Race Post-Mortem: Hartford Marathon

Me at around the Mile 26 mark.  I was in more pain than I appear to be in this photo.  Photo courtesy of Jeremy.

Saturday, October 10, 2015
Hartford, CT
Difficulty score: 2/10 (0 for weather, 2 for course profile, 0 for altitude)

When I told people over the summer that I was training for the Hartford Marathon, most of their reactions were "why Hartford?"  I had actually had my eye on Hartford for a few years now; the Hartford Marathon had received almost uniformly great reviews, and I knew that mid-October in New England would be gorgeous.  It was also a mid-size marathon of a few thousand people, and I tend to fare better in those.  Plus, Jeremy, in his quest to complete a marathon in all fifty states, still had not done one in Connecticut and also wanted to run Hartford.  We trained together for Chicago last year, and I thought it would be fun to train together again.  Unfortunately, he was unable to run Hartford this year, but he was an excellent supporter and cheer squad.

The course was not particularly easy, although it was not terribly difficult either (the difficulty score above should be more like 2.5 rather than 2).  The first half had a lot of twists and turns, and numerous rolling hills.  After the start in Bushnell Park, runners weave through downtown Hartford before running along the Riverfront Walk by way of a narrow bike path that, according to the elevation chart, had some pretty nontrivial uphills.  Then we would be back on the streets of downtown Hartford before crossing the Founder's Bridge and running through East Hartford and along the other side of the Connecticut River.  The second half, which consisted mostly an out-and-back segment down a semi-rural road through South Windsor, had fewer twists and turns, but had a surprisingly large number of uphills and downhills.  Regardless, I'm happy with my time and I am glad I did this marathon.  For those looking for a well-executed fall New England race in which weather is rarely an issue and that is quite aesthetically pleasing (especially with the leaves changing color; I'm referring to that long out-and-back part down that rural road between Miles 10 and 23 and some segments along the Connecticut River near Miles 3 and 4 in particular), I would recommend Hartford.

Also, I even got a video of me at various points throughout the race.  The video shows the start of the race and then me at Mile 6, at the halfway point, at around Mile 17, at around Mile 25, and my big finish.  I cannot help but notice how often I check my Garmin.

Map of the Hartford Marathon, from the race website.

Training

I basically repeated my training schedule for Chicago, thinking that it would bring me similar results.  My training started with a three-week lead-in in which I would run six days a week and ramp up my mileage from about 20-25 miles per week toward the end of May and beginning of June to about 45-48 miles (not something I recommend; I at least had cross-training from a lot of dancing during the month before, I was able to comfortably cover 50 or more miles in a week in recent months, and all of the running was done at an easy pace).

Throughout the remaining 14 weeks of the training cycle, I would run six days per week most of the time, with the structure being pretty uniform:

• Monday: 4-6 mile recovery run at 9:00/mile or slower to recover from the previous day's long run.
• Tuesday: my reserved quality running day.  After the initial few weeks, Tuesdays were speed workouts with intervals at current 5K pace (5:55 to 6:00/mile) that became progressively longer (5 x 960 meters for two weeks, 4 x 1200 to 1280 meters for the next two weeks, and then 3 x 1600 meters) and then turned into 8-10 mile runs with 4-6 miles at tempo (10-miler to half-marathon effort).
• Wednesday: OFF
• Thursday: mini-long runs of 8-14 miles.  Most of these miles were at long run pace (7:40 to 8:20/mile) with about 15 minutes of current marathon PR pace running (7:00 to 7:05/mile) toward the end of these runs every other week or so.
• Friday: 4-5 mile recovery run at 9:00/mile or slower to recover from the previous day's mini-long run.
• Saturday: 6-8 miles easy (7:40 to 8:20/mile) plus hill sprints.
• Sunday: long runs of 12-21 miles, mostly at 7:40 to 8:20/mile with a few miles of marathon PR pace running.

I also ran the Navy-Air Force Half-Marathon as a tune-up three weeks before Hartford.  I had actually completed the vast majority of my training plan as originally written, subject to some minor rearrangement to accommodate life and friends.  I did get hit with a pretty nasty summer cold during the second week of the training cycle, forcing me to take half of that week off and readjust everything else, but that did not have much of an effect on the rest of the training cycle.  I did also neglect the Saturday hill sprints during the later weeks; although I doubt that would come back to haunt me later, they could have really helped. 

I ended up hitting similar mileage to that from my training in Chicago.  After that initial mileage building, my weekly mileage was in the mid-fifties to mid-sixties, aside from a few reduced mileage weeks (40-45 miles) for recovery.  My peak mileage was one 65-mile week two weeks before the Navy-Air Force Half-Marathon, which immediately followed two back-to-back 63-mile weeks.

Race Day

I arrived in Hartford on Thursday evening with Jeremy, which would give me a full day in Hartford before the race.  The day before was intentionally kept low-key.  After a three-mile shake-out run in the morning with Brian, who was seeking to run a marathon in all fifty states and still needed to finish Connecticut, we met up with his parents and Jeremy for some breakfast.  I then went to the expo to pick up my bib and some Shot Blocks for the race.  Jeremy and I returned to the hotel for a few hours and rested before going to see the Mark Twain House (something I definitely recommend for those visiting Hartford).  After a pasta dinner in downtown Hartford, we parted ways and I went to bed relatively early.

At around 6:40 the next morning, I took the shuttle to the start line downtown (this was another thing I liked about this marathon: like in Chicago, the city really works with the organizers to accommodate the runners).  Weather was perfect: sunny, high forties at the start, and never really rising above the high fifties by the time I was finished.

My plan was to target 7:00 to 7:05/mile for the first 30K of the marathon and to speed up afterwards if I was feeling sufficiently strong.  My primary goal was to qualify for Boston by a large enough margin so that I would not really have to worry about whether I would actually get to run Boston in 2017.  This was actually the first time I would need to run a 3:10:00 rather than a 3:05:00 to qualify for Boston; although I was only 33 on the day of the race, I would be 35 on the day of the Boston Marathon in 2017.  I realized that if I could just repeat what I did in the Chicago Marathon and the Shamrock Marathon (and my training and tune-up half-marathon both indicated that I was perfectly capable of that), then I would be more than five minutes under my Boston qualifying standard (a BQ-5:00), which would allow me to register during the first week of registration and give me a very high chance of actually being able to run Boston.  If not, I probably would still have a very good chance if I got close to being five minutes under my standard.

Given how tricky I thought the first 15K was, I also decided that I should not freak out if some of those early miles were more like 7:05 to 7:10/mile pace.  This plan, I figured, would give me a shot in finishing in 3:04:59 or better.  Given my mileage during this training cycle and my time in the Navy-Air Force Half-Marathon (1:25:55, although under less than ideal conditions of 70 degrees with 62 degree dew point), my plan wasn't completely harebrained; I could expect my marathon time to be about 2.15 times my half-marathon time, or 3:04:43 (7:03/mile).

Miles 1 through 9 (7:13, 7:08, 7:03, 7:11, 7:09, 7:04, 7:03, 7:16, 7:10): The first few miles were very easy; most of the segment before the Riverfront Walk was flat, with some noticeable downhills.  The uphills were also quite pitiful.  That, coupled with the fact that everyone was feeling fresh at this point, made it very tempting to go a little faster and try to bank some time, and many people were doing exactly that; during these early miles, I got passed by dozens of people, including the 3:10:00 and the 3:20:00 pace groups.  That completely baffled me as even as early as Mile 2, I was going significantly faster than 3:10:00 pace and definitely much faster than 3:20:00 pace.  Actually, after the 3:20:00 pace group leader passed me, somebody asked if that was the 3:20:00 pace group; I told them that the pace groups were way off, and not to follow them.

After we got to the Riverfront Walk, the course became much more undulating.  Although some of the uphills were quite substantial, like that long one before the end of the Riverfront Walk, none of them felt difficult since I was still feeling fresh.  I was happy that my pace was exactly where I wanted it to be, but I was also annoyed since at the Mile 4 marker, my Garmin already indicated that I covered 4.05 miles; but I told myself that it was not too late to run tangents better in the next miles and avoid covering something like 26.50 miles.

Brian caught up with me around Mile 5, after the Riverfront Walk ended and we returned to the streets of downtown Hartford.  We ran together for much of the next several miles.  I was doing a great job of keeping my pace within the range that I wanted through East Hartford.  It turns out that I ran by my hotel at around Mile 7, and I did not even notice; I didn't know until my post-marathon walk around East Hartford later that afternoon, when I suddenly recognized the road outside my hotel as part of the course.

Then came an noticeable slowdown.  I finished Mile 8, along a narrow path on the other side of the Connecticut River, at 7:16.  I had no idea why; this mile was not particularly difficult, but for some reason, I slowed down.  Mile 9 was better, but still a little slower.  Perhaps this was just one of those patches in the middle of the marathon where I just seemed to have lost a little spring in my step.  But my pace improved in the following miles.

Miles 10 through 17 (7:07, 7:01, 7:08, 7:09, 7:06, 7:09, 7:05, 7:07): Now I was on that out-and-back segment through South Windsor.  The crowds on the side of the course, as expected, were not nearly as large as the ones from Chicago last year, but there were still quite a few neighborhood people out heckling the runners throughout.  As I crossed an at-grade railroad tracks crossing, I kept praying that no trains would come by.  I really hoped no trains would come by on the way back; a train crossing would have been the last thing I wanted to have 23 miles into a marathon.

Fortunately, no trains crossed on the way out or on the way back.  I passed quite a few people on this stretch, but I found this stretch much more difficult than I thought it would be.  The course elevation profile through this segment was quite noticeably up-and-down, with some substantial uphill segments including all of Mile 17.  My pace through these miles were mostly between 7:05 and 7:10/mile; not quite the 7:00 to 7:05/mile that I really would have liked, but this pace actually felt like I was pushing it a little.  I worried that if I tried to push the pace even a little at this point, I would risk imploding later in the race.

I crossed the halfway mark at 1:34:28, slower than my halfway split for the Shamrock and Chicago Marathons.  I was not terribly concerned; I have been able to execute negative splits during most of my marathons, and I had faith that I would be able to do it again this time.  Besides, I figured that much of my slower first half was due to me not running the tangents well; by the time I reached the 13.1-mile mark, I had covered about 13.25 miles.  As mentioned before, the second half of the race had much fewer twists and turns; thus, I would probably be covering less distance in the second half, which would also really help for a negative split.

I crossed the 17-mile mark at around 2:02:00.  As I rounded the turnaround, I did some math; a PR was probably not going to happen today, but I had 63 minutes to cover the remaining 9.2 miles in order to finish in 3:05:00 or less.  My plan for the remainder of the race was to keep my pace under 7:00/mile and aim for about 6:45 to 6:50/mile or slightly faster.

Mile 18 to the end (6:54, 6:53, 6:57, 6:46, 6:52, 6:49, 6:57, 6:54, 7:00, 2:34 for the last 0.43 miles): Mile 18 was a very long downhill, so it was hard not to actually accelerate during this part of the race.  I completed this mile in 6:54, about ten seconds faster than the previous several miles.  I tried to gradually bring my pace down to around 6:40 to 6:45/mile during this segment.  But my quads were screaming at this point, and 6:50 to 7:00/mile was the best I can manage for most of these miles.

These few miles were a blur; I was focusing on reaching the finish line in one piece, and preferably under 3:05:00.  I passed many more people that passed me in the early miles, but not a single person passed me.  But otherwise, I did not notice too much of my surroundings.  I did remember a small dog at around Mile 22 that kept growling and barking at all the runners passing by.  I did also remember quite a few rolling hills that I faced on the way out that were much more noticeable now.  By the time I arrived at the 24-mile mark, I had very little left.  I just focused on maintaining my pace until the end, and saving some strength for the very difficult last mile.

After the Mile 25 mark, I found myself face-to-face with an uphill entrance ramp onto Founders' Bridge.  This last mile consisted of two large uphill segments: the first being a half-mile long one along Founder's Bridge and the second one being a shorter one along Pearl Street.  We did get a relatively substantial downhill segment in between as a break, as well as a downhill finish, but putting this at the end of the race was cruel.  I passed quite a few people during this mile, but I slowed down by a few seconds; I was giving it all I had left at this point.  I kept glancing at my Garmin, watching it tick past 3:05:00.  Jeremy and Brian's parents were waiting near the Mile 26 mark, and seeing them gave me the extra boost I needed to carry me through the finish.

Even though I didn't run a BQ-5:00 or better, I got pretty close; 3:05:43 (7:05/mile), which was a BQ-4:17.  I ran a three-minute negative split; whereas I finished the first half of the race in 1:34:28, I covered the second half in 1:31:15.  Even though I could have run the tangents a little better, which was not an easy feat on this particular course due to the twists and turns in the early miles (I covered 26.43 miles during this race according to my Garmin), I think this was the best I was able to do on race day.

I first went to the results tent to get my preliminary results and then enjoyed the very well-organized post-race festivities, including some vegetarian chili.  After about an hour or so, Jeremy and I took the shuttle back to the hotel.

Next Steps

With a BQ-4:17, I have a much better shot at actually getting to run Boston in 2017.  But I would feel even better if I had a BQ-5:00 or better and was able to register during the first week of registration.  I feel like I am definitely capable of a 3:04:59 or better right now, particularly on a less hilly and straighter course.

Thus, I am strongly considering running the Shamrock Marathon again in March; this race features a fast and flat course with much fewer turns than Hartford, although headwinds can be a problem in Virginia Beach in March, and is just as well executed as Hartford was.  In the meantime, I plan on doing some shorter distance races, including the Veteran's Day 10K in the beginning of November, the Annapolis Running Classic 10K in mid-November, and the Jingle All the Way 5K in the beginning of December.  

Race Post-Mortem: Navy-Air Force Half-Marathon

Pre-race selfie with JR.  Photo courtesy of JR.

Sunday, September 20, 2015
Washington, DC
Difficulty score: 3/10 (2 for weather, 1 for course profile, 0 for altitude)

After learning that the Rock and Roll Philadelphia Half-Marathon, my fall half-marathon of choice for the past few years, had been postponed until October 31 this year due to the Pope's visit, I decided to do the Navy-Air Force Half-Marathon instead.  Some members of DC Front Runners that ran it last year spoke very highly of it, from the course to the organization.  It was also about three weeks before the Hartford Marathon, which would make it perfect for not only getting me back into the racing mindset after five months of no racing, but also for determining a realistic target time for my upcoming marathon.

One problem, however, was that it was a mid-September race in Washington, DC; this meant that there was a good chance that it would be warm on race morning.  And indeed it was; 70 degrees at the start with a 62 degree dew point.  Nonetheless, I still didn't experience too much of a slowdown: I ended up finishing in 1:25:55 (6:33/mile).  Even though it was about 40 seconds slower than my PR at the Philadelphia Half-Marathon last fall, it still indicates that I am where I want to be in terms of fitness for the Hartford Marathon.  Plus, this half-marathon time will give me 71.43 points toward my score for this year's DC Front Runners Race Circuit (age-graded score of 68.03 for a 33-year old male according to the 2015 age grade standards, times 1.05 in accordance with DC Front Runners Race Circuit scoring rules that half-marathon scores receive a 5% bonus).  That's a solid start to the 2015-2016 Race Circuit.

Training

My training leading up to this race consisted of six days per week of running during most weeks.  The structure of each week was pretty uniform:

• Monday: 4-6 mile recovery run at 9:00/mile or slower to recover from the previous day's long run.
• Tuesday: my one day reserved for faster running for the week.  During the first few weeks, Tuesdays were easy runs at 7:40 to 8:20/mile as I was building my mileage back up.  After that, Tuesdays became speed workouts with intervals at current 5K pace (5:55 to 6:00/mile) that became progressively longer (960 meter repeats for two weeks, then 1200-1280 meter repeats for the next two weeks, and then 1600 meter repeats).  Then Tuesdays became 8-10 mile runs with 4-6 miles at tempo (10-miler to half-marathon effort).
• Wednesday: OFF
• Thursday: mini-long runs of 8-14 miles.  Most of these miles were at long run pace (7:40 to 8:20/mile), but I did throw in about 15 minutes of current marathon PR pace running (7:00 to 7:05/mile) into the end of these runs every other week or so.
• Friday: 4-5 mile recovery run at 9:00/mile or slower to recover from the previous day's mini-long run.
• Saturday: 6-8 miles easy (7:40 to 8:20/mile) plus hill sprints.
• Sunday: long runs of 12-21 miles, mostly at 7:40 to 8:20/mile.

I had actually completed the vast majority of my training plan as written, subject to some minor rearrangements to accommodate life and friends.  I did have to take half of the second week off and readjust everything else due to a pretty nasty summer cold, but that really did not have too much of an adverse effect on the rest of the training cycle.  Unfortunately, I also did neglect the Saturday hill sprints toward the later weeks; although I doubt that would come back to haunt me later, they could have really helped.

During this cycle, I hit similar mileage to that from my training in Chicago.  After my initial mileage buildup in June, my mileage was in the mid-fifties to mid-sixties most weeks, with a few reduced mileage weeks (40-45 miles) to recover.  My highest mileage week this time was 65 miles two weeks before the half-marathon, which immediately followed two back-to-back 63 mile weeks.

The Race

Even two weeks before the day of the half-marathon, accuweather.com was calling for non-ideal conditions on race morning; a low in the low sixties for the evening of Saturday, September 19, plus the usual humidity.  As usual, the week before the race, I was checking weather pretty obsessively.  I did want a half-marathon PR and to break 1:25:00, and I felt that under the right conditions, I had a good shot at both.  Although I have run half-marathon PRs in worse conditions (e.g. Frederick in 2012), accomplishing that in what was predicted would have still been tough.

Race day weather roulette finally settled on low(er) humidity (70-80% humidity) but temperatures in the mid-sixties during the days immediately preceding the half-marathon.  At that point, I accepted that there was a good chance I would not run a PR, and that I should simply use this as a diagnostic for my Hartford Marathon plan; if I was able to run somewhere between 1:25:00 and 1:26:00, that would bode well for Hartford.  Fortunately, the race started at seven in the morning, so I would begin at what should be the coolest part of the day and hopefully be finished before the temperatures really started to rise.  And besides, at least we would have lower humidity, unlike in Philadelphia last year.

Except that actual race morning conditions turned out quite a bit worse than what was predicted, and was probably even worse than the weather in Philadelphia last year.  I woke up at around 4:00 am, about an hour before my alarm was to go off, and checked the weather, noticing that it was 73 degrees outside.  I didn't fall back asleep because I was so cranky that conditions were going to be noticeably warmer than predicted; at this point, it was likely that the temperatures would not even go below 70 during the race (and actually, according to my uploaded race results at garminconnect.com, the temperature at the start was 70 degrees).

After my pre-race preparations, I took a taxi down to the start line near the Washington Monument after deciding against dealing with the Metro.  With just my keys in my back pocket and my trusty SpiBelt containing my license, some cash, and my debit card, I bypassed bag check and met up with a few members of DC Front Runners before lining up at the start.

The Navy-Air Force Half-Marathon featured a slightly new course this year, due to the construction on the Memorial Bridge that prevented us from running to and around the circle near the entrance of Arlington National Cemetery.  The course consisted largely of one segment along the perimeter of Hains Point followed by an out-and-back segment along Rock Creek Parkway (very similar to the Cherry Blossom 9.39-Miler that I ran this past spring, with an additional segment down Rock Creek Parkway added on).  In general, this was a relatively flat course.  Most of it was either along the Potomac or shaded, which I expected would prevent it from getting too warm during the race.

Map of the Navy-Air Force Half-Marathon course for 2015, from the website.

Miles 1 through 6 (6:38, 6:30, 6:32, 6:31, 6:35, 6:28): After a little pre-race chat with Dan and JR in the starting chute and a pre-race selfie, we were off.  After rounding the first right turn, I moved over to the left side as we were going counterclockwise around this initial loop around the upper part of the Tidal Basin.  After about 0.4 miles, I saw that I was running around 6:34.  Even though it did not feel particularly comfortable as I felt like I was pushing my turnover a bit, I decided to stay at this effort.

But from Mile 2 on, I began to settle into around a 6:30/mile pace without much more pushing and was able to keep this pace for the next nine miles or so.  These miles felt quite a bit more comfortable than the initial mile did.  During these few miles, I found that I was passing quite a few people.  I ended up running behind this girl named Jess for the next several miles.  I decided not to make any moves just yet but instead, to just focus on hammering out these 6:30 miles.

Miles 7 through 10 (6:31, 6:31, 6:29, 6:27): Now we were beginning the out-and-back segment along Rock Creek Parkway, which was wide, was somewhat curvy, and featured a few mild rolling hills.  I tried to hug each of the curves as I ran, but I failed and ended up covering more distance than I would have liked by stupidly going back and forth between the two sides of the road.  Meanwhile, I was still running behind Jess for this segment of the race.  She would break away from me on the downhills, but I would catch back up to her on the uphills.  On the way out, I continued holding my 6:30/mile pace.

The turnaround point was placed about a quarter of the way up a very large and steep hill leading up from Rock Creek Parkway up to Woodley Park.  I was thankful we would not run up the entire thing, but still, what we did run up was quite substantial, and I was sure not something anyone would have liked to see at Mile 9 of a half-marathon.  Right before the hill, I passed Matt from my dance troupe; I knew he was running as I caught a glimpse of him at the start, but I had no idea he was this far ahead.  Seeing him gave me an extra boost of adrenaline which helped me up the hill, particularly thanks to my competitive, Type-A side that refused to let him do better than me in this race.

Jess slowed down noticeably as we were both going up the hill.  I wanted to pass her, but then decided against it, realizing that we were not going that much slower and that I still had four more miles to go; not blowing all of what I had left on this hill was a good idea.  But after we reached the turnaround point, I took advantage of the downhill and broke away from her.  I began accelerating and she could not respond.

I took advantage of my momentum to start speeding up for a small late-race surge, but at the same time, I held back so that I would avoid an implosion during the last two miles.  On my way back down Rock Creek Parkway toward the Washington Monument, I got to see many other members of DC Front Runners.  Seeing your friends on the way back is one thing I like about out-and-back segments.

Miles 11 to the end (6:22, 6:24, 6:18, 1:37 for the last 0.27 miles): The last part of the race was flat and fast with very few curves and turns.  However, it was also much more exposed than the preceding miles.  I could feel the effects of the warmth; I wanted to go faster, but even 6:22/mile felt like it was close to the limit of what I had in me.  As I was running these last miles, I was doing math for what I had to run for the remaining portion of the race; a PR was not likely, but sub-1:26:00 was possible.

At this point, I was in a no-man's land of sorts.  I was passing people, but I was running by myself, and by the time I rounded the turn off Memorial Bridge and back on Independence Avenue, the next person was about one hundred meters in front of me.  But at least there were still plenty of people around so it didn't feel too lonely; in addition to the runners on their way north on Rock Creek Parkway, people were still heckling on the sidelines, including Grace and Angela at around Mile 11.

I did manage to speed up a little on the last mile.  The very last stretch of the race along 15th Street was a very noticeable downhill.  I took advantage of this downhill and pushed it to 6:08/mile pace.  As I was running, I kept looking at my Garmin, watching it tick past my PR with about 0.1 miles to go and hoping that I would cross the finish line before it read 1:26:00.

After crossing the finish line, collecting my medal and water, and watching JR finish, I met up with some of the other members of DC Front Runners and checked my result.  I did not hang around too long; after all, I had preparations for the DC Front Runners anniversary party to complete.

Thoughts and Next Steps


My main objective for Hartford is to not only qualify for the Boston Marathon, but also qualify by a large enough margin so that I will not have to worry about whether I actually can get entry through the standard registration process for 2017.  I would like to run Boston, especially after two years in a row of qualifying but not making the cutoff.  I'm most likely out for 2016; even though as of now, I have not yet received the official notice, I'm pretty sure that being one minute and twelve seconds below my qualifying standard will not be enough for me to make the cut given the increased number of registrants so far and the higher number of qualifiers in some of the larger marathons such as Chicago and Boston this past year.

Fortunately, I will be 35 on the day of the Boston Marathon in 2017, so the Male 35-39 qualifying standards now apply to me, even though I will only be 33 on the day of Hartford (I love my March birthday!)  That means I would qualify with a 3:10:00 now rather than a 3:05:00; given my half-marathon time and my mileage during this training cycle, I know I have a good chance of running Hartford in 3:04:59 or better, which would be great since it would allow me to submit my application to Boston 2017 during the first week of registration.  But at the same time, 3:04:59 is no longer a hard threshold like it was the past few years.  I now have some wiggle room; I could run Hartford in 3:07:00 or so (three minutes faster than my qualifying time) and still stand a very good chance of getting in.

So provided that the weather on race day cooperates, I would like to run 3:04:59 or better in order to be able to register during the first week of registration; based on registration numbers from recent years, this would give me a very good chance of being able to toe the line at Hopkinton in April 2017.  If I run a PR (3:03:47 or better), that would be even better.  Otherwise, if I am able to run a 3:07:00 or better, I would be fine with that; although I would have to wait until the second week of registration, that would likely still be a large enough margin for me to be accepted.

Thus, my race plan is to settle into around a 7:00 to 7:05/mile pace over the first three miles and then to stay very close to that pace for the first 30K of the race.  If I am feeling strong at the 30K mark, then I will pick it up slowly (I learned from Chicago to gradually accelerate rather than just to suddenly drop my pace by 15 seconds per mile; I'd be able to run a stronger late-race surge that way).  Otherwise, I will try to stay at that pace.  This plan will get me very close to 3:05:00, or may even possibly result in a PR, depending on how well my surge after the 30K mark goes.  Given my mileage, my time in the Navy-Air Force Half-Marathon, and my experience level, I should not have a problem running at a 7:00 to 7:05/mile pace (2.15 times 1:25:55 is 3:04:43, or 7:03/mile pace; the factor of 2.15 is based on my mileage and my experience at the marathon).  This race plan may even be somewhat conservative given that the half-marathon was run under non-ideal conditions.  But that's okay, since my primary goal really is to qualify for Boston by five minutes or more.

And I need to learn how to run tangents better.  Based on this race and the DC Half-Marathon this past spring, whenever I make a conscious effort to hit the tangents, I end up covering more distance.  Perhaps I should just not worry about running tangents so much?

A Tale of a 9.39-Miler and Two Friday Night 5K Races

DC Front Runners post-Cherry Blossom photo.  Photo by Socrates, pilfered from the DC Front Runners Instagram page.

Cherry Blossom 10-Miler
Sunday, April 12, 2015
Washington, DC
Difficulty score: 1/10 (0 for weather conditions, 1 for course profile, 0 for altitude)

Crystal City Friday 5K
Friday, April 17, 2015
Friday, April 24, 2015
Arlington, VA
Difficulty score: 4/10 (2 for weather conditions, 2 for course profile, 0 for altitude)

The Cherry Blossom 10-Miler was actually to be my first 10-mile race ever.  I have wanted to do one for a while now, but it was never in the cards.  In 2012, the Cherry Blossom 10-Miler was too soon after the USA Marathon, and by the time I got around to registering for the George Washington Parkway Classic 10-Miler, it was sold out.  In 2013 and 2014, I didn't feel like I was in good enough shape to race either the Cherry Blossom 10-Miler or the George Washington Parkway Classic 10-Miler, having recently returned from vacations to China and not having run at all while there.

But this year, I decided I was going to do the Cherry Blossom 10-Miler.  I was not running a marathon at all this spring, so I didn't have the problem of having to recover from one in time to run Cherry Blossom.  I was not going to China this year, so I could keep training after I ran the Rock and Roll DC Half-Marathon and come into Cherry Blossom ready to race.

I did indeed run, and had a great experience, but I still do not have a 10-miler PR.  Unfortunately, somewhere on the course around Independence Avenue and East Basin Drive, an accident occurred that left a pedestrian injured; that portion of the course had to be closed off as a crime scene and the entire race rerouted, resulting in a shortening of the race.  The website reported that the length of the course was 9.39 miles, and I covered that in 1:00:29 (6:26/mile), which would have projected to a 1:04:25 for the full 10 miles.  I am satisfied with that time; going into this race, I was hoping to run 1:04:59 or better.  Breaking 1:04:00 would have been the cherry on top.

I ran the Crystal City Friday 5K five days later.  Pacers holds this series of four Friday night 5K races each week of April every year; I had done one of these races in 2012.  The best way to describe my race was "interesting."  I had not raced a 5K in almost a year and a half; I had decided to do this one because I thought it would be good for me to do a shorter race from time to time.  I was still recovering from the 10-miler earlier that week.  I also was suffering from allergies, and the evening was warm and humid.  I still managed to run 18:53 (6:05/mile); this probably was my weakest race performance this year and was not even close to my best 5K time, but given the conditions, and the fact that I think I have forgotten how to race a 5K, I was not totally surprised.

Initially, that 5K was to be my last race until September, but then after noticing that I had completely recovered by the end of the weekend and that weather conditions were supposed to be much better the next Friday, I decided to run the next 5K race in the Crystal City Friday series.  The weather was indeed much better; when I arrived in Arlington, it was in the high fifties, sunny, and negligible humidity.  I ran twelve seconds faster, in 18:41 (6:01/mile).  This was much closer to where I hoped to be for a 5K, but still nowhere close to my PR.

Training and the Days Leading Up to Each Race

I really did not do much mileage in the weeks between the Rock and Roll DC Half-Marathon and the Cherry Blossom.  Part of it was because I was tired; after taking a few weeks to recover from the Chicago Marathon, I increased my mileage again for the Rehoboth Half-Marathon, and then, after a break lasting only a week or so, I started training for the DC Half.  Also, I realized that my fitness during the DC Half was essentially going to be my fitness during Cherry Blossom.

My mileage was actually substantially less than what I should have been doing for a 10-miler or a half-marathon.  I think I ran twice the week right after the DC Half, and then during the following weeks, I ran four days a week.  This included eight to nine-mile long runs on Sunday, six to seven miles easy on Saturday, and six-mile runs with three miles at target 10-miler pace on Tuesday and Thursday.  I estimated that 10-miler pace should have been around 6:25/mile; I took the projection from the McMillan Calculator based on my time from the DC Half and made it slightly more aggressive, in part because the Cherry Blossom course was much easier and in part because I was convinced I could do better than 1:27:19 in the half.  Also, I'm 33 years old and male, and I, like the vast majority of people of my age and gender, have been disproportionately faster in the shorter distances relative to my half-marathon and marathon performances.  Weather was predicted to be ideal on race day, which convinced me even more that being aggressive was the right move.

After Cherry Blossom, I really could not do any quality workouts to prepare for the first of the Crystal City Friday 5K races.  I still was in recovery on Tuesday, so I could not run faster than 9:30/mile pace.  I did not dare to run quickly or do any more than three miles on Thursday because the race was the next day.  I simply went into the 5K not knowing what to expect.

After doing the first Crystal City Friday 5K, like I had said before, I recovered relatively quickly; one advantage of a 5K race is that you can certainly recover within a few days and be ready to run another one the very next week.  In the middle of the week, I decided to run the next Crystal City Friday 5K in the series.  Again, I really did not do much thorough preparation; I only did a six-mile run the preceding Tuesday with the inner four miles at approximately 10-miler effort.  Besides, it was too close to race day for me to make any substantial gains in 5K race fitness.

Cherry Blossom 10-Miler

The night before the Cherry Blossom, I got about three hours of actual sleep.  I set my alarm for 4:55 am and went to bed at around 9:30 pm, but I spent much of that time lying awake in bed and tossing and turning.  That may have had something to do with pre-race excitement, but I also did fall asleep at 2:00 pm the day before and slept soundly until 5:00 pm; that definitely had something to do with my insomnia.

I woke up when my alarm went off at 4:55 in the morning.  I felt a little sleepy on my way to the start of the race, but I still felt like I could function.  I was a little concerned about how my lack of sleep would affect me during the race; when it comes to lack of sleep, I really am a delicate flower.

The weather was just as nice as predicted.  After arriving at the Washington Monument, I checked my bag, which was when I heard the news that the course was going to be shortened due to that accident.  I then ran into Socrates and we talked a little; we both had to go to our starting corrals as we had little time left.  I positioned myself about halfway between the front of the Yellow Corral and the start of the Red Corral, on the right hand side as the first turn was a right turn, and began thinking about how I would handle race circuit scoring for this race.  Thorne and Daniel joined me as I waited in the corral.

The Cherry Blossom 10-Miler course has very few difficulties.  As I mentioned before, it is very flat, aside from a few slight hills near the Memorial Bridge shortly after the one-mile mark and right at the end, after exiting the Hains Point loop and just before the finish.  The hairpin turns at around Mile 3 on Rock Creek Parkway near the Kennedy Center and at around Mile 4 on Ohio Drive could also be a little tricky; hairpin turns often cause me to lose my rhythm.

Cherry Blossom 10-Miler map, from the website.

Mile 1 (6:36): I had to do quite a bit of weaving during the initial first mile.  Looking back, I should have gotten as far onto the right side as I could, even in spite of the gentle left turns that followed that first right turn; I would have had a much easier time settling into my target pace.

The Yellow Corral was for people predicted to complete the race in under 1:12:24, and within the corral, I could not find any subdivisions indicating where we should line up given our target pace (e.g. signs indication where to line up if you want to run a 6:00/mile pace, where to line up if you want to run 6:15/mile, etc.)  I probably had encountered some people who were actually going to run 1:10:00 to 1:12:24 that had lined up in front of me; since I was going much faster, I had to do a lot of accelerating and decelerating and weaving in and out of people.

But after the first half of this mile, I was finally able to hit a good pace.

Miles 2 through 8 (splits: 6:26, 6:23, 6:26, 6:21, 6:25, 6:22): For the next several miles, my goal was to settle into about a 6:25/mile pace.  Every so often, I made sure that I was keeping my body relaxed, and that I was not pushing the pace or making any moves just yet.  Keeping my pace and effort level constant did take a bit of self-control, particularly when I saw Daniel ahead of me (sorry, Daniel; I was not about to let you beat me in this race).  I also tried to resist the temptation to glance at my Garmin too often; during the days preceding this race, I made the realization that my best races were ones where I just ran at the pace my body told me to run and let whatever pace I ended up running be the pace for that day.  After all, checking your pace too often can cause you to panic that you are not running fast enough, and freaking out takes energy.

I was very happy to see that many of the DC Front Runners that were not running the race were on the side of the course at various locations spectating and heckling.  As I turned onto Memorial Bridge after the Mile 1 marker, I heard Ricky and Dave call out my name.  At around Mile 5, I encountered Chris, JR, and Lavar.  And then shortly afterwards, I saw Brian on Ohio Drive on the way to Hains Point.  That is one thing I like about local races, particularly the larger ones such as this; you get to see many familiar faces on the side.

My pace remained quite constant through most of these miles.  I did end up passing Daniel at around 6.5 miles into the race.  Shortly after that, I started to push the pace just a little.  But at the same time, I held back so I would not surge too much too soon.

Mile 8 to the end (splits: 6:22, 6:12, 2:58 for the last 0.48 miles): I kept at around a 6:22/mile pace for another mile, deciding not to accelerate just yet; I was not sure if I had three miles (or, rather 2.48 miles) at sub-6:20/mile pace at this point.  Once I passed the Mile 8 marker, indicating that I only had two more miles left, I finally went into Beast Mode and started pushing the pace to 6:12/mile.

I ended up passing quite a few people during this portion of the race, but I did not remember anyone passing me.  But I won't lie; these last two miles hurt.  I focused on keeping as relaxed as I could and making it to the finish.

Seeing Jeremy, Joel, Nate, and JR at that turn back onto East Basin Drive gave me a boost of energy; I threw in a surge as I ran past them.  I maintained my 6:11/mile pace down East Basin Drive, up that last hill right before the finish, and then down 15th Street to the finish.

I waited for the other members of DC Front Runners running this race near the bag check area, after which we all invaded the Starbucks on 14th Street and New York Avenue in accordance with our post-Cherry Blossom tradition.  After a few hours there, I went home and tried, unsuccessfully, to take a nap.

April 17 Crystal City Friday 5K

When I stepped out of my apartment on Friday to head over to the race, I was taken by surprise at how warm it was.  When I checked my phone, I saw it was 76 degrees.  I had been teleworking that day and did not leave my apartment until it was time for me to go down to Arlington.  I had my windows open, though, and it didn't feel particularly warm.  When I arrived in Arlington, I could see some relatively ominous-looking rain clouds near us; the impending rain storm caused Arlington to be very humid.  Even after my warm-up, I felt like sweat was dripping down my body.

The Crystal City 5K route had changed slightly since I last did it in 2012.  We still started and finished in the driveway of this office complex.  After the start, we would do this loop down Crystal Drive and then down this service road behind these office complexes; in 2012, we also ran down this service road, but I remember there being a hairpin turn somewhere in that area.  The subsequent segment north on Crystal Drive was still the same.  After winding through some side streets, we would then come to a hairpin turn at the Mile 2 mark, which definitely was not there in 2012.  Finally, we would come back down Crystal Drive for the finish.

Crystal City Fridays 5K map, from the website.

Since this was a circuit race, several members of DC Front Runners were running also.  Harry was running this race as well; we had trained for the Richmond and Baltimore Marathons together, but this was the first race in a long time in which both of us were running.

Pre-race photo with the long lost Harry.  Photo by Jeremy.

Miles 1 and 2 (splits: 5:54, 5:50): I spent the first mile doing a bit of weaving and not running the tangents well, which was probably the reason why I ended up running 3.16 miles instead of closer to 3.1.  I was considerably faster than some of the people who had lined up in front of me at the start.  Also, the route was quite twisty in the beginning; no matter where we lined up, we would be running the tangents for half the turns and not for the others.

But once I escaped this twisty part and began my trek up that service road behind the office complexes, I started to settle into a pace.  It felt manageable; I even felt like I was holding back a little.  When I looked down at my watch near the Mile 1 mark, I noticed I was running at a 5:54/mile pace.

JR, Jeremy, Alfie, and Andy were near the Mile 1 marker, cheering on the runners.  I was also pleasantly surprised to see Thorne on the side of the course shortly afterwards; I had no idea he was coming down to watch.  Lavar was not much further, taking pictures of us as we ran by.  I had so many friends watching me run today.

The second mile was even faster and felt even easier.  The segment down Crystal Drive had no twists and turns like the first mile, and was slightly downhill.  I dropped this woman and man with whom I ran most of the first mile, and caught several people ahead of me.  Some other people passed me, but I didn't care; I felt I was still keeping a much faster pace than I predicted I would.  As I approached the Mile 2 marker, I looked at my watch and noticed that I was now running closer to 5:50/mile pace.  This still felt manageable; as I ran, I began thinking that I might actually run a 5K PR today.  However, once I hit that hairpin turn, that changed.

Mile 3 to the end (splits: 6:11, 0:57 for the last 0.16 miles): The third mile did not feel nearly as good as the first two; I felt like I was working, yet I could not get my pace down to sub-6:00/mile.  I did have trouble getting back on pace after rounding that hairpin turn, but I also think that the warmth and the humidity were catching up to me at this point.  Plus, much of this part was a slight uphill.

I just kept running at the same effort to the end.  A few people passed me and I just did not have the energy any longer to respond.  I ran past JR, Jeremy, Andy, Alfie, Thorne, and Lavar again and threw in a little sprint as I turned onto the driveway and toward the finish line.  

Me struggling during the last stretch of the 5K.  Photo by JR.

I got some water, walked around a little, and watched John finish his race, as well as get a PR and achieve his goal of breaking 20:00.  I then returned to where JR, Jeremy, Andy, Alfie, and Thorne were standing to watch Harry and Matt finish their races.

April 24 Crystal City Friday 5K

As mentioned before, the temperature was much cooler and the humidity was much lower.  We were also blessed with abundant sunshine and a cloudless sky.  The weather would have been perfect had it not been for those headwinds.

The course was exactly the same as the preceding week's Crystal City Friday 5K, so I came in with a better idea of what to expect and where the difficult parts lay.  JR was also running this time, and again, we had our cheering squad of Jeremy, Andy, Alfie, and Eric.

Mile 1 (6:08): I did a much better job of running tangents through that initial twisty part of the race this time, and I managed to do it without cutting anyone off.  As I turned onto the service road behind the office complexes, I glanced down at my Garmin and noticed I was running a 7:30/mile pace.  That was definitely not right; looking back, what I think had happened was that the GPS still had trouble resolving even after I started the timer.  This sometimes happens on runs; the GPS is unable to find signal during the first thirty seconds or so, which makes the timer continue even though my Garmin is not recording any distance.  Essentially, thirty seconds pass before my Garmin detects the first hundredth of a mile that I run, ultimately resulting in an unusually slow first mile.

But I knew by feel that I was definitely not running 7:30/mile pace.  I focused on running the tangents and not freaking out; I kept my effort at a constantly hard level, but not too hard that I would implode on the much more difficult third mile.  On the service road, I ended up passing this guy with whom I would be locked in a struggle until the very end of the race.

My Garmin read 0.96 miles and 5:54 when I crossed the first mile marker.  This also gave me reason to believe that my Garmin had a signal issue at the very beginning of the race.

Mile 2 (5:53): I kept my pace controlled throughout this section, remembering from last time that this mile was much faster than the other two.  However, headwinds began to strike.  Remembering my experience from the Shamrock Marathon, I tried not to charge into the headwind.  Even in spite of the headwind, I was able to keep my target pace.

Aforementioned guy that I passed in the first mile on that service road passed me right back halfway through this mile.

Mile 3 to the end (splits: 6:02, 0:38 for the last 0.11 miles): Just like last time, the third mile was the most difficult for me.  Again, I felt like I was working harder, even though my Garmin indicated I was running slower.  The headwinds certainly did not help; I still encountered them, even though I was going in the opposite direction of what I was running in the second mile.  But I did pace the third mile much better than I did last time; 6:02 versus 6:11.

At the very end of the race, I did end up passing that guy that I passed in the first mile who passed me back in the second mile at the very end.  From this mile until the end of the race, I also had my sights set on JR; he was several seconds ahead of me, and I tried to catch up to him, but I could not close the gap (I am truly happy that he got a PR in this race; I'm just a little competitive).

During dinner, every few minutes I would have these chesty coughs where I would hack up slime from inside my lungs.  It all cleared up shortly afterwards.  I was not concerned; this is typical after I run a shorter race.  I've always believed that if you're hacking up a lung after a 5K race, then you did it right.

Next Steps

I have forgotten how much 5K races hurt.  They are so short and fast that it is pretty much a lung-searing effort the entire way.  It isn't an effort that is strenuous enough to have you tasting that coppery lactic acid taste afterwards like an 800 meter or 1600 meter race would, but it will still leave that burning sensation in your lungs for hours afterwards.  Also, due to their length and speed, they leave very little room for error.

I'm finished racing until the Navy Half-Marathon in September.  I don't really like racing during the summer, not only because the heat and humidity leads to slower finishing times, but also because I have been training for fall marathons every year these past several years, and I typically prefer to spend my time and effort for training.  Doing the shorter races that are typically held in the area in July and August still require at least a few days of recovery, but are not as helpful to my marathon training as long runs or high mileage would be.

This also marks the end of my 2014-15 DC Front Runners race circuit season.  Spring 2015 was not exactly a season of jaw-dropping personal bests or breakthroughs, although I did manage some perfectly satisfactory performances.  But I have been a little lazy this winter; my mileage has been lower.  But in general, this race circuit season was a good one; my total score (431.00) was similar to last year's (430.24), I did get a half-marathon and marathon PR, qualified for Boston again (although by 72 seconds, which still probably would not be enough to allow me to run Boston in 2016), and won an age group award.

Now, I plan to take a bit of a break from running.  I will still run, though, but just the bare minimum to maintain some fitness so I don't go into training for my next marathon completely out of shape.  I will be doing a lot of dancing for the DC Front Runners Dance Troupe, which was excellent cross-training, did help maintain fitness also, and exercised muscles I usually don't work when I'm doing heavier mileage.  But I did like having a season of less running and no marathons.  Come June, I will be ready to start training for Hartford.  I'll aim for a PR, and hopefully, I will run a fast enough time to allow me to run Boston in 2017.

Race Post-Mortem: Pi Day Half-Marathon

Pre-race picture of me, Thorne, and JR waiting in the rain in our corral.  Photo by JR.

Saturday, March 14, 2015
Washington, DC
Difficulty score: 3/10 (0 for weather, 3 for course profile, 0 for altitude)

No, there is no Pi Day Half-Marathon.  This was actually the Rock and Roll DC Half-Marathon (formerly the Rock and Roll USA Marathon, which before that was the Suntrust National Marathon), which just happened to take place on 3/14/15.

For Spring 2015, I decided I was going to break the cycle of fall and spring marathons that my racing scheduled had been centered upon for the past three years or so and devote a few months to focus on half-marathons, 10-milers, and perhaps some shorter distances.  I had planned a series of shorter races for the spring, including the Rock and Roll DC Half-Marathon, the Cherry Blossom 10-Miler in mid-April, and possibly one of the Crystal City Friday Night 5K races shortly after that.

I finished in 1:27:19 (6:40/mile), about two minutes slower than my PR at the Rock and Roll Philadelphia Half-Marathon this past September.  This was also the "longest" half-marathon that I had ever run; when I crossed the finish line, my Garmin read 13.29 miles.  Usually, I am pretty good about covering close to 13.1 miles in a half-marathon race; for other half-marathons, my Garmin reads anywhere from my record "shortest" half-marathon of 13.12 miles in the Colonial Williamsburg Half-Marathon to about 13.19 miles.

Oh well; I guess not every race can be a PR.  I did have a number of factors working against me this time around.  My training did not go as well as I would have hoped, nor was my mileage nearly as high as it was preceding my half-marathon PR.  Even though I did have a string of four or five weeks with mileage in the mid to high forties, I did have a few short interruptions to my training: once at the beginning of the cycle due to some bad hamstring soreness and twice due to sickness.  Also, the DC Half-Marathon course is tricky, and certainly more difficult than the Philadelphia Half-Marathon course; in addition to a very nasty hill near the 10K mark leading up to Woodley Park from Rock Creek Park, other tough segments featured in this course include a long uphill from Woodley Park to Columbia Heights and the rolling hills near the Mile 12 mark.

But at least I did get a good amount of points toward the DC Front Runners race circuit with this race.  A 1:27:19 half-marathon for me would translate into 71.22 points (that half-marathon time from a 32-year-old male gives an age grade score of 67.83; the extra 3.39 points comes from the 5% bonus added to scores for half-marathons in accordance with DC Front Runners race circuit scoring rules).  In general, I'm happy with any score above 70.

Training and the Days Leading Up to the Race

When I sat down to plan my training for this race back in December 2014, I set a goal range of 1:24:30 to 1:24:59 (6:27 to 6:29/mile).  This would have been slightly faster than my half-marathon PR from the Rock and Roll Philadelphia Half-Marathon this past September, and I thought a small improvement like that would have been reasonable.  My mileage was not going to be nearly as high as it was when I was training for the Chicago Marathon, but I figured that would not be a problem; after all, I was training for a half-marathon right now.

Also, I planned to include a little more speedwork, which I pretty much completely neglected while training for Chicago.  During marathon training this past summer, I sacrificed some of my quality workouts to be able to increase my weekly mileage to figures quite a bit higher than I had ever done.  The increase in mileage certainly did help; I got a 59-second PR in Chicago and ran a four-minute negative split.  But including a little more speedwork probably would have led to an even greater improvement.  Three weeks later, when I ran the Veteran's Day 10K here in Washington, DC to capitalize upon my fitness from marathon training, I found I had lost quite a bit of speed.  I finished in 38:55, 45 seconds slower than the 10K PR I ran just seven months before following the Shamrock Marathon (this never happened before; usually, when I run a shorter race after a marathon, I get quite a few PRs, or at least very close to my PRs).  So this time, I included weekly hill sprints and some add-on 300-meter intervals, following suggestions from Brad Hudson's and Matt Fitzgerald's book Run Faster from the 5K to the Marathon, hoping that the bit of additional speedwork would help during the half-marathon.

The weekly structure of my training for this half-marathon consisted of five days of running per week at the beginning, moving toward six days per week in the middle of the cycle as my mileage increased.  Each week looked like this:

• Monday: easy recovery runs of four to five miles.
• Tuesday: half-mile intervals done at 3K pace (5:40/mile) at the very beginning of the cycle; in the middle of the cycle, these would be replaced by six to seven mile runs containing tempo portions at 6:25 to 6:30/mile pace, with four 300-meter intervals done at 3K pace added on every other week.  
• Wednesday: OFF
• Thursday: seven to eight-mile runs up the very hilly segment on Massachusetts Avenue Northwest toward American University, in which I progressed to tempo pace for the last few miles.  
• Friday: OFF in the early weeks of the cycle, easy recovery runs of four to five miles later.
• Saturday: seven to nine-mile easy runs plus hill sprints, going from four by six seconds at maximal effort uphill to six by eight seconds.
• Sunday: long runs of between 10 and 14 miles.  The ones toward the end of the cycle included segments at target race pace.

With this training plan, I would have a string of weeks with mileage in the mid to high forties, with two 49-mile weeks in the middle of February.

However, as mentioned before, training did not go as smoothly as I would have liked.  I was dealing with a very sore hamstring during the first few weeks of the cycle, which prevented me from doing the Tuesday half-mile intervals.  I caught two pretty nasty colds; one in the middle of January and toward the end of February, both of which interrupted my training and forced me to forego a few of my Tuesday evening tempo runs as I recovered from sickness.  The snow and ice we got here in Washington, DC made some of the tempo runs and the add-on intervals infeasible, forcing me to replace them with slower runs.  

But I did, like I also mentioned before, get a good string of weeks where I was able to execute my training plan as originally written, and fortunately, these coincided with the highest mileage weeks.  Thus, I was able to get several weeks with mileage in the mid to high-forties, just like I had planned. 

The Race

Race day morning was cold and wet, with temperatures remaining in the forties or low fifties and a steady light rain lasting for the entire morning.  But I found that the rain did not really affect my performance, provided I avoided running through puddles; my shoes were like a sponge, and I really did not want to carry any additional water weight.

When I arrived at the start line, I immediately stripped off the jacket I was wearing and went directly to bag check.  Once again, bag check was arranged by last name in alphabetical order, which is one of the most thoroughly stupid race day logistics decisions I have ever encountered.  They had done this when I ran the Rock and Roll DC Marathon in March 2013, and this not only caused a number of people in the earlier corrals to miss their start, but also resulted in some misplaced and lost bags.  I had always asserted that arranging by bib numbers is the logical way to go for bag check.  This way, at the beginning, more volunteers could direct their attention to the lower-number bibs (i.e. the earlier corrals), making sure they get to the start line on time.  Then afterwards, they could gradually move more volunteers toward the higher-number bibs.  Also, ascertaining a runner's bib number takes a fraction of the time that it takes to ascertain their last name; all the volunteer needs to do is look at the runner's bib, rather than ask "what's your last name?"  Then there is the tendency of a surprisingly large number of people to mix up the second and third letters of my last name for some reason.  Then they would tell me I'm at the wrong truck when I'm really not; that's another bit of time spent dealing with bag check and holding up the line.

But regardless of my ranting, bag check was not a complete disaster.  I was able to drop off my bag within three minutes.  Of course, given that I had very little faith in bag check at this race, one logical thing to do could have been not to check any bags; but given the rain, a change of clothes for after the race and an umbrella would have been really nice.  Besides, I wanted to see my friend Socrates finish his first marathon, and I probably would not have been able to go home and change between when I finished and when he would pass the 26-mile marker where I planned to see him.  I was also very glad that bag check did not lose my bag.

As I mentioned before, the Rock and Roll DC Half-Marathon course is quite difficult.  The first six miles or so, in which runners go east on Independence Avenue, out and back along Memorial Bridge, and then north on Rock Creek Parkway are relatively flat and fast.  The next three miles are difficult; at around the Mile 6 mark comes the Shoreham Drive hill connecting Rock Creek Parkway to Calvert Avenue, which is short but steep.  But even though this hill is what most people bring up when talking about the difficulty of this course, the more difficult segment is yet to come; the next two miles through Adams Morgan, Columbia Heights, and Howard University feature quite a few uphills and downhills.  Once runners reach North Capitol Avenue, the course does allow for faster running again, but it does become a little tricky again once we turn onto K Street; the course features several turns and uphill segments.  The last mile, fortunately, is mostly downhill.

Map of the Rock and Roll DC 2015 course, from the website.

The start through Mile 6 (splits: 6:40, 6:33, 6:35, 6:38, 6:38, 6:40): After a slightly slower first mile to transition into my target pace, I wanted to aim for about a 6:30 to 6:35/mile pace during these first six miles.  I managed to run in that range for the second and third miles, which covered Independence Avenue and Memorial Bridge, but once I turned off Memorial Bridge onto Rock Creek Parkway, I found that keeping that pace was actually a little difficult.  As usual, I focused on hitting the tangents and not wasting energy weaving in and out of people.

For the first half of the sixth mile, I was able to drop back down closer to 6:30/mile.  But when I arrived at the hill, my pace slowed again, as expected.  I tried to relax and just make it to the top.  I was able to pass several people on this hill, but I did not focus too much on that.  I did not want to expend too much energy on this hill; after all, I still had seven more miles to run.

Miles 7 through 9 (splits: 7:02, 6:40, 6:34): After finally reaching the top of the hill, I ran down the Calvert Street bridge toward Adams Morgan, through a large inflatable plastic blue arch that seemed to serve no purpose whatsoever.  Mile 7 is, in my opinion, easily the most difficult mile of the half-marathon.  In addition to the fact that it starts on that nasty uphill, runners have to face two more hills in this mile: one on Calvert Street right before the large intersection with Columbia Road and one on Columbia Road right before entering Columbia Heights (and right past where I live).  Neither of these hills are super long or steep, but they are enough to slow most runners down quite noticeably.  What also makes this part of the race difficult is that here, runners would need to find their rhythm again after that hill up from Rock Creek Park; that was what I spent much of this mile doing myself.

Mile 8, which features a long downhill stretch followed by a relatively substantial uphill segment, is less difficult, but still not a particularly fast mile either.  I was able to get closer to my original pace during this mile.  By the time I was running through Howard University at Mile 9, I was right back to where I was before that hill.  But during this portion, I began to encounter some slight headwinds, which were somewhat annoying, but not debilitating.

Miles 10 through the end (splits: 6:24, 6:20, 6:27, 6:23, 1:44 for the last 0.29 miles): North Capitol Street is a much faster segment than any part of the race that preceded it; even though the course does have a few uphills due to the underpasses, most of this part is straight and gradually downhill.  I took advantage of that and sped up once I turned onto North Capitol Street for a bit of a late-race kick.

I had managed to bring my pace down to 6:20 to 6:25/mile here, which was pretty uncomfortable.  During these miles, I was focused on just maintaining this pace until the end of the race.  But I did notice a few familiar faces on the sides during this portion, including the DC Triathletes spinning on the side of the course who called out my name as I passed, and Chris from DC Front Runners.  I did also notice that some people that I passed back near Howard University or Columbia Heights were now flying past me.

I had forgotten about the tricky parts of Mile 12 until I was staring at an uphill portion on 13th Street Northeast leading up to Constitution Avenue.  I was really pushing, and an uphill at this point was not something I wanted to see.  After reaching the top of this hill, I began the gradual descent toward the finish line at RFK Stadium.  Shortly afterwards, the marathoners, after having run with the half-marathoners for the past 20K or so, split off.  As I watched some of the marathoners turn right toward the more desolate and less aesthetically pleasing second half, I began to feel glad that I was not them.  At this point, I had barely enough to get me to the finish line.

I kept my pace steady for the thirteenth mile, although looking back, I could have been more aggressive, particularly since it was almost entirely downhill.  But I did put in one last surge once my Garmin indicated I had run 13 miles.  I gave it all I had and tried to break 1:27:00, but unfortunately, there was too much distance between myself and the finish line.  After finishing, I waited for Thorne, JR, and other members of DC Front Runners who were also running the half-marathon to finish before claiming my bag from bag check and changing into a dry shirt and dry socks and shoes under an overpass.

Remember, kids: let this be a lesson to you to always try to run the tangents.  The difference between running 13.29 miles and 13.15 miles can be a minute or more.  In this race, I finished that last 0.29 miles in one minute and 44 seconds.  Compare that to the last 0.17 miles in the Rehoboth Half-Marathon and the last 0.15 miles in the Rock and Roll Philadelphia Half-Marathon, both of which I ran at a similar effort and covered in 58 seconds and 51 seconds respectively.  Had I hit the tangents better and ran 13.15 miles rather than 13.29, I probably could have finished closer to 1:26:00.

Next Steps

My next race will be the Cherry Blossom 10-Miler on April 12, which will actually be my first 10-miler ever.  I'm not sure what to expect during this race; I'm currently not in as good shape as I hoped I would be, and with less than three weeks to go, whatever shape I am in right now is essentially the shape I will be in on race day.  But I do think I can still manage a pretty good finishing time; I hear from many people who have run that race that the course is fast, flat, and very conducive to a PR (or in my case, a strong 10-miler debut).

I am also considering doing the Crystal City Friday Night 5K five days after the Cherry Blossom.  For this race, I'm even less sure what to expect.  I have not raced a 5K in over a year, and I have not done any training for a 5K.  I suppose I could expect to finish in under 19 minutes; under 18:30 would be fantastic.  But after this 5K race, I would be done with racing until fall.

Gargoyles, Opus 29 (Liebermann)

Gargoyles is a set of four short, contrasting character pieces written by American pianist, conductor, and composer Lowell Liebermann.  Gargoyles was composed in 1989, when the composer was 28 years old, and is actually the first twentieth-century piece post-dating Impressionism that I have done in a long time.  Numerous sources indicate that the four movements of Gargoyles were meant to evoke imagery associated with the carved stone figures, but none of the movements refer to a specific gargoyle.

What exactly each of the movements depict is largely left to the listener's imagination.  For example, the first movement (0:00 to 2:02) conjures up images of more demonic-looking gargoyles; marked Presto (fast), this movement is a scherzo whose mischievous and sarcastic character is achieved through the irregularity and unpredictability of the music.  The movement is full of dramatic and sometimes schizophrenic dynamic changes, including the sudden loud dissonant chordal outbursts throughout the middle section and the long crescendos.  Liebermann also employs hemiola to add rhythmic unpredictability, switching from the standard 3 + 3 partition of the 6/8 time signature to a 2 + 2 + 2 division.  The polyrhythms in the beginning and at the end also contribute to the irregularity of the rhythm.  The unresolved dissonances everywhere and the lack of familiar major and minor chords also gives this movement its demonic character; out of the four movements, this one is the most dissonant.

Meanwhile, the second movement (2:08 to 4:41), marked Adagio semplice (slow and simply played), is a pensive aria that could depict a weathered, time-worn gargoyle, like one fixed to the ruins of a church.  Here, Lieberman uses stagnant harmonies, sparse textures, and dynamic markings that never rise above mezzo-piano to achieve an atmosphere of desolation and loneliness.  The beginning and end of the movement feature a chromatic melody in the right hand, played in bare octaves, over a plodding accompaniment in the left, whereas the contrasting middle section features a chorale over slow repeated notes.  The music is also entirely restricted to the higher registers of the piano, giving the movement not only a lonely character, but a haunted feel as well.

The third movement (4:50 to 7:24), marked Allegro moderato (moderately fast), is a reverie that seems to evoke imagery of water; after all, many gargoyles serve not only as decorations, but also water spouts to drain rainwater from the roofs and walls of buildings.  The water-like feel of the music is achieved through a continuous harp-like arpeggio accompaniment, which, throughout most of the piece, is to be shared between the two hands, which makes for a challenge in maintaining evenness.  Most of this movement consists of a simple theme in F-sharp major with some chromatic inflections in the soprano over this arpeggio accompaniment, with the addition of a bass melody that imitates the soprano toward the end.

Finally, Liebermann returns to a depiction of the demonic in the last movement (7:32 to 10:31), marked Presto furioso (fast and furious).  This movement, which most pianists would agree is the most technically demanding of the four by far, takes the form of a very energetic dance.  The unrelenting triplet rhythm, abundance of accented notes, extended passages in the lower registers of the piano, and the dense chordal passages gives this movement a savage character also.
  

Gargoyles is perhaps the most frequently performed work for solo piano piece of Liebermann's, and has been becoming a favorite among pianists and audiences; threads on piano-related message boards requesting recommendations for modern piano pieces often get Gargoyles as an answer.  Its outwardly virtuosic writing certainly could be a reason for its popularity; technical challenges in Gargoyles include rapid scales, arpeggios, broken octaves, dense chords, and leaps, which not only give the pianist an opportunity to show off a diverse set of techniques, but also are almost certain to excite audiences.  But more notably, Gargoyles is a twentieth-century piano piece with Romantic Era characteristics.  The twentieth-century characteristics of Gargoyles, particularly its use of atonality in the first, second, and fourth movements (none of these movements can be reliably said to have a central key) and its avoidance of more traditional harmonic progressions of the Classical and Romantic Eras, including unresolved dissonances in the first movement and the use of exotic scales in the first and fourth movements introduce some variety and novelty, but the resemblance of Gargoyles to character pieces and etudes of the nineteenth century prevent the piece from becoming so foreign that it would be unenjoyable to most pianists and audience members.

Each of the four movements of Gargoyles have structures similar to those of piano pieces from the Romantic Era.  The first, second, and fourth movement are all in ternary (ABA) form, namely an exposition of the main theme (A), followed by a central section in which either a contrasting theme is presented or the main theme is developed (B), and then a reprise of the main theme (A) with a coda.  The third movement is in rondo (ABACA) form, consisting of three statements of the main theme (A) with two intervening interludes (B and C).  As a result, these pieces are easier to follow than most others composed around the same time, which often tended to be more loosely structured.  Dissonances, although prevalent in all four movements, serve functions such as for shock purposes in the first movements and to build tension in the second, third, and fourth movements, similar to those they serve in the pieces of eighteenth and nineteenth-century composers.  In the third and fourth movements, dissonances even resolve into familiar major and minor chords.  Thus, these dissonances, although sometimes jarring and even harsh at times, are placed in contexts where they make sense to most listeners, which would tend to make the audience more receptive to these harmonies.

The video above is of my performance of the complete set at the Washington Music Teacher's Association (WMTA) Annual Adult Recital on January 17, 2015 (thanks to Thorne for taking the video.)  Somehow, I managed to accurately execute the runs in the first movement cold; I did not get much of a chance to warm up prior to the recital, and I was fifth in the lineup, which meant I had to wait another forty minutes or so before I finally performed (do not try this at home, kids.  Always precede a performance of this piece with a proper warm-up or a performance of a more sedate piece.  I'm just a little crazy and have a tendency to do dumb things like this.)

Overall difficulty assessment: Difficult.

First movement (Presto):

• Technical difficulty assessment: Difficult.  Technical problems include rapid scale passages, alternating martellato chords, and irregular polyrhythms.
• Musical and interpretive difficulty assessment: Straightforward.  This is not a subtle movement.  Playing the notes and following the dynamic markings will result in an effective performance, but pianists should be sure to play the soft passages very softly and hammer the louder chords to make the contrasts more stark and the crescendos more dramatic.

Second movement (Adagio semplice):

• Technical difficulty assessment: Easy.  This movement contains no substantial technical difficulties.
• Musical and interpretive difficulty assessment: Tricky.  Attention to phrasing and dynamic detail and a sensitive touch is needed to create the atmosphere that Liebermann wants from this movement.  Any carelessness in phrasing, touch, and dynamics will be exposed due to the sparse textures and lack of technical fireworks.

Third movement (Allegro moderato):

• Technical difficulty assessment: Difficult.  Technical problems include maintaining evenness in the arpeggiated accompaniment shared by both hands, balance of sound between the melody and the accompaniment, and voicing of the canon in the recapitulation.
• Musical and interpretive difficulty assessment: Tricky.  In addition to difficulties in voicing and balance of sound, the pianist must put some thought into how to prevent this movement from becoming monotonous.

Fourth movement (Presto furioso):

• Technical difficulty assessment: Very difficult.  This movement contains alternating martellato chords, leaps, dense chordal passages, and rapid repeated octaves, all of which are to be executed at a very fast tempo.  This movement also requires endurance.
• Musical and interpretive difficulty assessment: Straightforward.  Like the first movement, this is not a subtle piece of music.  Here also, playing the notes and following the dynamic markings will capture the spirit of this movement and result in an effective performance, but performers should be sure to really capitalize on the softer passages for make the contrast with the louder passages more pronounced and the crescendos more dramatic.