My training for a 3-hour marathon
As most of you know, I recently took a break from triathlon to train up from my first marathon – the Chicago Marathon on Oct 13, 2013. The training went really well and the race met all of my expectations for my finishing time (2:57:47) and more importantly my overall race experience. The race report is here: http://members.endurancenation.us/F...fault.aspx.
A bunch of folks have asked me what training plan I used building up to the Chicago Marathon so I thought it'd be worth documenting in some detail what I did and why and how I felt about it all. Documenting this is not meant to imply that my approach is the best way to prepare for a marathon or even that I'd advocate for using this approach. At the end I've made an honest evaluation of how I think this approach worked for me and there are certainly positive and negative aspects.
INTRO AND BACKGROUND
I am currently a couple of weeks away from my 38th birthday, 140lb and 5'5". My background is not at all athletic (I was on the debating team in high school) and I've only been "off the couch" since 2010 when I set a goal of doing an Olympic-distance triathlon the year after starting to make changes in my diet and begin a substantial (70lb) weight loss journey. The following year I joined EN and stepped up to the half-iron distance. I have always followed the EN plans in my training so the "EN style of training" is pretty much the only way I've ever trained.
I have never really done dedicated run training, it has always been running in the context of triathlon. That said, I do a handful of shorter (usually 8-10k) running races each year, mostly in the spring to test my VDOT or in the fall to give me some motivation. Now 37 years old, I have been lucky to be able to set PBs in many races since my starting point was so low, the EN triathlon plans have been effective at improving my fitness and EN execution has allowed me to optimize my performance. Also, as mentioned above, I have lost a lot of weight since 2009 although I've been stable in the 137-142lb range since 2011 so the more recent performance increases have been fitness-driven.
My wife was expecting our 3rd daughter in June of this year and I wanted to free up some time. As most people here in EN know, I'm an "all in" sort of person so half-assing my triathlon training wasn't an option. Cutting out biking and swimming and setting a goal around running – a MUCH less time-intensive sport – seemed like a good option. Since I had never run a marathon before I thought running the Chicago Marathon would be a good goal to set. My season plan was to do two early-season triathlons – the American TTT and 70.3 Kansas – then take a week off before dropping into week #2 of an 18 week marathon build.
I expected (and still expect) to be a "one and done" marathoner, so my goal was to run to my capability and be able to look back at the experience and know that I didn't leave much on the table. In that vein, I took a very methodical approach to preparation:
- I did a ton of research into selecting and constructing my training plan (lots of detail on that below), and included several aspects beyond running workouts, for example strengthening and massage
- My training was super-consistent with virtually no missed workouts over a 17 week training plan (obviously with running this is much easier than in the other triathlon sports as long as you don't get injured)
- I spent a lot of time figuring out what pace I should target in the race, and solicited advice on that and a number of other issues over the course of my training from you guys as well as non-EN sources
- For the first time ever I actually wrote down on paper my very detailed race plan (http://members.endurancenation.us/F...fault.aspx)
- Knowing that race-day conditions play such a pivotal role in performance and that lack of experience could lead to execution failure, I even signed up for a "backup marathon" in November in case I boogered the race and didn't perform consistent with how my training should have set me up
In short, I really approached everything very comprehensively and didn't leave much to chance.
TARGET PACING
In my research, most books I read about training had the concept of a "target pace" featured in the plans and workouts. Daniels is the exception since he uses VDOT whereby target paces and training paces are derived from performance. I had run a 1:26:10 half-marathon back in March (at the time it was a 7:20 PB, but then again I hadn't run a half-marathon other than during a triathlon since September 2010), so based on VDOT I knew a 3-hour marathon was within reason. Validating that as a target for the actual race took some time, and when I started training in the heat of the summer I was certainly not performing to the VDOT set in a March race in 40-degree weather. But as a rough goal for setting up my training and workout target pacing that was what I used.
TRAINING PLAN SELECTION AND CONSTRUCTION
I considered 4-5 of the "classic" runner's training books and bought all of them and read them before deciding:
- Pete Pfitzinger
- Jack Daniels
- Hal Higdon
- The Hansons (famous for featuring a maximum long run of only 16 miles)
After a lot of consideration, I initially elected to follow a Pfitzinger plan…although that's not what I ended up doing once everything was said and done. The thinking that led to my initial approach was something like this:
- I figured my biggest risk was injury. I had never done any kind of volume before (I'll discuss this below) and I wanted to ensure that the frequency, overall volume, long runs and workouts were all manageable.
- I had some schedule peculiarities, in particular that I wanted to take Friday off…Friday is typically a busy day for me and I have trouble fitting in a workout. Also I wanted to run 5x/week most weeks, and not 6x/week or every day. Mostly this was due to schedule issues but also due to the volume concerns mentioned above. Finally, two weekend workouts would only work if I could have the option to do the Sunday one in the evening after the kids were in bed…two weekend mornings reserved for workouts wasn't part of my reality this summer. So plans that had a workout the day before the weekend long run would be tough…Friday/Saturday would be inconsistent at best, and although Saturday/Sunday could work, the thought of spending the whole weekend with the long run hanging over me then heading out the door at 7pm on Sunday to do 18 or 20 miles wasn't palatable at all.
- I like threshold work and it has always worked to improve me as a runner
- I wanted a plan that was pretty prescriptive. Without any experience in marathon training, I wanted a plan and not guidance to support my own creativity.
With respect to how that thinking reflected in my evaluation of the plans:
- The Higdon approach whereby you did weekly marathon-pace runs followed by a long run didn't work schedule-wise
- Daniels wasn't prescriptive enough for me. His plan prescribed some quality workouts (some of which I seriously doubted I'd be able to complete after just reading them), and then essentially said to fill in extra mileage up to your selected weekly mileage via your other runs, with each week having a "% of peak mileage" that dictated your target. This was too loose for me. Plus there was all sorts of stuff around the long runs being either % of a mileage target or a specific mileage…it would have taken spreadsheet work to figure it all out. This wasn't what I was looking for. Finally his book heavily features shorter-distance races and essentially has a single plan for long distance that is meant for both marathon and half-marathon training. So it didn't feel purpose-built.
- Pfitzinger had both 55 mile-per-week plans and 70 mpw. The 55 mpw plan was 5 days of running most weeks, which I thought would be great for me since it was essentially the lowest mileage plan. I didn't love the fact that a massive % of the runs were easy…coming from EN where there are so many quality workouts I was surprised that the LT workouts were essentially once every 10 days or so. But some of the workouts appealed to me, and looked pretty challenging.
- Hansons intrigued me the most. As most of you know, the plan is that you run a lot but the longest runs are only 16 miles. But you run 6x/week and you have no rest before or after your long run days, and there are a bunch of back-to-back long run days as well as hard sessions the day after pretty long runs. The overall weekly mileage peaks in the mid-60's. I really loved the structure of the quality workouts ("strength" and "tempo"…more on this later). But at the end of the day, I thought that as a first-timer it would be too risky to run a marathon without any really long runs under my belt. Also I didn't know how I'd be able to hold up to 6x/week of running (plus the difficulty in scheduling that 6x/week requires).
I picked the Pfitzinger plan and started my training. However, within a couple of weeks there were a few problems:
- The long run was on a Sunday, and there was a short Saturday run of easy mileage. I was okay being a bit inconsistent about completing the Saturday run (hey, you need to make tradeoffs and adapt), but for a bunch of other reasons I wanted to move the long run to a Saturday. But doing that meant almost never getting in the day-before run.
- I felt like there were very few workouts. Again, maybe it's the EN approach that I'm used to, but it felt like almost workout was 10 miles "general aerobic" or 10 miles "general aerobic plus 10 strides" or "5 miles recovery"
- This will sound silly, but I didn't love the fact that there was minimal consistency around days of the week. Long runs were on Sunday and Monday was a day off, but all other days could be "anything". Speed, LT and easy workouts could be on pretty much any day.
- Most critically, when there were quality workouts, I found them incredibly difficult to complete. The LT work is structured as tempo runs at half marathon pace (~6:32 for the 3-hour target). Some of these were 7 or 8 miles of HMP. Maybe it was training in the heat of the summer, but I was looking at a litany of failed workouts.
So, after 5 weeks of Pfitzinger, I did a pivot and decided to create a training plan based on a hybrid of Pfitzinger and Hansons. Essentially the hybrid looked like this:
- Overall mileage based on Pfitzinger
- Long runs based on Pfitzinger
- 5 days running per week (consistent with Pfitzinger)
- A weekly structure based on Hansons
- Most importantly, each week had 2 quality runs from Hansons which were very EN-ish and suited me well
- I also kept a few of the really "key workouts" from the Pfitzinger plan, for example an 18-miler with 14@MP
All of this translated into a very specific and consistent calendar for the last 12 weeks. I'll describe the actual workouts in more detail below but the general structure looked like:
- Monday – LT run (Hansons calls this a "strength" workout)
- Tuesday off
- Wednesday – MP run (Hansons calls this "tempo" but it is MP, I will explain below)
- Thursday – easy mileage
- Friday off
- Saturday – long run
- Sunday – short mileage recovery
TOTAL VOLUME / MILEAGE
The summary numbers from Monday June 17 through Sunday October 13, 17 weeks:
- 87 runs, 32 rest days, out of 119 days in total, implies ~1.9 rest days/week
- Only 1 run per day…no doubles!
- 100:05:20 total running time, 859.57 total running miles
- Average run length of 9.88 miles and 1:09:02, quite high but counts a weekly long run that was usually 16-20 miles, and hey, to hit 50mpw running 5 days a week, the math has to work…
- Amazingly, and in stark contrast to tri training, the weekly time commitment was under 6 hours/week (5:53, although that only includes the actual running and not the cooling down, showering, uploading workouts, etc…&hellip
- Average pace 6:59/mile (higher than I thought or targeted, but really is the output of doing the workouts, ~22-24 miles of which averaged ~6:40s)
- Average HR of 162 bpm, 90spm cadence
- According to Garmin estimates, 97,235 calories burned!!!
As I mentioned, the total mileage was based on the Pfitzinger 55mpw plan. The below chart shows my weekly mileage over the build period. The bars are what I did and the black line is the plan. You'll see I exceed the plan on a few weeks, and the reason for that is largely circumstantial – I did my long runs with a group and on a couple of weeks the group ran longer than my plan called for, and also on some Sundays I increased my mileage because I ran at night (i.e. more rest after Saturday morning's long run) and/or I knew Monday's workout would have to be pushed to Tuesday for schedule reasons so I could do a few more Sunday miles without risking the next quality workout.
As you can see in the top right corner, the weekly average over the training build was 49.5 miles. From July 22 to Sept 27 – i.e. the 10-week "heart" of the build – the weekly average was 55.76 miles with every week over 50 miles. That is a far cry from the higher-mileage plans that go to 70, but there was very strong consistency in the mileage week-in and week-out. Also, to put the mileage in a personal context, consider that from January 2011 through until the start of the above chart, my weekly average was 19.98 miles with only 7 weeks above 30 miles and with the highest-ever week being 40.35 miles:
Going so quickly into such a higher-volume training scheme worried me a lot and I kept waiting for an injury. But luckily I was able to accommodate the volume and didn't get injured. The whole process has given me incredible confidence in being able to hold up to running volume that is greatly in excess of the triathlon training norm. Going forward I almost can't imagine running less than 30 miles/week as the "new normal", and will consider overall run volume in how I'll train for triathlons going forward.
In terms of time commitment, running is of course a very low-overhead sport. All you have to do is walk out the front door! I travel a lot for work and running his highly compatible with travel (unlike biking and to a large degree swimming). I can squeeze in a 55-minute run into an hour and a half at lunch or between meetings even in the heat of the summer where I need time to cool down after the workout. As shown in the numbers above, my average weekly volume was just under 6 hours, excluding the "overhead". About 20-30% of that time was on Saturday morning, so weekdays were obviously less.
Finally, I should mention that biking in that 17 week period was exactly ZERO. In fact I did no aerobic crosstraining, just running.
REGULARITY / CONSISTENCY
As mentioned, I was highly compliant to the plan. Although some workouts were failed, I always got in the mileage with revised goals. In 17 weeks, I only missed a single scheduled run workout…a 5-mile recovery run on July 16. Achieving this consistency required:
- Flexibility to move workouts around within the week, including doing back-to-back hard days quite a few times
- Willingness to do the workouts at odd hours. Many times this meant running very late at night after my kids were in bed and my work email was cleared. I actually did quite a few 10-14 mile weekday runs beginning after dusk.
- Willingness to run in virtually any weather. I didn't do any treadmill running so in heat waves with 90-100 degree temps and lots of humidity I headed out the door. If it was raining and dark, I headed out the door. It if was 30mph windy, I headed out the door. In the heat waves I tried to look at the weather forecast and shuffle workouts to move the hard stuff when it looked to be cooler and/or in the evening, but that didn't always work. Some of my interval workouts were failed due to weather, but I always got out there and attempted them, and I ALWAYS got the mileage done.
- Mental resolve to get the workouts done on days when I didn't feel like it. I recall an elite runner quoted in Runner's World once saying something along the lines of "listen to your body when it comes to injury and illness, but when it comes to being tired, don't listen too much or you'll end up missing too many workouts"!!
- An accommodating and understanding and supportive family!!!
LONG RUNS (and group training)
For my long runs, I decided to run with a group organized by the Chicago Area Runner's Association (CARA). They have a highly organized marathon training program with hundreds of participants. Most of them follow a Hal Higdon plan since Higdon is one of the CARA forefathers, but my objective was just to get some good pacing help and company for the long runs so I stuck with my plan and just showed up at 6:30am on Saturdays to run long with the group. Due to vacation or schedule issues I did do some of these runs alone or with a training partner (including 2 of the 20-milers), but in general I tried to get out with the group.
Using the group for pacing discipline was key. Those who know me know that I'm pretty much a single-gear runner and always do my workouts pretty hard. I knew that getting into serious mileage would require turning down the dial and learning to run at EP. With that logic in mind, and with the books saying that EP for me should be around 8:00/mile when targeting a 3-hour marathon, I started with the 8:00/mile pace group. Honestly, it was too slow. It just didn't feel right. Then as I started reconsidering my Pfitzinger plan I read Hansons saying that for a 3-hour marathon my "easy runs" and "long runs" should both target 7:29/mile.
So taking Hansons into consideration I ran with the 7:30/mile pace group for my long runs. That said, the group almost always ran around 7:20 for the first half and negative split the run. Furthermore, most folks practiced finishing strong by speeding up in the last 4 miles and targeting MP for the last 2-3 miles. So in practice my long runs were strongly negative split with the last 2 miles in the 6:40's. Average pace for these runs typically ended up between 7:10 and 7:15/mile. Looking back, these were not at all EP runs, and frankly averaged only ~20 sec slower than targeted MP and ~30 sec slower than actual MP. But they were certainly low HR and very conversational until the last 4 miles or so, so it was more of "easy running then finish strong", which I guess might get credit for allowing me to maintain consistent splits and lay down a couple of 6:30 miles at the end of the marathon on race day.
The above pacing was also probably facilitated by taking most Fridays off. Most of the plans have a shorter run the day before the long run. Some are pretty substantial, for example Higdon who puts a weekly MP run the day before the long run or Pfitzinger who inserts Saturday tune up races up to 10k into the plans on some weeks before a Sunday long run. I didn't do that for the most part and only had a run the day before my long run 6 times:
- Twice when I was still adhering to Pfitzinger (he calls for that, I guess you might say I missed 3 of 5 rather than say I did 2!)
- Twice because my schedule backed up and I did the MP run on Thursday so needed to get the easy run done on Friday or else miss it
- Once when on vacation to scout a route for the next day's solo 20-miler in Michigan
- Once because I ate a gigantic meal the day before and felt the need to burn some extra calories (bad logic, I know)
In terms of run length, I used the Pfitzinger 55mpw plan for long runs, which oftentimes meant I had to tack on some extra miles after the group runs (usually 2-4, never more). That said, when the group ran longer than Pfitzinger I ended up running whatever the group was going to run that week. The Pfitzinger plan includes 3 20-milers in the build and the CARA/Higdon plan had 2 20-milers and a 22-miler. The plans lined up on some weeks but overall I ended up doing 4 runs of 20 miles or more (3x20's and the 22). There were also 3 runs of 18 miles. Overall the average long run length was 17.3 miles over the 17 weeks (again, the bars are actual and the black line is planned):
LT WORKOUTS
For the weekly LT workouts (always Monday or Tuesday), I used the Hansons "strength" workouts with a slight modification. These workouts are very similar to threshold work in the EN plans. In the Hansons plan the workout is a 10 mile run with 6 miles of running at just a bit slower than TP. Each workout starts and ends with an easy 1.5 miles and the middle is the hard stuff – 6 miles of intervals and around a mile of recoveries interspersed. Actually the recoveries usually added to more than a mile so the overall workout was 10.5 or 11 miles. Hansons lays in 7 weeks of "strength" workouts starting with 8 weeks remaining prior to the race. The main sets progress as follows: 6x1mi (400m), 4x1.5mi (800m) , 3x2mi (800m), 2x3mi (1000m). Then in the remaining weeks they ladder down symmetrically: 3x2, 4x1.5, 6x1.
Since the Hansons progression doesn't start until 8 weeks before the race and I cut over to my hybrid plan with 12 weeks to go, I just ramped up the "strength" workouts earlier and a bit slower, starting with 2 weeks of 4x1mi then 2 weeks of 5x1mi before hitting the Hansons progression that starts with 6x1mi. It ended up being a good transition from the more familiar EN workouts.
As you can see, the Hansons "strength" workouts are very similar to EN workouts, except the EN workouts have longer recoveries and less interval mileage (EN is usually 3x1mi, 2x1.5mi or 2x2mi). But there is another difference, which is that Hansons has you targeting 10 sec/mile faster than MP for these (~6:40 in my case), and not targeting TP as the EN plans would or HMP as Pfitzinger would do. Of course Hansons also has you go into these workouts after two back-to-back days of long runs and I didn't have that fatigue in my legs. So for thee intervals I compromised targeted around HMP (~6:30) but ran a lot by RPE and generally ended up in the low 6:20s on good-weather days and mid-6:30s on hot and humid days. So in short, I ran these pretty hard. I tried not to take any inactive recovery but on the hot days I did stop at a water fountain during many of the recoveries.
Knowing how hard the EN workouts can be, including the 2x2mi OS workouts, I was really worried about being able to complete more intervals with less recovery. Also the total mileage of these runs ended up being pretty significant. However, the workouts ended up looking harder on paper than they were in reality. I should note that the one LT workout I absolutely dreaded was the 2x 3-mi. I really psyched myself out on that one and it turned into a fail where I ran the day's mileage but didn't complete the main set. In retrospect that was probably psychological (although it was damn hot (90+) and humid the night I attempted the run, and was 2 days after a Pfitzinger "key workout" of 18 miles with 14 at MP). I failed 2 others, both on days with over 90-degree heat, but didn't worry too much about that given the conditions.
Overall my average pace for the LT runs, including warmup and mileage at the end, usually fell somewhere in the 6:40s in terms of average pace for the 10-11 miles.
MP WORKOUTS
The second Hansons element I included in my plan was a weekly marathon pace (MP) run. Like the "strength" runs, this workout starts and ends with 1.5 miles at an easy pace. In between you run at marathon pace. Like the LT runs, there is a progression. It starts with 5 miles at MP, then ramps up. In the last 12 weeks before the race, the MP progression is 2 weeks of 5 miles, 3 weeks of 8 miles, 3 weeks of 9 miles, then 3 weeks of 10 miles. By the time you get to 8 miles of MP the run is 11 miles in total and is no joke.
Psychologically these workouts were a bit damaging because they were very difficult, yet less than half of the actual marathon distance. If I was dying trying to do 9 miles at MP during a workout (usually allowing myself to stop at a fountain every 3 miles), how on earth was I going to be able to do a whole marathon at that pace??
In any event, some of these workouts I felt great and others were much more difficult. A couple were fails, although not many…more often I would pause every 2-3 miles to drink water at a fountain, which I guess isn't a fail although it isn't a nonstop run as the Hansons prescribe or as Pfitzinger prescribes for his MP runs. One of the fails was the one the week before race week which I was nervous about having in the plan to begin with – 10 miles at MP seemed like a pretty hard workout 10 days out from the race so after attempting the MP running I did most of the 13 miles at or around EP and called it a day.
Notwithstanding the above, I certainly didn't shortchange all of the MP workouts. In fact, I tried to hold high 6:40's instead of the 6:51 MP because I knew I'd have to run faster than MP given imperfections in running the course, etc. One of them I did with a buddy at night and we nailed 12 miles with 9 at around 3 sec/mile faster than MP nonstop in ~80 degree heat. Another was a Pfitzinger "key workout" that I really wanted to include in my hybrid plan…an 18-miler with 14 miles at MP. I did this on September 8 (week 13 of 18) in lieu of my long run that week and in concert with a local half-marathon. I showed up to the race early, ran 5 miles at around 7:10 pace, then went to the start corral and ran the half marathon in 1:28:57 (after ripping the timing chip off my bib to ensure I wouldn't be tempted to go too hard). It was a humid day, not really a PR day for most people, and I felt good at the end and not spent. I took Monday off and failed my LT workout on Tuesday (in 90-degree temps), but did 55 miles that following week including the weekly MP run (12 miles with 9 MP) and a 17-miler on Saturday. Although I didn't finalize my marathon pacing plan until a few weeks later, that 18-miler which included the MP race was probably the key workout that cemented in my mind the validity of the 3-hour marathon goal.
Similar to the LT runs, overall my average pace for the MP runs, including warmup and mileage at the end, usually fell somewhere in the 6:50s for the 11-13 miles.
EASY RUNS
The LT and MP runs accounted for around 20-24 miles/week. The long run was ~16-20 most weeks. So that meant the remaining mileage up to the mid-50's was done in 2-3 "easy pace" runs. One was almost always on a Thursday after the Wednesday MP run, and one was almost always on a Sunday the day after the long run. In my hybrid plan I had to allocate the mileage to these runs to get to the Pfitzinger weekly total, so I usually tried to have Sunday be 5 miles and put the remainder on Thursday (maybe 6-9 miles or so, never higher than 9). I targeted 7:29 pace for these but usually ended up between 7:05 and 7:15 or so. If I did a true "recovery run" on Sunday morning it was very slow and rarely more than 5 or so miles (other than 2 weeks where I needed to do 7 to avoid loading up the Thursday). If schedule pushed the Sunday run to the evening I'd run some extra miles, usually a bit faster, and then postpone the Monday LT workout to Tuesday (implying back-to-back hard days of LT then MP on Tuesday/Wednesday).
I was highly compliant with running the core Thursday "easy run" quite slowly…not hard given the LT and MP runs stacked earlier in the week.
On 3 or 4 of the weeks I also added an extra EP run of 5-6 miles, usually because my schedule had me thinking I'd miss one later in the week. On most of the weeks I was over Pfitzinger's 55 miles, this was the reason (the other reason being a long run with the group that was longer than the one in my plan).
VO2 WORKOUTS
I didn't do much VO2 work at all during the training. Hansons had VO2 speedwork in the first few weeks of the program, but at that time I was following Pfitzinger. In the last 8 weeks Hansons had no VO2 work at all. Interestingly Pfitzinger had a few VO2 workouts sprinkled into the plan after the first few weeks, but by then I was doing all the Hansons workouts.
I do wonder if I ran some of my LT workouts at efforts approaching those that would create beneficial VO2 adaptations – from an HR perspective I got up to the mid-to-high 180's in some of those workouts for sure.
STRIDES
Pfitzinger is a big fan of strides and many of the "general aerobic" runs included 10x100m strides. So I did those at the beginning of the plan. Once I was into the Hansons workouts, however, strides weren't part of the program on a regular basis.
TAPER
As you can see by the above charts, I tapered significantly for the race, essentially beginning 3 weeks out after the last 20-miler. I kept up intensity but overall volume come down as did the volume of individual workouts. For race week I created a workout schedule based on both Pfitzinger and Hansons and then further adapted it for my work schedule. It ended up being 4 runs:
- Monday (6 miles with 4x half miles at TP)
- Tuesday (5 miles with 4x400m at IP, around 5:45/mile)
- Wednesday (6 miles with 2 miles at MP)
- Saturday (2 miles easy).
The Saturday run went against all of my prior history – I never run the day before a race – but most plans called for something on Saturday so I did it.
I think the taper had a huge positive impact on my performance. I don't usually taper much for triathlon because mid-season races are usually in the middle of training not at the end of it. So my "taper" is typically a full workout load up to race week or even up to the Tuesday or Wednesday of race week then a significant reduction (or no workouts) up to the race. This experience has taught me that a well-executed taper can be a powerful performance enhancer.
FUELING WORKOUTS
Put simply, I fueled all workouts. Not "go crazy and carb load the night before all of my long runs", but certainly I fueled every run over 5 miles (i.e. virtually all of them other than the Sunday recovery runs which were usually after a dinner out with my wife on Saturday night that was larger than it should have been!). I adopted the "fuels all workouts" approach since the 2013 OS, motivated in large part by Coach P's practice during his very successful IMTX build in 2012.
For my LT, MP and easy runs I had a gel before heading out the door and post-workout had a 12 oz or 16.9 oz Gatorade plus a PowerPar Protein Plus (those bars are 20g protein / 210 calories). For MP runs I sometimes did the gel after the 1.5 mile warmup instead of pre-run. For the long runs I had a good dinner the night before – nothing crazy but not a salad either, then a ritual of 3 shot bloks before the run followed by gels at miles 4, 8 and 12.
I didn't do any "carbohydrate staved workouts" in an attempt to teach my body to burn more fat. Honestly the instructions for those as published in various places are so precise as to make me really suspicious as to if it really does anything other than make you feel like crap during the workout, or to make you feel really okay about underperforming a workout. In the book Racing Weight the concept of nutrient timing resonated with me, and the practical application of that is essentially concentrating a lot of your eating before, during and after workouts. So with all that in mind, I fueled all workouts and honestly I think that had a lot to do with performing well in most of the workouts. Essentially all my failed workouts correlated with ridiculously hot and humid weather, and virtually none with feeling tired or "off". So I think the fueling really helped.
FLEXIBILITY TRAINING / STRETCHING
I didn't do any stretching or flexibility training. When I first got off the couch I used to do static stretching after my runs. Then I got injured and seriously agitated the injury by stretching. Ever since then I've done zero stretching. As far as I can tell, the research is very mixed as to if it aids in injury prevention or performance improvement. In my past it has been a risk more than a positive factor, so I left it out of my training.
STRENGTH TRAINING
I'll preface this by saying I've historically never done any strength training as part of my triathlon training. Mostly because I don't find it fun or engaging…it's a chore. But for the marathon training I observed that a lot of the books promote core strengthening. I decided to include some strength training into my plan in the hope it would do two things:
1. Reduce the likelihood of injury
2. Help me maintain running form and mechanics when fatigued later in the race
Consistent with the above, my strengthening program had two elements:
1. A set of "myrtle exercises". This is a lot of the glute medius stuff that is meant to prevent running injuries. I quickly worked my way up to doing this 5-6x per week. Two circuits of:
a. Clamshells - 25 reps each leg
b. Lateral leg raises - 15 reps each leg
c. Donkey kicks - 15 reps each leg
d. Fire hydrant - 15 reps each leg
It takes 12 minutes to do the above, so I really forced myself to do it every day. I made sure to do the clamshells and the leg raises with my back and butt agaist a wall, so the movement is totally lateral and it isolates the muscles as well as can be done. There are more myrtle exercises you can do, check out the following video. I picked the ones I picked because (a) I did them in PT after prior injuries, so I figured they are the best for injury prevention, or (b) they look to be focued on glutes which are important for running mechanics and often weaker than they should be in runners: http://www.youtube.com/watch?v=2GLrKr54yA0
2. A set of core-strengthening exercises. These are meant to increase your ability to hold together your running form late in long races. I did these 3x per week. The ab stuff was painful even after many weeks of doing it so either I'm didn't make good progress or they are just difficult (although I increased the reps and hold times of some of the exercises per the book I got them from). Attention to the proper form when doing these is really important. Two circuits of:
a. Abdominal crunch - 20 reps
b. Leg pushaway - 20 reps…these are REALLY TOUGH if you do them right, and really easy if you don't!
c. Staff - 6 reps of 5 sec
d. Single leg bridges - 3 reps of 15 sec per leg
e. Back extension - 5 reps of 10 sec
f. Side hover - 3 reps of 15 sec per side
The above takes 20 minutes in total and is a pain in the ass, but I stuck with it.
Overall my compliance to these workouts was okay but not great. Motivating myself to do these was tough because I really didn't enjoy them at all. I basically had to give myself a talking-to each time and say "you're doing all the hard training, why throw away your chance at running the race you want because you can't do 10 mins of leg raises!". You can see by the below that despite coming in 10-20 min increments, the total volume added up to ~1.5 hours/week on average. That's a lot! I usually did the routines before bed. Again, the bars are actual time spent, the line is planned workout time:
The thing that really annoys me is that I can't say for certain if these exercises really made a material difference in my race performance. I just can't get a handle on the ROI at all. In triathlon where I'm starting to see the input-output equation get more and more challenging it becomes tempting to add things like the above to the overall training regimen. But I don't enjoy the exercises and can't be certain of the ROI have some decisions to make about doing these in the OS and/or next season. Tough call…
OVERALL TRAINING STRESS
Qualitatively I'd say I felt pretty good throughout the training. Some days were better than others but my body held up pretty well to all the workouts. Quantitatively the plan and my compliance to the workouts yielded a very consistent PMC profile. I'm not sure exactly how TrainingPeaks estimates rTSS (based on HR? on pace relative to the my zones?), but the bottom line is that every week had a pretty similar IF in the 0.96 to 0.98 range, and so the volume profile is what drove the ATL and CTL. Because the overall volume is pretty steady, the ATL is pretty steady and the CTL, after declining at first (because the moving average had a lot of bigger volume tri training in there for a while), drifts up a bit from 92 in mid July to 108 before the beginning of the taper in late September. But overall the CTL doesn't move a lot…I was building endurance for sure but my overall fitness probably didn't increase all that much (see the last comment in this write up for more thoughts on that). Note in the PMC screenshot I went all the way back to the beginning of the OS so you can see the remarkable difference in profile between tri training and run training (btw the massive spikes are the American TTT that I raced in May, followed by 70.3 Kansas in early June).
RACING SHOES
I did virtually all my training in my Asics GT-2000 shoes with green Superfeet inserts. Very stable and lots of support. I raced in my Addidas Adizero 2.0 shoes. They have a decent amount of drop (9 or 10mm I think) but have a lot less support. I have raced in these shoes before (the spring half marathon I mentioned) and I did one training run in them (the 18-miler "key workout" with 14 miles at MP). That gave me the confidence that I could wear them for the marathon. I like racing in shoes you don't wear often since I think there's a psychological benefit. That said, if I did it all over again I'd probably do more training in my racing shoes since my feet were absolutely killing after the race and my ankle muscles get taxed in all sorts of new and different ways when I wear those shoes. It didn't impact me during the race but after the race (and after the 18-miler too) I was in rough shape for a few days.
FAILED WORKOUTS
I certainly had a handful of failed workouts. That said, it wasn't more than 5 or 6 out of 87 runs, and I certainly completed all of my long runs feeling great. So when I failed a workout I just slowed down, put in the day's mileage to complete the run and let it go. Psychologically the fails didn't impact me as much as I thought they might, possibly because some of the successful workouts were ones I thought I might fail (e.g. 4x1.5 mi and 3x2 mi on hot days), or 9 mi at MP hitting mid 6:40s).
SETBACKS: INJURY AND ILLNESS
Amazingly, I had no material injuries during my marathon training. This stands in contrast to every triathlon season where SOMETHING happens! There were 2-3 points where I had what I thought were calf strains, including one that felt pretty bad after my first 18-miler. I was aggressive with ice, skipped a day then did a hard LT run on it. It didn't feel great but it didn't feel worse. So I skipped a day, continued ice and did another hard workout (MP). Again it didn't feel worse and eventually it cleared up within a week. The same strain re-appeared again later my training but I had the confidence to handle it the same way (and of course always being ready to stand down completely if it was getting more damaged). I has similar strains in my Achilles and hamstring at some point but they were not bad and cleared up in the same way. In short, I was very lucky.
The one real setback I got was in early September. It was a really stubborn cough/cold that turned out to be bronchitis. I trained through it and actually did some pretty epic workouts during that timeframe (a few times needing to stop during the recovery after an LT interval to hork (sp?) up some pretty "interesting" things!!). During this time I was really truly sick and my HR in many workouts was simply higher than it should have been and higher than the norm. Eventually I went to the doctor and got some antibiotics that cleared it up fairly effectively, but I went to the doctor about 3 weeks later than I should have. Lesson learned. I was lucky.
Overall I was extremely lucky not to have any real setbacks.
MASSAGE
On the referral of a colleague of mine I have come to know a woman who does deep-tissue massage. She comes to my house, brings a table and all the equipment, and charges $60/hour. Sign me up!!
The one problem with deep-tissue massage is that it is about as destructive to my muscles as a hard workout, which is to say I can't do it without a bit of recovery after my last workout and I certainly can't run the next day. So scheduling the massages was a bit tough esp since when I'm at home there are a lot of things to do to keep the family running. But I tried to get a massage with some regularity and managed to do 10 in the 17 weeks of training.
NUTRITION
This is almost a different topic given the detail I could easily get into. But it is critically important and featured heavily in just about every running book. For better or for worse, I didn't make any major dietary changes during my training other than eating less (because I was working out a lot less than during triathlon training). I eat reasonably healthy but I'm not fanatical and don't subscribe to any major dietary protocols (for example I do eat "recreational sugar" although I try to avoid desserts in general, I do drink alcohol but not heavily, I do eat red meat but try not to eat gigantic steaks too often, etc.). For social reasons I'm a bit lax about diet – I do a lot of client dinners for work, and if I go to a steakhouse and order a piece of fish there are some negative implications. Likewise my wife and I get a babysitter almost every Saturday night and eat out, and is not fun for her if I'm starving myself or special-ordering a dry roasted piece of fish. Honestly, it's no fun for me either. I could have done better on this dimension and may choose to in the future.
One thing most of the books discuss is the need to have a diet skewed more towards carbs. In general up to 70% carbs. Regardless of if you count by weight (58.3%) or by calories (53.9%), I was well under-target for carbs:
You might think I wasn't trying…but I was!! I ran the above charts for Jan 2013 up until the start of marathon training and they show 49.4% and 55.2% carbs by cals and weight, respectively. So I certainly increased my carbs…but not by even close to what would have been required to get to 70%. Clearly hitting a 70% carb target would require a FUNDAMENTAL re-think of my diet. It's certainly something to consider trying…could be an interesting experiment…
For the record, I seemed to do pretty well on micronutrients, but I assure you that is luck because I do NOTHING to manage to those numbers:
WEIGHT AND CALORIC BALANCE
As I mentioned in my race report, I didn't lose much weight during the marathon build. Now that I've run the numbers, it's not too surprising. The above chart shows I consumed ~290k calories during the 119 day training build. Assuming my RMR is ~1,600 cals/day I burned ~190k calories from my RMR, leaving a surplus of ~100k calories. Garmin says I burned ~97k in my workouts, so there you have it. Almost perfect balance!! In fact I probably ate a bit more (there are some unlogged meals I have written down but not yet entered due to their complexity), the RMR is actually a bit less than 1600 plus it double-counts time spent working out, and all this sit in some plus/minus error window. But the point is, the balance was, well, pretty balanced. And I was already in pretty good shape with low BF% before I started the training, so I ought not to have expected to lose weight.
I did apply a lot of my usual eating "rules" during the build, for example no food after 8pm, practice "nutrient timing", etc. So I got some relief from the harshness of the numbers I think as a result of that.
Although the chart above looks like a big success notice a few things:
- the y-axis shows only a 4lb range
- the starting point reflects a week of doing nothing but eating after 70.5 Kansas
- the weekly weigh-in and my ability to "make weight" by choosing what day I weight in each week(!!) masks a lot of variation and a lot of days over 140lb
To reinforce the second point above, consider the chart for all of 2013 to-date:
As a side note, the BF% is probably too high. When I set the scale to "athlete mode" it shows a BF that is ridiculously low. I don't believe either number so I just go with the setting I've always used for consistency.
The final piece to the "you don't lose weight when you don't implement weight loss practices" story involved daily caloric balance. By the above you can see I didn't run a systemic deficit and didn't really alter my macronutrient balance. Well to top it all off, I also got into the typical cycle of running deficits then eating too much on weekend, at meals with clients and colleagues and family, etc. I haven't counted the number of 3500+-calorie days or days where the balance was over 1000 surplus, but it's not insignificant:
OVERALL ASSESSMENT
At the end of the day I think my training worked out pretty well. I feel like my preparation allowed me to race to my capabilities. My marathon time of 2:57:47 was 2 minutes faster than that implied by my spring half marathon (VDOT 54.3 vs. 53.6). I certainly learned a ton about myself and about run training. That said, now that I've had more time to reflect, I'm not sure I truly improved much as a runner…less than a 1 point increase in VDOT. In fact, I really wonder if my VDOT improved at all, considering that my racing history seems to show consistently better performance in longer distance races. In other words, I'm pretty sure I'm physiologically wired as a "diesel engine" – lots of endurance and not much top end speed – and I wonder if I ran a half marathon instead of a marathon on October 13 if I would have beat the 1:26:10 I ran back in March. Of course back in March I likely wouldn't have been able to complete a marathon in 3 hours, so the training really was acting to facilitate the VDOT at a longer distance. But boy that's an awful lot of training for not much "improvement". So notwithstanding that I feel I had a great race and performed to my ability, I have some nagging uncertainty as to if the training actually improved me or not.
Oh, one final note, if you have questions about how I logged the data to create the charts above, send me a PM or start a different thread. I didn't mean for this thread to be about logging and charts. The short answer to 90% of the questions is (a) I used TrainingPeaks which costs me $119/year and (b) I am OCD about tracking and logging everything.
Comments
I really don't understand why the first image doesn't appear. It is a table with a calendar showing 12 weeks or workouts, so pretty key! I've tried 4 times but it won't show up when I submit the post, despite showing up iin the preview. I'll try one more time here but then am giving up. This site is very tempremental and my patience is officially gone...
Totally awesome - will be a pleasure to run the relay-marathon "with" you next week
1. Congrats again .
2. You lost 70lbs wow!
3. Man are you detailed and organized.
okay 4. great write up and thanks
Awesome resource, thanks for sharing
This definitely needs to be archived in the Wiki.
Wow that is some amazing detail! Really enjoyed reading your approach.
And wow a sub 3 on your first marathon is pretty amazing. I know people who struggle for years. One and done? You have to run Boston at least once.
Thank you Matt, this post is a thing of beauty! AND very, very valuable for both those of us interested in doing a Marathon (I have never done one) and in getting the most out of our running for HIM/IM. I agree with you that some of us are wired to be diesel engines, and so our Vdot improvements come slow and over a long period of time. I struggle with this as well. My instinct from my previous life as a runner is that this kind of focus period is critical for making a quick gain in Vdot if you were to follow it up with a second focus period on speed. It creates an endurance base/depth that would allow you to hit it really hard focusing on 10k kind of speed work and fight your way to a new level. Great Stuff!