12

What's a good strategy for pacing oneself on hills during a marathon (versus flats or downhills)? (Read 773 times)

    I'm running the ING Georgia Marathon in about 11 days. There are about 3-3.5 miles worth of 'significant' hills where the elevation gain is something like 100 to 200 feet per mile. Also, during the last 4 miles, the elevation goes up about 200 feet - no big hills here, but a steady slow uphill. So if I'm shooting for an average pace per mile for the race, what's the best strategy? Should I try to keep the same pace even on the hills? Or instead, should I try to keep an even effort throughout the race, and expect to run the hills a little more slowly, and the flats and downhills a little faster? Or something in between - where I slow up on the hills, but still put in a little more effort than on the flats or downhills? Or am I thinking too much? Shocked
    Trent


    Good Bad & The Monkey

      Have you trained on hills? MTA: looked at your log. As long as the weather cooperates and you can run a smart race, you are gonna ROCK!


      Kings Canyon NP 07'

        Find three or four people that are in your pace set and stick to them............then pick three or four people that look weak and pass them.

        left-right-left-right-repeat

          Although it's easier said than done, you want to keep the effort as even as possible. And, as I'm sure you realize, there is nothing that will better prepare you than practicing on the type of terrain you will be racing on. For this next one it's a little late for that if you haven't already been doing it, but you could squeeze one in if you do it soon. I'm not so hot on the idea of picking out other runners and trying to stick with them unless you are very familiar with their running styles and know that they are smart racers. I think it's better to run your own race during the early miles and rein in the competitive juices until later in the race.
          Age 60 plus best times: 5k 19:00, 10k 38:35, 10m 1:05:30, HM 1:24:09, 30k 2:04:33
            Trying to maintain a constant pace on hills is a mistake. The energy that you expend to maintain pace going uphill is not offset by that saved or returned on the downhill. Plus, braking to hold the pace down on the downhill is hard on the quads. Try to maintain an even effort on hills, which will mean slower going uphill and faster going downhill than the pace on flat terrain. Think of it as driving on hills with a constant accelerator position, as opposed to pushing on the accelerate on the uphill and easing off on the downhill to maitain constant speed. For detailed comments on hill running, whether in training or in a race, see http://mysite.verizon.net/jim2wr/id18.html.
              I agree with both Jims. You have to maintain an even effort throughout the race. You will have to slow down on the uphill sections and then run the downhills faster. If you have been training on hills and working on your downhill technique, you should able to fly downhill without expending additional energy. I think it helps if you can find a few people planning to run the same pace as youself so you can stick together but in the end you have to run your own race. Good luck! My Blog http://breakhearttrailrunning.blogspot.com/
              Only the lead dog has a different view. My Blog http://breakhearttrailrunning.blogspot.com/
                Go by effort. Whether it's hilly or not.

                Runners run


                #2867

                  As stated, effort is a good gauge. I'm also running ING Georgia, and I'll probably pick up the pace on the hills and run closer to even times. I'm sure I'll slow down a bit, but not so much as normal. Then again, this is a warm up race for my goal marathon on May 4, so I'm treating it as a tempo run.

                  Run to Win
                  25 Marathons, 17 Ultras, 16 States (Full List)

                    As long as the weather cooperates and you can run a smart race, you are gonna ROCK!
                    I hope you're right Trent! My definition of ROCK would be if I finish in 3:30 or less, since that will qualify me for Boston (I'll be 45 next year). Various 'race time predictors' predict I can run something like a 3:17 or 3:18, but I realize those things are only accurate if your running a flat fast course with the wind at your back when the temperature is 50F the entire race, and everything else falls your way and you run a perfect race. Still, I think a 3:30 might be doable, but with the Atlanta hills to negotiate, it ain't gonna be easy, that's for sure. By the way, thanks everyone for the great info on running hills - it has definitely been helpful.
                      Various 'race time predictors' predict I can run something like a 3:17 or 3:18, but I realize those things are only accurate if your running a flat fast course with the wind at your back when the temperature is 50F the entire race, and everything else falls your way and you run a perfect race.
                      ....and if you have sufficient training base to support the predicted time. I don't know what your training has looked like, but race time predictors are most accurate for those running 70+ miles/week and less so for those running lower mileage. See http://mysite.verizon.net/jim2wr/id70.html. If a 3:30:59 BQ is your primary goal, then I would ignore the 3:17-3:18 prediction and base your race plan on 3:28, at least for the first 18-20 miles. If you can push the pace in the last 10k to bring it home faster, that will be icing on the cake. But going for 3:20 or faster could prove disastrous.
                      JakeKnight


                        I don't know what your training has looked like, but race time predictors are most accurate for those running 70+ miles/week and less so for those running lower mileage. See http://mysite.verizon.net/jim2wr/id70.html.
                        Hey Jim: I found that article really interesting and helpful and I intend to use it in coming weeks to pick a realistic marathon goal for next month. I actually have a goal in mind but may have to adjust it (there's a small wager in play) so I really don't want to over extend and blow the race. I'll run a 10k time trial soon to get a more accurate idea of where I am. But the one thing I didn't understand was what exactly to base your weekly mileage on. Admittedly, I skimmed parts of the article, so I may have missed it. But what exactly does "mileage" mean? Average miles over the course of the entire preceding year? Average training miles during some specific pre-race period? Or is it peak weekly mileage prior to the race? Which mileage do I use to pick the ratio you discuss? MTA: Never mind. Just found this:
                        "Mileage" is the average weekly mileage I ran during a 16 week period preceding the marathons, which generally was in the 40-47 mile range.
                        So average the preceding four months or so?

                        E-mail: eric.fuller.mail@gmail.com
                        -----------------------------

                          Hey Jim: I found that article really interesting and helpful and I intend to use it in coming weeks to pick a realistic marathon goal for next month. I actually have a goal in mind but may have to adjust it (there's a small wager in play) so I really don't want to over extend and blow the race. I'll run a 10k time trial soon to get a more accurate idea of where I am. But the one thing I didn't understand was what exactly to base your weekly mileage on. Admittedly, I skimmed parts of the article, so I may have missed it. But what exactly does "mileage" mean? Average miles over the course of the entire preceding year? Average training miles during some specific pre-race period? Or is it peak weekly mileage prior to the race? Which mileage do I use to pick the ratio you discuss? MTA: Never mind. Just found this: So average the preceding four months or so?
                          It's really difficult to be precise on something like this, Jake, because we are all so different. As I said in the essay, it's best to learn what works best for you over a period of time. But, as a rule of thumb while working on your individualized guidelines, I think that, if one's training mileage varies seasonally during the year as mine did (spring 10k and fall marathon seasons), average mileage over 12-16 weeks preceding the marathon, excluding the taper weeks, is a reasonable guide. The guy who conducted the 10k vs. marathon performance survey that the guidelines in my essay reflect asked respondees to report mileage during a 16-week marathon training program, not including taper weeks. Of course, if someone trains at about the same mileage level year round, then a weekly average of annual mileage should be fine. Remember, these predictors aren't precise, partly because of the many variables that come into play. You might be interested in an early (1973) attempt to correlate marathon performance to several variables, including total mileage during the 8-weeks immediately preceding the marathon....http://mysite.verizon.net/jim2wr/id53.html. At least, I found it to be interesting. I heard from the original author a couple of years ago. He said that he liked my analysis of his article. Smile
                            Jim, your article on marathon pacing was very interesting. It definitely makes sense to me to include the extra variable of ‘training intensity’ to ‘correct’ the race time predictor times. I've only run one marathon (last Nov), and I started out way too fast and as a result I hit the wall after about 19 miles. So my final time, I’m pretty certain, was much slower than it would've been if I'd run a smarter, even paced, race. Even still, when I calculated my ratio it was 4.85 (I used a 15k time that I'd run about 3 weeks before the marathon, and converted it to an equivalent 10k time as suggested in your article to calculate the ratio). My mileage per week wasn’t high either (about 35 mpw), since my training was disrupted for a period of about 2-3 weeks in the 16 weeks prior to the race. It was more like 50 mpw the 8 weeks before the taper though. Anyway, although this is an experiment of 1, it appears that my ‘personal ratio’, like yours, might turn out to be something closer to what the race time predictors use (i.e., 4.6 to 4.7) than would be suggested by Brian’s table. I guess my upcoming race will give me my second data point! By the way, it seems to me that two other key factors for accurately predicting a marathon time are 1) course difficulty and 2) weather (mostly temperature and wind). Seems like there should be an algorithm that can take the elevation profile of a course and estimate a correction factor! Wouldn’t it be cool if all (or some) of the huge amounts of data in this website’s database could be used to develop a new calculator….someday maybe….
                              Jim, your article on marathon pacing was very interesting. It definitely makes sense to me to include the extra variable of ‘training intensity’ to ‘correct’ the race time predictor times. I've only run one marathon (last Nov), and I started out way too fast and as a result I hit the wall after about 19 miles. So my final time, I’m pretty certain, was much slower than it would've been if I'd run a smarter, even paced, race. Even still, when I calculated my ratio it was 4.85 (I used a 15k time that I'd run about 3 weeks before the marathon, and converted it to an equivalent 10k time as suggested in your article to calculate the ratio). My mileage per week wasn’t high either (about 35 mpw), since my training was disrupted for a period of about 2-3 weeks in the 16 weeks prior to the race. It was more like 50 mpw the 8 weeks before the taper though. Anyway, although this is an experiment of 1, it appears that my ‘personal ratio’, like yours, might turn out to be something closer to what the race time predictors use (i.e., 4.6 to 4.7) than would be suggested by Brian’s table. I guess my upcoming race will give me my second data point!
                              You are just another example that many (most?) individuals are not exactly "average". All such "predictors", whether calculators such as McMillan's or tables such as Brian's, reflect average performances. But individuals vary all over the place. I think that's why it is best to just use them while developing personalized factors.
                              By the way, it seems to me that two other key factors for accurately predicting a marathon time are 1) course difficulty and 2) weather (mostly temperature and wind). Seems like there should be an algorithm that can take the elevation profile of a course and estimate a correction factor! Wouldn’t it be cool if all (or some) of the huge amounts of data in this website’s database could be used to develop a new calculator….someday maybe….
                              I could not agree more. Environmental variables can have a major impact on race performances....and it is magnified in the marathon distance. The Runworks calculator is the only one that I know of that provides a means to estimate those impacts.
                              Hannibal Granite


                                Remember, these predictors aren't precise, partly because of the many variables that come into play. You might be interested in an early (1973) attempt to correlate marathon performance to several variables, including total mileage during the 8-weeks immediately preceding the marathon....http://mysite.verizon.net/jim2wr/id53.html. At least, I found it to be interesting. I heard from the original author a couple of years ago. He said that he liked my analysis of his article. Smile
                                Thanks for posting the link to that study and the formulas. Just for giggles I looked back through my logs for each of the four marathons I have done and compared to each of the 8 formulas in that 1973 study and I got some interesting (to me) results. Everything rounded to the nearest minute Marathon #1 actual finishing time 3:15 fastest prediction 2:59 (formulas #3&4) slowest prediction 3:15 (formula#6) correct prediction - formula #6 Pro: One of the formulas did predict my time Con: had this not been my first marathon all the predictions would've been considerably too fast Marathon #2 actual finishing time 2:49 fastest prediction 2:45 (formula #3) slowest prediction 3:02 (formula #6) correct prediction - formulas #1 & #4 Pro: With the exception of #6 all predictions within 10 minutes of each other Con: Would've all been to slow had I paced myself a little better Marathon #3 actual finishing time 3:15 fastest prediction 2:42 (formula #3) slowest prediction 3:01 (formula #6) correct prediction - none all predictions too fast Pro: With the exception of #6 all predictions within 12 minutes of each other Con: Didn't take into account the pulled hamstring I had 2 weeks before race, of course I wouldn't expect any formula to do this Marathon #4 actual finishing time 3:02 fastest prediction 3:10 (formulas #1,#2,#3) slowest prediction 3:28 (formula #8) correct prediction - none all predictions too slow Pro: If I had these formulas before the race I would've felt good about beating all of them Smile Con: I didn't do any 20+ mile runs in the last 8 weeks, but did several of them in the 8 weeks prior to that, which weren't accounted for. Also, I cross-trained more for this race than in the previous races which were also not accounted for. So what does this all mean? Probably nothing, but it gives me another tool to play with and helped me kill an afternoon during Spring Break

                                "You NEED to do this" - Shara

                                12