The amateur marathon runner always believes the first three miles are a prophetic vision of the remaining twenty-three.

They check their watch at the five-kilometer mark, they see a pace that suggests a record-breaking morning, they ignore the slight twinge in their left hamstring, and they proceed as if the physics of exhaustion simply do not apply to their specific biology.

We do this in construction and facilities management every single Tuesday. We sit in windowless conference rooms with lukewarm coffee, we look at Gantt charts that possess the aesthetic clean lines of a Mondrian painting, we listen to contractors promise that the system impairment will be a surgical strike, and we all collectively agree to pretend that the shutdown window is a fixed, immutable law of nature.

01

The Vertical Tomb Paradox

Last month, I spent twenty minutes stuck in an elevator. It was a small, brushed-metal box that smelled vaguely of industrial floor wax and the anxiety of the three people who had occupied it before me.

When the car lurched and stopped between floors, the first thing I did was check my phone, which had no service, and the second thing I did was press the alarm button, which emitted a sound that was less an emergency signal and more a polite suggestion that something might be wrong.

“The intercom crackled to life after four minutes. A voice, distant and strangely bored, told me that a technician was on-site and the delay would be ‘minimal.'”

I spent the next sixteen minutes contemplating the difference between a “minimal” delay and the reality of being suspended in a vertical tomb by a series of steel cables that I had previously given zero thought to.

In those twenty minutes, I realized that my entire day-a series of tightly packed meetings, a livestream I was supposed to moderate, a grocery list-was predicated on the assumption that the elevator would behave according to its programmed intent. It didn’t. It had its own timeline.

02

The Catastrophic Quick Fix

The industry treats fire system impairments with the same delusional confidence I had in that elevator. We plan a sprinkler head replacement or a backflow preventer test, we notify the monitoring station that the system will be offline for a specific duration, we scope our safety protocols for that specific duration, and we act shocked when a rusted valve snaps at 4:00 PM on a Friday.

The shutdown window was supposed to be six hours. Now the building is vulnerable for the entire weekend. Everyone in the room has lived through the “quick fix” that morphed into a catastrophic week-long ordeal.

We have seen the blueprints that didn’t account for the 1970s-era wiring hidden behind the drywall, we have felt the panic of a fire panel that refuses to reset after a simple test, we have watched the sun go down on a project that was supposed to be wrapped up by lunch, and we still sign off on the same optimistic estimates for the next job.

To plan for an overrun is to admit that we lack total control, and in a field built on the illusion of mastery over the built environment, that admission feels like a defeat. But the building doesn’t care about our feelings of mastery.

When the pipes are dry and the smoke detectors are bagged in plastic, the building is no longer a structured environment; it is a pile of potential fuel waiting for a thermal event. We treat the impairment as a technical hurdle to be cleared, rather than a period of high-stakes vulnerability.

We scope the safety measures, like fire watch, to the optimistic estimate because the optimistic estimate is what the budget allows. If we admit the shutdown window might double, we have to admit the cost of the safety net might double. So, we lie. We lie to the insurers, we lie to the tenants, and most dangerously, we lie to ourselves.

03

Buying Against Your Own Optimism

The logic of the industry is a race to the bottom of the best-case scenario. We choose the fiction that costs the least on paper, ignoring the massive, unhedged risk of the “gap”-that period where the planned coverage ends but the system is still offline.

I have sat in meetings where the safety plan was debated with the intensity of a trial. The engineers argued over the pressure thresholds, the architects argued over the aesthetic of the replacement fixtures, the accountants argued over the overtime rates, and the fire watch was squeezed into a tiny, two-day box that everyone knew was insufficient.

We treat fire watch as a line-item expense to be minimized, rather than the only thing standing between a minor electrical short and a total loss of the asset. When you hire a Fire watch security company, you aren’t just buying a pair of eyes and a flashlight; you are buying an insurance policy against your own optimism.

The reality of facilities maintenance is a series of “known unknowns.” We know the system is old. We know the parts are backordered. We know the crew is stretched thin. Yet, we still write “48 hours” on the permit.

This is the base rate fallacy in action. We ignore the historical data of our own failures in favor of the specific, flattering story we tell ourselves about this particular project. “This time,” we say, “we have the right team.” “This time,” we say, “the weather is on our side.”

But the weather changes. A team member gets sick. A tool breaks. And suddenly, that forty-eight-hour window has evaporated, leaving the building exposed. The danger isn’t the delay itself; it’s the fact that the delay was never provisioned for. We didn’t book the guards for the third night because the third night “wasn’t supposed to happen.”

04

The Shift to Reality

When I was in that elevator, the bored voice on the intercom was my only link to the world of functioning systems. I wasn’t thinking about the cost of the repair. I wasn’t thinking about the technician’s hourly rate. I was thinking about whether or not the building had a plan for when things went sideways.

I realized then that most of our systems are designed for the “expected,” and we are woefully under-prepared for the “probable.” It is probable that a construction project will run long. It is probable that a system shutdown will encounter complications. It is probable that the best-case scenario is a lie.

The shift that needs to happen in facilities planning is a move toward “reality-based scoping.” This means looking at the forty-eight-hour estimate and asking, “What does the building look like at hour seventy-two?”

It means acknowledging that the most dangerous part of any impairment isn’t the first hour, but the hour after the planned coverage has packed up and gone home. We need to stop treating safety as a variable that scales with the budget and start treating it as a constant that scales with the risk.

If the system is down, the risk is up. It doesn’t matter if it’s down for an hour or a month. The building’s vulnerability doesn’t diminish because the project manager is stressed about the deadline. In fact, the risk often increases as the project drags on.

Fatigue sets in. Debris accumulates. Temporary electrical setups become “permanent” solutions. This is exactly when you need the most oversight, yet it’s usually when we have the least, because we’ve already exhausted the “safety budget” allocated for the optimistic window.

I’ve seen facilities where the fire watch was dismissed at 5:00 PM because the permit said the work would be done. At 6:00 PM, the welding slag was still smoldering in a wall cavity. At 8:00 PM, the building was empty. At 11:00 PM, the sirens started.

All because someone believed a piece of paper over the reality of a construction site. We have to stop being fans of our own schedules. We have to start being critics of our own assumptions.

The elevator eventually moved. It didn’t move because of my frantic button-pushing or my internal monologue about industrial safety. It moved because the technician ignored the “minimal” estimate and took the time to actually fix the underlying sensor issue.

When I stepped out into the lobby, the world looked the same, but I felt a profound distrust for anything labeled “automatic” or “scheduled.” We should apply that distrust to our project planning.

We should scope our fire watch for the “worst-case likely” scenario, not the “best-case possible” one. We should assume that the pipes will leak, the sensors will fail, and the weekend will arrive before the work is done.

By provisioning for the overrun, we remove the “gap” where catastrophes happen. We turn a potential disaster into a mere scheduling inconvenience. Ultimately, the cost of being wrong about the shutdown window is too high to keep pretending we’re always right.

The extra shift of coverage, the additional day of monitoring, the realistic buffer in the timeline-these aren’t “extra” costs. They are the true cost of doing business in an unpredictable world.

We can keep lying to ourselves in those windowless conference rooms, or we can start planning for the world as it actually exists: messy, delayed, and perpetually prone to catching fire when we least expect it.

I’ll take the extra coverage every time. I’ve spent enough time in the dark between floors to know that “minimal delay” is just something people say when they don’t want to admit they’re lost. Let’s stop being lost in our own optimism. Let’s build a safety net that actually catches us when the schedule falls apart.