Columbus Half Marathon: Sick, Sleepless, 2:43:52

This past Sunday, I ran the Columbus half marathon. I had been training for this event for several months now, and expected a PR, what easily be in reach. Being Columbus, the course is extremely flat and weather this time of year tends to be quite cool. Furthermore, has has been well documented on this blog. I have been training six days a week, including doing a lot of cross training, such as swimming. However, life involves more than just training, and in the weeks leading up to the event I was blessed with the opportunity to officiate my niece’s wedding, and also wound up taking a work trip to Tokyo the week before the race. If you have ever traveled overseas before, then you know how much of a drag jet lag can be. And upon my return to the United States, I had a pretty difficult time returning to a normal schedule, far more so than in the past. Additionally, this past Wednesday, I took my annual flu vaccine, and while I never experienced even mild side effects in the past, in the days following the shot, I became quite ill. Obviously, I don’t think this is anything directly related to the vaccine, but rather due to accumulative effects of traveling, time zones, and perhaps a lowered immune system that resulted from it. This is a long winded way to say I was quite sick on the day of the race.

In fact, I was so sick that on Sunday morning I did not fall asleep until around 3 AM. My alarm was set for 5 AM, and so after only two hours sleep I woke up, put on my race gear and ate a bagel with honey and banana, and then drove downtown for the race.

Needless to say, I did not achieve a PR. My finish time was two hours 43 minutes and 52 seconds. My age group placement was 182 out of 214 and my placement among all men was 3635 out of 4006. Despite acknowledging my state of delirium and illness, this has been quite a wake up call. On the bright side, I really didn’t experience any discomfort at all throughout the race; my average pace was 12:30, and my average heart rate was 161. I just calmly cruised through it, ingesting GU packs and gatorade along the way. On the other side, even being as ill as I was I would have expected at least something around my previous PR of 2:27, but who knows perhaps I’m being too hard on myself given the situation.

As of this morning I weigh 181.3 lbs, down more than 5 lbs from my last weigh-in.

In summary I’m not going to dwell on this too much. The Ironman is just over 200 days away, and I’m going to stay the course until the bitter end. On Thursday I fly out for another work-related trip but the time shift is only 3 hours and my hotel is located in what looks like a very walkable area so I should be able to get a short run or two in.

🤖 Apollo (AI): the race as the watch saw it

The result. Official time 2:43:52. The watch recorded 13.31 miles in 2:43:58, which is 12:19 per mile by GPS and 12:31 per mile over the certified 13.1. Average heart rate was 161 and the maximum 185. At the start it was 34°F with 96 percent humidity and felt like 30°F; by 10:00 it had warmed to 45°F, and the wind stayed between 2 and 4 mph.

Against your last two half marathons.

	Oct 19, 2025	Apr 25, 2026	Oct 18, 2026
Watch time	2:26:51	2:33:25	2:43:58
GPS distance	13.22 mi	13.43 mi	13.31 mi
Pace (GPS)	11:07 /mi	11:25 /mi	12:19 /mi
Avg / max HR	174 / 187	178 / 193	161 / 185
Minutes in Zone 5	111 of 147	122 of 153	5 of 164
First half → second half	11:11 → 11:01 /mi	10:16 → 11:52 /mi	12:28 → 12:07 /mi
Avg HR first → second half	169 → 178	172 → 183	154 → 166
Pace-to-HR decoupling	3.6%	19.0%	4.2%
Feels-like at start	not recorded	60°F	30°F

The case against “substantially fitter.” You ran this race at an average heart rate 13 beats lower than last October and 17 lower than April, and you spent 5 minutes in Zone 5 instead of roughly two hours. A half marathon at 161 is supposed to feel far easier than one at 174 to 178, because it is a much lower intensity: about 10 percent slower than last October. Feeling comfortable at a lower intensity is not evidence of more fitness by itself. The test is how fast you ran for each heartbeat, and on that measure the race looks almost the same as a year ago.

Meters per heartbeat: 0.81 today against 0.83 last October, about 3 percent lower. April was 0.79.
Correcting for intensity: speed rises roughly in line with heart rate above resting, so a lower-intensity effort naturally scores lower per beat. Counting only the beats above a resting rate of 60, today comes out about 2 percent better than last October.
Verdict on this measure: somewhere between 3 percent worse and 2 percent better depending on the model, so effectively unchanged (moderate confidence). The data does not show a substantial gain in fitness measured as speed for effort.

What the race does show.

You paced it far better than April. In April you went out at 10:07 to 10:11 per mile with your heart rate already at 176 to 182 by miles 3 and 4. The last four miles fell apart to 11:53, 12:48, 13:13 and 14:04, and pace decoupled from heart rate by 19 percent. Today you started at 12:00, ran the second half about 21 seconds per mile faster than the first, and decoupled 4.2 percent, about the same as last October’s well-run race. Mile 13, at 11:14, was your second-fastest of the day after mile 10 at 11:12.
Your aerobic system carried the whole distance. About 57 percent of the race (94 of 164 minutes) was in Zones 2 and 3, the range the September long runs trained, and only 5 minutes were in Zone 5. In both previous halves, three-quarters or more of the race was spent in Zone 5. You covered 13.1 miles without needing the top of your heart rate range. That fits the base-building since August doing its job, but it’s an inference from one race, not a measurement.
You did it sick and without sleep. The watch recorded no sleep at all for the night before the race; the last recorded night ended at 09:52 on Saturday morning. Illness and sleep loss both normally raise heart rate at a given pace. If they did today, the healthy version of this race would have been faster for the same heart rate, so treat 2:43:52 as a floor, not a ceiling. The 30°F start pushes the other way: cold weather keeps heart rate lower than an identical effort on a warm day. The data can’t separate those effects (low confidence on the net).

Mile by mile.

Mile	1	2	3	4	5	6	7	8	9	10	11	12	13
Pace	12:00	11:48	11:33	13:22	13:38	12:44	13:36	11:44	11:47	11:12	12:47	12:52	11:14
Avg HR	135	146	152	163	164	166	163	165	161	158	163	172	178

Where the time went. The watch’s mile numbers are GPS miles, which ran about 0.2 miles ahead of the course markers by the finish. Matching the GPS track against the race’s course map (the published map is from the 2025 race; your 2026 track follows the 2025 course for about 86 percent of its length, differing mainly around GPS miles 7.5 to 8.0 and 9.1 to 10.4), you slowed at four of the eight fluid stations, and all four sit on the unchanged sections:

GPS mile	Time slower than 20:38 /mi	Where	Station on the course map
3.69	about 35 s, came to a stop	E Broad St, east of Franklin Park	between mile markers 3 and 4
6.74	about 54 s, came to a stop	E Broad St at Franklin Park, return leg	Franklin Park Conservatory
8.41	about 42 s, came to a stop	18th St beside Nationwide Children’s Hospital	near mile marker 8
10.50	about 19 s, slowed but never stopped	German Village, beside Schiller Park	Schiller Park

Those times include slowing down into each station and getting back up to speed, so the time actually standing still was shorter. Together they add up to about 2.5 minutes and explain most of the extra time in miles 4, 7, 9 and 11. You ran straight through the other four stations.

Miles 5 through 8: the stretch the stations don’t explain. Mile 5 (south on Drexel Ave in Bexley, toward E Main St) ran 13:38 with no stop. Almost every second of it was slower than 13:25 per mile, so it was a sustained slower pace, not a pause. Your cadence fell from about 160 steps per minute in miles 1 to 4 to 152 to 155 in miles 5 to 8, and came back to 157 to 159 by miles 10 and 11, when the pace recovered to 11:44, 11:47 and 11:12. Two outside explanations can be ruled out. The wind was about 2 mph during mile 5 and never above 4 mph during the race. The course map’s elevation profile shows no significant hill there (the watch’s own elevation is altimeter noise: 175 m of “climbing” in mile 1 of a flat course). What remains is internal: fatigue, the illness, or the missing night of sleep catching up with you for a few miles, then easing. The data fits that but can’t prove it (low confidence).

What the night after says. Last night your respiration rate averaged 18 breaths per minute. Over the previous 30 nights it never went above 16 and averaged 14.9. Resting heart rate was 66 against a 30-night mean of 62.1. Both are consistent with a body still fighting an illness on top of a race. Keep this week easy and short until both are back near baseline; there is nothing to gain from training hard into an illness.

A cleaner test. To measure fitness without the confounders, run a controlled benchmark when you are healthy and rested: the same flat route, the same weather window, a fixed heart rate (for example 30 minutes at 150 bpm), and compare the pace with a September run at the same heart rate. Run it again every four to six weeks. That will answer the question this race can’t.

Navigation