Was my negative split during my marathon too big?

slingrunner wrote:

Well without knowing your makeup of fast-twitch and slow-twitch fibers its hard to know what the best split is for you. However, like most people, you probably would benefit from a small positive split, since your LT pace most likely decreases the longer you run. It's very rare that a negative split is the winning strategy if you want to achieve your fastest time, although that conservative approach is also less likely to backfire spectacularly.

If you really take this seriously, you might want to consider running a tune up 10-15K before your marathon to get a better idea of where your fitness is.

Well, I certainly don't now that myself, but anecdotally I don't feel like I'm a fast-twitch dominant person and my Vo2 Max is certainly nothing great. I think it was tested at 157 years ago and may barely be over 160 now. I have to work on speed a lot and have gotten better at it, but I can't run sub-70 quarters all day, I underperform at 800s and I might barely be able to break 5 in the mile (never tried).

You have an interesting and contradictory opinion to pretty much everything I see. I don't know if I've ever seen any coach or publication recommend a positive split, but I know sometimes it does actually win. Some courses that have early downhill and late hilly sections often run at positive splits, but I've always heard a negative is the ideal method.

My read of your first post is that you ran a near-perfect race. You set out on pace for what you hoped you could run, and found you had more gas in the tank, so you ramped it up and finished strong.

You never really know the right pace to set out at, and guessing the other way is much worse. Knowing in hindsight you had 2:37 in you, maybe you should have set out at 1:18.5 for the first half. But you didn't know that beforehand. Had you actually been capable of 2:40 on the day, setting out at 1:18.5 likely could have had you finish at 2:45-55, not 2:37.

Great job, and quit overanalyzing this. Hindsight may be 20-20, but it's ultimately very nearly useless for helping you look ahead...

I agree that +/- 1 minute is ideal in principle. However in reality, my theory is that if you run maybe 30 seconds to 1 minute positive and are just starting to fade during the last mile or two (even with the benefit of rolling up people), then you know you left it all on the course. Not fading so much that you're actually bonking, but just starting to lose 5-10-15 seconds per mile close to the finish and that feels like you're maxed out.

As for Boston, it's a special case since the 1st half is so much easier than the 2nd. Something like a 2 minute positive split at Boston is about the same as even splitting a flat course like Chicago or Berlin.

I'm a fan of a big negative split. PR many years ago came with a 5 minute negative split.

1. Makes the race more enjoyable. You're not going to win. Why not enjoy blowing by people for the last 10k?

2. Removes doubt. I know I can hammer at a faster pace with fresher legs than dead ones. Staying in control for the first 17-20 is a huge confidence booster.

3. The math works out. Dropping 30+ seconds a mile for the last 10k works greatly in your favor mathematically to reach a goal pace.

4. Enables maximum effort. A true balls-to-the-wall effort can't be had on dead legs. Acceleration can't be done properly on fat metabolism ("the wall"). Preserving glycogen stores for an explosive last 10k sets you up to deplete everything.

I've run over a dozen marathons with just about every combination of split pairings, from 5 minute negative to 20 minute positive. I enjoyed the big negatives the most, and at the end of the day that's what this sport is about (again, not going to win any marathons).

I think a key that enabled me to have these sorts of races has been the depletion run. I don't take any fuel on long runs, just water. I pick up the pace at the end. Hurts like hell, but teaches your body how to handle fuel efficiently. Then when you fuel with gels in the marathon your body eats it up. Adding caffeine to the mix sure helps too.

After every PR race I've had in every distance from 800 to marathon, I've always felt that I can do better...wanted more. Many people I've spoken with have this same feeling. That's the hunger. Negative splits keep that going for me.

slingrunner wrote:

Well without knowing your makeup of fast-twitch and slow-twitch fibers its hard to know what the best split is for you.

I'm curious to know what you think the relationship is between a) fast-twitch versus slow-twitch and b) how to split a marathon.

To the OP:

I've run 7 marathons. In descending order from fastest to slowest:

PR: negative split by ~3:30 (Shamrock)

2: positive split by 21 seconds (Chicago)

3: negative split by 2:30 (Hartford)

4: positive split by 11 seconds (CIM)

5: positive split by 1 minute (Grandma's)

6: positive split by 10 minutes (ouch) (Philly)

7: positive split by 6 minutes. (Philly)

I find it interesting that my PR is the race with the biggest negative split. Shamrock is a double out-and-back pancake, so the negative split wasn't due to course profile.

Hartford was warm and humid, and so I intentionally went for a hard negative split. My theory is that in warmer or humid weather, a negative split is better - you stay conservative and manage your heat and hydration for a long time, and then hammer for the last 60-70 minutes . In great weather, even splits are the way to go.

Heimes57 wrote:

There is such thing as being more insecure and less insecure. What if I meant the insecure douches on another forum were LESS insecure vs MORE insecure? But then there are a variety of douches as well. Some are not insecure at all, just douches.

Good point. I don't really think you were looking for a forum of douches slightly less insecure than here, but I'll allow it. Carry on!

Insecure Douche wrote:

Heimes57 wrote:

I'm honestly looking for advice because I want to be the best runner I can be. Should I post there because there are less insecure douches?

Fewer. Fewer insecure douches, not less.

That depends. Maybe he was trying to suggest that the douches at reddit were less insecure than the douches here. Maybe the absolute quantity remains the same, just they are qualitatively different. Perhaps the OP would be kind enough to enlighten us as to his intention?

OP: did you mean the douches there were less insecure? Because I feel like most of us douches here on LRC do not suffer from a lack of confidence in our douchebaggery. Also, I think the bigger sign in your marathin that you went out too slow is that most of your negative split came in the last handful of miles. I do think you would have run faster had you gone out in about 1:18. Maybe you wouldnt have won (the winner may have been pushed by you in his second half), but I believe you would have ran at least 1 to 2 minutes faster. Good job on PRing, you should be proud of PRing, even if it turns out that it wasn't a perfectly executed race.

Shamrock negative split was likely due to tailwind on second half. Most years that is the case.

Well done. Nicely raced, and nice improvement over the years! I was very close to running Harrisburg as I had a cold for my goal marathon. But I ended up deciding it was too long of a drive and my taper was kind of screwy. Oh well it seems like a good course, next year.

In my last a marathon I ran a lot slower than you, but also neg split. I would agree with the sentiments that 2-3 minutes falls into the perfect zone of a good race. Anything more, like 4-5 minutes, and that's probably when you can be more conclusive in thinking that you underperformed in the first half.

Slight positive split is optimal on flattish course.