I was in the best shape of my life. By the numbers, I was projecting a time in the low 9s at Ironman Cozumel 2012: swim around 59 minutes, bike at 240 watts for a 4:40 split, run at 7:00 per mile for a 3:10 marathon.
The preparation was obsessive. 15-20 thousand meters of swimming per week. 40 miles of running. Most bike training on an indoor trainer in the aero position. Most weeks were 20-plus hours, often starting at 4am. I had no social life. I carried hand sanitizer everywhere.
This was my comeback race. Five weeks before Ironman Coeur d'Alene earlier that year, I'd been hit by a car and broken two ribs. I still raced, but missed a Kona slot by 15 minutes. Cozumel was meant to be redemption.
I arrived early to reduce travel stress. I brought all my own pre-race meals. I only drank bottled water. I even brushed my teeth with bottled water.
The preparation was perfect.
Then Everything Changed
Two days before the race, my one-year-old son was up all night vomiting. It lasted less than a day for him. My wife got violently ill on Friday night and couldn't move all of Saturday. I was still feeling fine, though at the back of my mind I had a sinking feeling.
By 6pm Saturday I really didn't feel well. I went straight to sleep without eating. I woke at 11pm with a bloated gut, weird since I hadn't eaten in over 8 hours. Then it hit. I spent most of the night on the toilet.
At 4am on race morning I couldn't face solid food. I drank two pre-made Starbucks frappuccinos for 400 calories and took two Immodium. I was determined to at least complete the swim and attempt the bike. I couldn't let all this prep go to waste without even trying.
That last sentence is the one I want you to pay attention to. It's the sunk-cost trap in its purest form.
The Race: Watching Fitness Evaporate
The swim was 1:10. One of my worst IM swim times ever, in the best swimming shape of my life. The current was brutal, but the number shocked me. My coach Kevin shouted that it had been a very slow swim for everyone, which helped. In T1, I got on the bike and started riding before realizing I still had my swim skin on.
On the bike, I was struggling to hit my power numbers. 240 watts usually feels easy and I was working hard just to get near 230. Through 56 miles I was still in contention at 2:25, so I told myself it was fine. Then it fell apart.
I couldn't keep anything down. Water came back up. Bars came back up. EFS came back up. I started bloating badly. I could no longer stay in the aero position. Each lap got progressively worse: 205 watts, 184 watts, 155 watts. Average speed dropped from 23 mph to 17.5 mph. Five porta-potty stops on the bike, each one worse than the last. Bike time: 5:38.
About halfway through the final lap, I decided I would not start the run. I was severely dehydrated and running on zero fuel. A marathon would not only be miserable but dangerous.
I limped through the final miles, relieved to be done. DNF.
The Lesson Most People Take Away
The obvious takeaway is "sometimes you get unlucky." Virus hits, race is ruined, nothing you can do about it. Bad luck. Move on.
That's true but incomplete. The more useful takeaway is about race planning and specifically about what happens when single-day variance strikes.
Single-Day Variance Is a Category
In my racing framework, I separate the things that can go wrong into two categories. The first is preparation failures: things you could have controlled but didn't. Not enough training. Wrong nutrition strategy. Equipment not dialed in. These are your fault and they're fixable.
The second is single-day variance: things that are outside your control on race day. Weather. Illness. Equipment failure that wasn't preventable. Another athlete crashing into you. These are not your fault, and the only thing you can do is have a plan for when they happen.
Most athletes plan obsessively for the first category and barely acknowledge the second. They train for months, dial in nutrition, practice transitions, fine-tune equipment. Then they show up on race day with zero plan for what to do if something goes wrong that isn't their fault.
Cozumel 2012 was a single-day variance event. I did everything right in preparation. The virus was a random input that overwrote all of it.
What I Should Have Done Differently
Here's where it gets uncomfortable. I should not have started the race.
By 4am on race morning, the data was clear. I'd been up most of the night with GI distress. I hadn't eaten in over 20 hours. I was dehydrated. The probability of a good race was near zero. The probability of a mediocre race was low. The probability of a DNF or a medical incident was high.
The rational decision was DNS. Don't start.
But I started anyway, because of the sunk-cost trap. "I can't let all this prep go to waste." "I flew all the way to Mexico." "I'll feel like a quitter if I don't start." "Maybe it'll clear up once I get moving." All of these are the brain rationalizing a decision that the data doesn't support.
Starting that race cost me: a miserable 7 hours of racing. Severe dehydration that took days to recover from. A full week of being unable to eat normally. And roughly 2-3 weeks of additional recovery time versus a DNS.
If I'd DNS'd, I'd have been training again within a week. Instead I was wrecked for three weeks. The sunk cost of "all that preparation" actually increased by starting.
DNS Criteria
Since Cozumel, I've built explicit DNS criteria into every race plan. These are pre-committed decisions that I make before race week, when my judgment isn't compromised by adrenaline and sunk-cost reasoning.
The criteria are simple. If any of these are true on race morning, I don't start:
Any GI distress in the 12 hours before the race that prevents eating or hydrating normally. Any fever or confirmed illness in the 48 hours before the race. Any injury that would alter my gait or stroke mechanics enough to cause a secondary injury. Any resting heart rate more than 15% above my baseline on race morning.
These aren't negotiable. They're pre-committed. I decided them when I was thinking clearly, specifically so that I wouldn't have to decide them when I'm standing on a start line at 5am with thousands of dollars of entry fees and travel costs whispering "just give it a shot."
The Aftermath
The next day I still couldn't keep anything down. On Tuesday afternoon, 48 hours after the race, I finally managed to eat something. A 48-hour stomach virus on the only two days that really mattered.
But within a few days, I was online, booking my spot at the inaugural Ironman Los Cabos. I would return fitter than ever. I called it my Mexican Revenge.
Three months later, I won my age group at Los Cabos in 9:42, qualifying for Kona. The fitness that produced the DNF at Cozumel was the same fitness that produced the PR at Cabo. The only difference was single-day variance.
That's the thing about single-day variance. It doesn't mean your preparation was wrong. It means race day delivered an input your plan didn't have a protocol for. The fix isn't "prepare harder." The fix is "build DNS criteria and recurring-fault protocols into your plan, so that when variance hits, you have a pre-committed response instead of a panicked improvisation."
Peak fitness and a DNF can coexist on the same body, on the same day. The question is what you learn from it.
Want to work with me? I coach athletes from first-time Ironman to Ultraman.