The following only applies if you are doing sufficient mileage. And by “sufficient mileage” I mean something north of 60 miles per week minimum. 80+ is better. If you tell me you can only do 40 miles per week, forget it. It is completely pointless to have a “goal” for a race distance that’s more than half your weekly mileage. Maybe you’ll make it, maybe you won’t but there is no reliable way to tell.
Here’s the “method”.
Make a reasonable guess at your goal pace.
Early in your training cycle, try 2 x 4 miles at that pace. Preferably on your own, on a relatively flat course. If it’s a hot humid day, maybe 2 x 3 miles.
If you come near to successfully hitting thatpace, then it’s probably not out of reason.
Later in your training buildup, try 1 x 8 miles at that pace. Make it happen or come close.
Very late in your training buildup (2 or 3 weeks before the marathon), Do a 15-19 mile run — all alone, no help from anyone-- with 10-12 at that pace. Make the pace happen, no matter how tough it is.
If you can do the last run, even if it wipes you out, you have a chance at holding that pace.
It would not hurt you to perhaps enter a 10 miler or half marathon during buildup in which you try to do a big block of miles at your goal pace. But that last bit (#5 above) should be all alone, in a non-race situation. That’s the litmus test.
On race day, go out easily. Don’t aim for a pace. Aim for 2 miles of “this is too slow!”. Give yourself 2 miles at whatever pace feels “too easy” and let the race pace slowly come to you. If it’s going to be your day, mile 3 will be right at the goal pace you want. Then you have 23 miles to slowly chip off any seconds you might have lost in miles 1 and 2. And if it is not your day, you will be very glad you didn’t insist on pace X:YZ at the start.
Here are some benchmark workouts I find fairly accurate. These should be done in normal training, not rested/tapered for.
6x1600 at 8% faster than MP, 400 recovery jog
6 miles at 4% faster than MP (no breaks)
20 miles with average pace 94-95% of MP
5 miles easy, 10 at MP, 5 easy
How do you make use of that if you’re trying to figure out your MP? do the first or second workout ‘all out’ but as consistently as possible and do the math to see where you should be. Use workouts 3 and/or 4 the following week(s) to validate your ability.
I do two benchmark tests with a few days rest in between:
Run 4x1600m with 90-seconds rest in between at max effort. The average of these 4 efforts is used to estimate 5K pace.
Run 800m max effort, 90-seconds walking recovery, then 1600m max effort. After those, run 4-miles easy, then repeat the test (this normalizes the results). Average the two 800m efforts, and then average the two 1600m efforts and determine the % difference between the two. This is your regression rate which is the pace drop off for each doubling of distance (the drop off from 5K pace to 10K pace, for example).
Use that regression rate to extrapolate all the way out to 40K to get your estimated marathon pace.
I realize it’s basic and prone to error, but it worked pretty well for me—I ran a few halfs and then used its pace estimator to get an MP for training. I then validated those numbers by doing some training runs (like a 10 mile) to see how things compared. However it calculates things–it assumes that you’ve got a fair amount of miles under your belt and you’ve got things figured out well enough that you don’t bonk.
I’m a pretty good runner (2:40) and what’s worked for me to determine marathon pace is doing (usually about 2 weeks before a race) a 16 mile hard run. Hard. It shouldn’t kill you. But haaarrrrrddddd. If you can hold whatever pace for 16 miles in the middle of a hard training week, then you should be able to hold that for a marathon when you are rested and tapered
How about racing a 10k? Get your Vdot and do some MP runs (5-10miles). You’ll know if it’s sustainable or not. None of this way over complicated tests.
Calculators can be decent for giving you a ballpark, but their value declines as you use shorter races and will vary greatly by the individual. So the longer the race the better.
I think people forget that these calculators are really not predictors of what you can run. They tell you times across all distances that are equivalent to the time for the distance you entered. So there are other important points you’d want to consider. What’s your current training like? How long have you been running? Are you more slow twitch or fast twitch?
For example, a talented high school kid can run 16 flat off of pretty low mileage (30-50 miles a week) but would be reduced to walking if they tried to run the equivalent 2:33 marathon.
I’m more slow twitch then fast twitch and I ran about 2.5% slower in my marathon compared to what the calculator predicted based off a half marathon two months before the race and a 5k 4 months before the race. I was also in better shape for the marathon and my training for the two months leading up to it was geared towards the marathon and mileage was higher.
I’d say your best bet is to run 26.2 miles… as quickly as possible.
Then rest a little bit, train some more and try again, preferably aiming to go a little bit faster. Repeat until your knees no longer work.
He’s kind of trolling you guys anyway, since he’s already gone 2:52 for an open mary.
For me it was a pessimistic look at a recent half time (double +20 minutes) and that was ballpark for my first time. Even that wasn’t pessimistic enough tho lol
Vdot table from a 5k was about 40 minutes optimistic of where I ended up. I don’t do a lot of long runs, and had poor preparation/disclaimer