Should I go for sub 3 or sub 2:55 in a marathon

Are you also a time traveller from the past? In 2025 they're looking for someone to go sub 2 or 1h55. You may still be a winner at the 1908 Olympics if you go sub 2h55 to beat me. But elsewhere and elsewhen buddy, this timeline belongs to yours truly.

He didn’t ask what to tell everyone he asked what he should actually do, yall get too consumed with what is broadcast to the world about your life. Newsflash: the world doesn’t care!

1:22 half is in line with 2:50 marathon. Go for BQ! Good luck!

If you need any coaching help, lmk

I'd probably go for 2:55, I ran 1:24 HM last october and then a 2:53 at the same event this october. And that was coming off being sick.. all the training and tune ups were pointing towards 2:50 being really reasonable.

And that was with a really mild bump in mileage, mostly just continued consistency. I was running around 55mpw all of last year and this year, and did pfitz 12/70, which I think averages around 60-65mpw.

General consensus seems to be to go for 2:54, especially if I gain any fitness between now and race day

It feels like you know what you should do and are just seeking our confirmation.

If it were me and I felt like 2:55 was going to be possible, that's how I would train. You're going to know 4-6 weeks out if it's realistic given any complications that came up through training. At that time, you can adjust your target time/pace accordingly.

What are your times on other distances?
Indeed, many people can run sub 2:55 with a 1:22 half and some can break 2:50, but there is also people that is stronger at shorter distances that will have trouble breaking 3 out of 1:22 half marathon fitness.
In my example of one, I am a miler and run a 1:23 half and then tried to break 3 in my first marathon, just to bonk badly and split 1:29-1:55 (ouch!). After 1 more year of training, I run 2:50-low, but by then, I was breaking 1:18 in the half. My advice would be to look at your shorter distances and try to figure out which type of runner you are and plan accordingly. If you can run a 16-minute 5K, you may find it difficult to break 3 hours, but if your best 5K time is 18:30 minutes, then you may be able to pull out a 2:50.

All super in line with each other. Seconds under 18 for 5k, around 37:30 10k, 5:13 mile

You've got some time to work with. Start where you are, then re-evaluate in a few months. Re-evaluate again when you're 3-4 weeks out.

DON'T re-evaluate at mile 5-10 of your race when you're coming off a taper and feeling fantastic and sub-6:30 pace just feels super natural. That's the mistake people make.

Then 2:55 is within the realm of possibilities.

Agreed.

No reason not to start off with a sub-2:55 goal, especially with short distance PBs that line up. Take your cues from running marathon pace on tired legs versus being more fresh in your taper. Be honest with yourself pre-race & give the weather/course some thought if you need to be a touch more conservative.

When I 1st went under 2:55, I had run a summer 18:40 5k. I ran a somewhat surprising 1:20-mid half in October on an honest course on a cool weather day. Then I ran a horribly paced 2:53. I was college-aged & went out at 6:15ish pace, feeling great on tapered legs/not knowing how to run the distance. Definitely had sub-2:50 in me. Running 60-70+mpw or so at that time. Ran sub-5 & 17-mid the following spring. 2:46 in the fall, again with sub-par pacing. This was all at a point where I was making some rapid progression, with increased mileage. Just be consistent. If you can hang out in the 60-75mpw range then there's no reason you can't go after a strong marathon time.

How old are you?

27

Train for 2:55.

Your total mileage can be the same whether you are "training for 2:55" or "training for 3:00"; the main difference is of course the pace of your race-specific workouts.

There is no harm in running a bit faster than marathon pace for earlier workouts in the build, like 10xMile, 6x2 miles, etc. With 1:20-1:22 fitness none of that will be a problem.

As you get into tougher more specific marathon workouts like 20-22-mile runs containing 3x5 miles or ending in a 10-miler at 6:40/mile for a 2:55 goal, it should become clear if this is feasible or not.

If it's too much of a reach this build, still try to train at that pace or at least 6:45-6:50 per mile, and you will be well-prepared for the 6:52/mile needed for sub-3:00.

If you can run 1:22, then you can run 2:44 just double it

Agree that you don't need to decide now--I often found my marathon pace training runs were the best feedback I could get on how fast I could run a marathon. If I could struggle through but was beat the next day, then the pace was too fast for a whole marathon.

Anyway, I strongly lean toward 3:00. I think I had run 1:18 before I finally broke 3 hours, after 4 tries. There's a big difference between a half and a full, and not everyone makes the jump well.

True that there’s a big difference. True that not everyone “makes the jump” well. Hence the questions directed at OP to see whether he might be more or less likely for his half time to translate up, and hence the suggestions about managing training to give him the best shot at doing so (under present circumstances, that is).

Here, on Reddit, and on Strava, I sometimes see people report running marathon times slower than indicated by their half results. A certain percentage look as though they may possibly be more suited to shorter distances than longer ones. But it looks like a higher percentage are people who, upon providing additional information, clearly had certain missteps either in training generally, in the last two weeks, or on race day that were the culprit rather than it being an issue of just having innate difficulty making the jump in distance.

And I wouldn’t doubt in the least if for some people (probably not a very high percentage, the whole phenomenon of the race being different just gets in their head and becomes a self-fulfilling prophecy.

Updating now. I ran a 37:41 10k, which "converts" to 2:53 mid roughly. I have my doubts about my ability to really match the conversions, however, not all courses are equal, so that should be considered. I'll be racing Mesa.

This course had about 115 feet of climbing, with 2 back to back steep up and down, plus one more steep up and down. It also had very janky turns, with 2 straight up 180 degree turns, 6 90 degree turns, and maybe 4 kind of tight curves. A 10k equivalent of Mesa would have slightly less uphill, and said uphill would be less steep, a lot more downhill, and some weird turns but not as many, so I'm sure I would have been significantly faster on a course like that.

With all that said, do I go for 2:55 or 3 in about 6 weeks?