

I ran 1:36/1:40 in the only marathon I ran, so right on the positive split they suggest. I would suggest the course accounted for at least half of the difference as there were two significant uphills in the last five miles that cost me quite a bit of time. I followed Pfitzinger/Douglas for my race prep.

I know that is horribly slow by Letsrun.com times. I only make ~100,000 as well.

I've run 29 marathons; my first marathon, my first sub3, and my two fastest times were all negative splits. It's just awesome to stay strong right to the finish. Wish I could do it every time. Go out easy!

I think the real take away from that section is to run even splits(which he says is ideal). A 2% difference really isn’t much.

If your splits are within 2%(positive or negative), you ran a great race.

If you hugely negative split, you ran too slow at the beginning, if you hugely positively split, you ran too fast.

There are too many race day variables to justify religiously following positive or negative splits.

I just looked at my splits again from my first and only marathon. For some reason I thought I really blew the second half and remembered a large positive split. Those last five miles were soooo painful. Inner thighs were not happy with me.

Anyway, the second half was only 1:35 slower than the first half. Not too bad. Had to use the porta at mile 20, so I guess I can blame it on that. :)

Training for my second marathon. I guess I did okay with the first one and will use the same strategy for this one coming up.

I've run a very slight positive split three times

- the first was 3:09:21 split as 1:34:30/1:34:51;

- the second was 3:11:11 split as 1:35:30/1:35:41

-the second was 2:57:42 split as 1:28:48/1:28:56

Each race I felt good about - I thought I had run pretty much what I was capable of that day. My plan for each race was a negative split. I started slow and then built. In the second and third races, I had a slight, unintended fade in the late miles. (the first race - the 3:09 - was a negative split because I had to pull off to the side and do a clothing readjustment in the second half).

I never intended to run a slight positive split in any of these - I went for the negative, and then minor things happened and I got a slight positive. And that's really my point - I think a slight positive split is a very good way to execute a marathon, but you can't accomplish it by targeting a slight positive split.

Plan on a slight positive split, and I'm fairly certain you'll end up with a big positive split. But...go for a negative split, and if you have a good race, you'll have something between a negative split and a slight positive split.

I think this is basically right -- it's not easy to plan the pace that will have you positive split by 30s instead of 5 minutes. Ideally you should have a good gauge on your effort, not the time, and then if you positive split by a bit that's fine. Intentionally choosing a time you can't hold seems like a recipe for a big blow-up. I've always found it tougher than expected to pick up the pace the last 6 miles -- at that point you're low on fuel and your muscles are beat up. I thought I made a big move the last 6 at Boston last year, and passed lots of people, but my pace didn't change at all.