Race Post-Mortem: Rock and Roll Philadelphia Half-Marathon

Some of the DC Front Runners who ran the Philadelphia Half-Marathon.  Can you see the humidity in the air?  Photo courtesy of Marcel.

Sunday, September 21, 2014

Philadelphia, PA
Difficulty score: 1/10 (0 for weather, 1 for course profile, 0 for altitude)

This past Sunday, I ran the Rock and Roll Philadelphia Half-Marathon as a tune-up for the upcoming Chicago Marathon.  My intentions in running this race was not only to get back into the racing mindset as my last race was the Pike's Peek 10K this past April, but also to get a half-marathon time to serve as a basis for determining a realistic goal for Chicago.

I finished the half-marathon in 1:25:12 (6:29/mile), an improvement over my PR on the same course last year by 13 seconds.  This came as a pleasant surprise for me as race morning weather, 68 degrees and 88 percent humidity, was far from ideal for a PR.  I was prepared to finish this half-marathon one minute or two slower than my half-marathon PR.  Initially, I thought that was going to happen when I finished the first mile in 6:41.  But during the next few miles, I settled into a 6:30/mile pace and was able to maintain that pace until mile 10, at which point I accelerated to a 6:25/mile pace.  I cannot say exactly how well I would have done had race day weather been the ideal racing conditions of this race last year.  But my ability to run a half-marathon in this warmth and humidity faster than I did last year in temperatures in the low fifties gives me hope that I will be able to run a marathon PR in Chicago.

Training and the Days Leading Up to the Race

In preparation for Chicago, I had increased my mileage quite a bit.  During my training for Shamrock, my weekly mileage was in the high forties and low to mid fifties for most weeks.  But during this cycle, my weekly mileage was in the high to mid fifties and low to mid sixties for most weeks.  I had hoped the increase in mileage would help me improve my marathon time even more.  My weeks consisted of one tempo run on Tuesdays, a mini-long run of eight to fourteen miles on Thursdays, an easy run of six to ten miles on Saturday, a long run on Sunday, and recovery runs of five to six miles on Monday and Friday.  I had to sacrifice one tempo run each week in order to be able to achieve this kind of mileage; whereas during my training for Shamrock, I did two tempo runs per week, during this cycle, I only did one tempo run on Tuesdays.  The remainder of my miles for the week were easy or recovery runs, which meant that most of the time, more than 90 percent of my weekly mileage was run at 7:40/mile pace or slower.  The mileage itself was strenuous; I could not imagine running a second tempo run each week.

The Philadelphia Half-Marathon took place the week after a 61-mile week, my last truly high-mileage week for the entire cycle.  After a tempo run on the Tuesday before the race and a recovery run on Wednesday, I began a brief, four-day taper for the half-marathon consisting of an eight-mile run at around 8:00/mile pace on Thursday and a slow, four-mile shakeout run on Friday.

Before the half-marathon, I had been checking the race day weather obsessively.  The forecast at the beginning of the week indicated sub-optimal conditions; 62 degrees at the start of the race and relatively high humidity.  As the week passed, the forecast progressively got worse; by the day before the race, most weather sites predicted the temperature at the start of the race to be 68 degrees and the humidity to be in excess of 80 percent.  What was especially annoying was that the weather on the morning of that Saturday before and on the morning of that Monday after were both predicted to be much better; lower humidity and temperatures in the high fifties.  I adjusted my goal time for this half-marathon from running a PR to a time between 1:26:00 and 1:27:00.  This was somewhat annoying, but this was not my target race.

The Race

The morning of the race, I told myself that I should just power through the unpleasant racing conditions.  Warmth and humidity should not have had as drastic effect on my performance in a half-marathon as it did on my performance in a marathon.  Even if I was unable to run a PR this time, I should have been able to run a 1:27:00 or faster on this course.  Besides, the Frederick Half-Marathon in 2012 took place under similar weather conditions, yet I was still able to run a PR there.

Philadelphia is a fast course, and I knew that I at least could use that to my advantage on race day.  During the first four miles, we weave through Center City, down Broad, Market, and Arch Streets, before returning to the start near the Art Museum and heading north on Kelly Drive along the Schuylkill River.  We run along the Schuykill River for about four miles before crossing a bridge to the other side.  We the run south on Martin Luther King Drive along the other side of the river before the uphill finish near the Art Museum.  For the most part, this course is flat, aside from a gradual but long downhill segment between miles 5 and 6, a gradual and long uphill segment before the bridge crossing the river, and the relatively steep but short uphill finish.  Not only was more than half of this course adjacent to the Schuylkill, but much of it was also shaded and surrounded by trees; I anticipated the river and the trees to be very helpful in staving off overheating during this race.

Map of the Philadelphia Half-Marathon course with mile markers from the website.

Originally, I made a plan to take the first two miles a little slower and settle into a 6:35-6:40/mile pace, which I would sustain until after the bridge across the Schuylkill at around mile 9, at which point I would reassess and decide whether to accelerate or hold that pace for the remainder of the race.  My first mile was indeed my slowest at 6:41; many people flew by me during this first mile, but I ended up passing many of them back later in the race.  After training through the hot and humid DC summers for the past four years, one thing I had learned about my body was that in order to run my best on hot and humid days, I needed to start slow.

The second mile was somewhat faster at 6:36 and the third mile was significantly faster than my original target pace range at 6:28.  I finished the fourth mile, meanwhile, at 6:19, which would turn out to be my fastest mile during this race other than the last mile.  This was mostly an effect of adrenaline; as the runners starting in the earlier corrals return to the start during this mile, they get to see those in later corrals starting their races.  Many of the runners from the later corrals were quite enthusiastic, and this had an energizing effect; due to this, together with the flat and straight course in this segment, this mile has been fast for me all three times I ran this race.

Me at around the 5-mile marker.  Photo courtesy of Lavar.

As I headed toward the river, my pace slowed slightly to a 6:28/mile pace.  Lavar was on the side of the course near the mile 5 mark, watching the race; the photo to the right is of me at this point in the race.  Over the next three miles, I focused on maintaining about a 6:30/mile pace and on running the tangents.  Surprisingly, even in a race of more than 15 thousand participants, I had plenty of room around me throughout most of the race and I was able run most of the tangents.

Mile 9, not surprisingly, was the slowest mile aside from the first one.  This mile consists of a long, gradual uphill leading to the bridge across the Schuylkill, and many finishers of the Philadelphia Half-Marathon would agree that this one is one of the more difficult miles.  But having run this race twice before, I knew what to expect during this portion of the race, and told myself not to panic when my Garmin indicated a slower pace than the preceding miles and to conserve energy for the remainder of the race.  After all, the last four miles down Martin Luther King Drive were very fast and perfect for a late-race kick.

After I crossed the bridge, I began to accelerate, aiming for faster than a 6:27/mile pace.  However, this was where I began to really notice the effects of the humidity.  I felt like I was working; I was able to go a little faster, but I struggled to go much faster than 6:27/mile pace.  Miles 10 through 12 were all between 6:22 and 6:27.  Even though these miles felt difficult, I was happy to find that I was still passing plenty of people during this part of the race.

When I crossed the 12-mile marker, the clock read a little over 1:18:00.  I then realized that I actually had a chance at a PR during this race; I needed to run the thirteenth mile in about 6:15 to 6:20 and cover the last 0.1 miles in about 45 to 50 seconds.  After doing the math, and seeing Lavar again near the mile 12 mark, I picked up the pace, completing the thirteenth mile in 6:18.  At the hill leading up to the Art Museum and the finish line, I gave it everything I had.

I waited in the finish area for the other DC Front Runners to finish their race.  I was happy to report that even in spite of the warmth and humidity, several other members also got half-marathon PRs.  We all walked around the finisher's area before we returned to our hotels to shower.  Since Thorne and I were leaving at around 7:30 pm that evening, we decided to have lunch with some of the other DC Front Runners and explore Philadelphia a little.  We got to visit the Reading Terminal Market and the City Tavern, supposedly one of the most haunted places in the United States.  But otherwise, we really did not do much because we were so tired.  Note to self: if I want to explore a city, I should spend at least one more day there following the race, since I will not have the energy to explore the day of the race.

Next Steps: Chicago Marathon Time Prediction

As mentioned before, I had intended to use my time in this half-marathon as the basis of a prediction for an appropriate marathon pace in Chicago.  Since the weather was not ideal and hopefully was much warmer and more humid than the weather in Chicago would be on race day, my performance in this race may not have as much predictive power for my marathon.  I could adjust my race time for temperature and humidity and use the corrected half-marathon time, which would most likely be a minute or more faster, as the basis for my prediction, or I could be conservative and use the 1:25:12 as the basis.

I decided to do the latter.  Not only do I believe that being conservative in marathon pacing would hurt much less than being aggressive, but I do not know how reliable calculators to adjust for temperature such as Daniels Running Calculator are.  I cannot imagine anyone has actually validated these calculators; I actually am having trouble thinking of a study design that would be able to help assess whether the adjustments for temperature and humidity are anywhere near the true effect of these conditions.  The last thing I want is to believe a calculator that tells me I should have been able to run a 1:23:30, for example, in optimal racing conditions and pursue a goal based on that, only to have that adjusted time actually be an overestimate of my current fitness.

According to the Maclin calculator, a 1:25:12 on the Moderate setting of aggressiveness translates to a 3:02:04, or 6:57/mile.  I selected the Moderate setting given my mileage throughout this cycle.  If I did manage to run that time in Chicago, I would be very pleased.  I had qualified for Boston by 13 seconds during my last marathon, but due to the number of applicants exceeding the available number of slots in the Boston Marathon, only those who ran more than 62 seconds faster than their qualifying standards will be able to actually run Boston in 2015.  If I were to finish about three minutes faster than my 3:05:00 qualifying standard, then my chances of being able to run Boston in 2016 will be so much better.

So in Chicago, I will plan to target a 6:55 to 7:00/mile pace.  I will reassess at the 30K mark, and if I have the energy, I will accelerate at that point.