Wednesday, October 04, 2006


Today’s discussion focuses on the marathon and pacing. I love this time of year; cooler temps, fall colors, marathons taking place every weekend, etc. Call me a sadist, but two of my favorite “hobbies” this time of year include 1) browsing marathon results and 2) reading race reports.

Yes, I know other people do this stuff too. However, when I look at race results I’m usually comparing first half and second half splits to see how badly people blew up. And when I read race reports I especially like to find sentences like “I felt really good at mile 2 so I picked up the pace…” Then I scroll down to about mile 16 in the report and look for the gory details.

I haven’t officially kept track of results but the same things appear in every marathon, every year, no matter where you are in the pack – people slow down. Looking at the first 500 finishers of TCM and I’d guess about 7% ran even or negative splits. A positive 5-minute split seems normal.

Since I’m a math geek, I thought I’d break that down into pace. Say you want to run 7:00 pace and finish in 3:04. Assuming you go out on-pace (which is probably a bad assumption) you’d go through the half in 1:32. The 5-minute positive split would give you a 1:37 second half for a time of 3:09. That means your pace dropped from 7:00 to 7:24 pace during the second half. Assuming you held 7:00 pace all the way to mile 20 means you dropped from 7:00 pace to 7:54 pace for the last 10k.

Those seem like rather large increases when you look at it that way. But like I said, this 5-minute positive split is “normal”. I know I’ve done it. I’m just curious why it happens. Are people doing it on purpose? Falling apart? Not realistic in their goal pace? Not adjusting their goal to the conditions?

I don’t have the answers, but I have some ideas. I’m curious to hear other's opinions.

As for my training, I haven’t quite started my taper yet. In the past I think I’ve tended to over-taper for my races. I’d get to 3 weeks out and just cut my mileage in half all the sudden. This time around I’m going to see if I can manage something closer to a 70, 50, 30 drop in weekly mileage.

I jumped on the treadmill last night for an easy 5 miles. I had a dentist appointment this morning, so I’ll have to put off my run until tonight.

I was going to give an update on my other projects, but this is already getting to long. I close with a “funny” article on Marla Runyan at TCM. I particularly enjoyed the last sentence; “I can't complain.” Based on the article, I beg to differ.

Quote of the day;

“Show me a real runner and I will show you someone who has, at least on occasion, become too wrapped up in the sport.” – Mark Will-Weber, The Quotable Runner


pjm said...

I'm actually going to try to do some wholesale data analysis on 5k splits from Boston. I'm going to start with DNFs and I think then compare final results with seed times, and see what comes out in terms of "race profile." Should be interesting. I'll be able to tell more in a month or two.

UMaine Cooperative Extension said...

Nice article on Marla Runyan. Being blind in one eye, I can sort of relate.

I am not sure what is worse, living through the gory details of a poorly run marathon, or posting them for sadists to read.

Wayne said...

You brought up a great topic. I personally think it's because runners think they can sustain the pace for the entire marathon. The early miles feel so easy that we feel the need to pick up the pace. Marathon pace should not come by surprise. I've learned the hard way. I think i'm mastering the art of negative splits now.

Anonymous said...

As you may recall from when we used to work together, I did a study comparing 10k times from Get in Gear to Grandma's Marathon times matching people by name, and found how much "negative split" was optimal. As I recall (data is gone with quitting the job) it was about a 2% positive split that gave the smallest multiple of 10k to marathon time.

I'd guess the slowdown is caused by an increased fat to glycogen ratio, as the glycogen gets depleted.

Lawrence said...

I liked the article on Runyan. Your quote of the day is spot on!

Triseverance said...

Interesting, it is quite a decision especially for us first time marathoners.
I feel like I have nailed the training, hit all of the goal paces, gotten confirmation in MP runs and a tune up race.
Still I am unsure if I am setting my goal to high for a distance that is that far and foreign to me. One side of me says if you don't try you will never know, bonk or no bonk. The other side fears the bonk. hmmmmmm
I was hoping to run even splits. I want to be realistic but at the same time I don't want to sacrafice my goals.

Anonymous said...

Just got round to looking at the actual results from Twin Cities, sorting by time at half-way. That's some carnage there, with a lot of people well beyond the 4-8 minute positive split that one can reasonably attribute to going out 1-2 minutes too quickly. After a while the 1:25/1:38 type splits just seemed normal. That leads me to believe that the combination of direct sun, 60 + temps, and a last six mile headwind got to a lot of people.

Steph said...

I'm not sure what is typically used as a strategy, but it might depend on how one breaks the marathon down. I tend to use: 1-10 mi, 11-16 mi, 17-20 mi, last 10K. I adjust my pace in the race (and in training) to match up with those mile markers.

I have experimented with negative splits in two marathons, and it's a fine art to figure out how to run as even splits as possible and still run your goal time. Experiment #1, National Marathon, 1st half in 2:11, second half in 1:49 (with a 48:05 last 10K). Boston 2005 was a bit closer, 1st half in 1:58:58 and 2nd half in 1:53:42. Not a fast time, but it's really hard to figure out where that sweet spot is between running a fast 2nd half and blowing up.

Hopefully Marine Corps in a few weeks will be a negative split AND a Boston qualifying result, putting everything together.

mouse said...

I've finished two marathons, and in both I negative split (the second by a good 7 minutes, breaking my then-current half PR). but... I don't ever feel that i've really trained to "race" a marathon. those two times I went out at a pace that felt comfortable, and then found I had energy left at the end so I picked it up, so I kind of don't feel like the two I finished really "count".

the real point of this comment is to let you know that I laughed literally a good two minutes at this:

"And when I read race reports I especially like to find sentences like “I felt really good at mile 2 so I picked up the pace…” Then I scroll down to about mile 16 in the report and look for the gory details."

too, too funny.

Anonymous said...

From the "I've got wayy too much time on my hands" department, here's an analysis of the top 200 finishers @ the Lakefront Marathon on Sunday (about 11% of the total finishers)

-The average positive split for this group was 6:21, or 6.93%
-14% of these finishers ran even or negative splits (28 total)
-36% ran positive splits of 10 minutes or greater
-There was no real discernable difference between the 1st 50 finishers (6.3% positive split average) or any of the next 3 groups of 50 (7.2%, 6.8% and 7.4% respectively)
-The best negative split race was a 1:41/1:32
-The worst negative split race was a 1:23/2:01
-Kudos to the 27 people in this group who ran from 1 minute positive to a 1 minute negative split.

Having run Lakefront twice in the last 4 years, I'm aware that it is a course that can lull you into a fast 1st half. Having run a six minute positive split the first time and a two minute positive split the 2nd, I am very impressed with those who can run even pace throughout.

Chad said...

Parker, the geeks (me, Eric, Evan, Peter, Jack, etc.) will be waiting for a report.

Marc, I'd say "living through" is worse.

Wayne, nail = head when you said "Marathon pace should not come by surprise."

Eric, you mean "work" together. Yeah, I remember that study. Too bad we didn't have enough funding.

Bob, yeah I don't envy you first timers. It sounds like you are on the right track. As I've heard before, "no one" has ever finished a marathon and said "Damn, I wish I'd gone out faster."

Susan, I plead the 5th.

Evan, "After a while the 1:25/1:38 type splits just seemed normal." You're right, but it's still scary.

Steph, your pacing strategy is interesting. And while I like your two negative split results, they're a long way from your 3:27 PR.

Mouse, it's funny because it's true.

Peter, now you're scaring me.

Jack, stay tuned for Parker's "wholesale data analysis."

Arcane said...

I've only run three marathons and the closest I've come to even splits is a 3 minute positive split, but I finished that feeling great. The other two I crashed and burned with a 15 minute positive split in my first and a 10 minute positive split in my most recent (albeit a PR). I don't have much experience but I think even splits are the way to go.

qcmier said...

Great post. Here's some numbers to think about. I've run two marathons.

1st 1:34/1:46

2nd 2:18/2:05

On the second one, believe me if I felt I could have run my first split faster I would have. But everything gets skew because that was after a little swim and a bike ride.

We'll see third time around.

Anonymous said...


Your suggestion that many runners positive split because they are too optimistic about their ability doesn't really hold when you start looking at the marathon splits for elite and local champion runners. In the original research I have done, it is even rare for racers in these groups to negative split.

My goal is to evenly split at best and nearly collapse at the end of the race because I left it all on the course.

Good luck to everyone racing this fall.

ShoreTurtle said...

Numbers can be facinating. I too often look at race results and look for the gory details.

I'm running my first marathon in a couple of weeks. I'm hoping not to become a crash and burn statistic.