Sunk Cost Fallacy & AI
Don't Take A Horse & Buggy When Automobiles Exist
Sunk Cost Fallacy
Economists coined the term. You keep investing in something not because it still makes sense, but because abandoning it means admitting the time and money already spent no longer makes sense.
Sports performance coaches are not immune to this.
Sports scientists who spent years learning to code. Data analysts who built their entire identity around proprietary systems. Performance coaches who hand-built Excel files for a decade plus. AI is not a threat. It is a tool that if utilized correctly can save us all time and money. The threat is the decision to dig in and protect the past instead of leveraging what you already know to build something better.
“If I had asked people what they wanted, they would have said faster horses.” -Henry Ford
Nobody asks for the thing that makes their current expertise feel smaller. That does not mean the thing is not coming.
Teambuildr Vs Excel
I was an Excel guy for over a decade. Thousands of hours committed to building and iterating templates. Drop-down menus, automated print functions, a homemade tracking system that made my laptop sound like a Boeing 737 when I opened up the document.
It worked incredible well. I was and still am proud of it.
Then I found Teambuildr. This did not make my time building excel files wasted, but more time spent using them certainly would be. Exercise prescription, management, and tracking. All of it is better with Teambuildr. Cleaner, faster, and saves me countless hours per week.
The easy move would have been to dismiss it. To say the custom file was more specific to our needs, that I already had a system, and that the learning curve was not worth it.
That would have been the sunk cost fallacy in action.
The right move was to use the better tool and master it.
Our field loves to talk about being overworked and underpaid, but don’t ask the important question often enough. What am I doing daily that is a complete waste of time?
What Actually Threatens People
Two things drive the resistance to AI in this field.
One is the sunk cost fallacy. Years of learning a skill, building a system, or mastering a platform make it psychologically painful to accept a faster path exists. The investment feels wasted. It is not. The knowledge transfers. The reps you put in made you capable of asking better questions and recognizing bad outputs. That does not disappear, it gives you the best chance to maximize the use of it.
The second is more honest: some people are threatened by the fact that others can now do what took them years in a fraction of the time. A coach with no coding background can build a functional AMS.
IF and only IF they know what questions to ask. This does not empower those who are lazy. It empowers those with years of experience in this field, who know what questions to ask, but have not had the skills to build.
This does not take away experience. That will always remain important. The ability to recognize when something is wrong before it fails.
The AMS Case Study
This winter, a vendor quoted us $40,000 for an off-the-shelf athlete management system. Centralized data, but stock options, just a cookie cutter software.
I built our own in under a week. Custom to our exact needs. On a $20 monthly subscription.
The $40,000 system still exists and will for years, because like most technology it will be resisted by many. It is the same reason people still print off excel files weekly, or coaches hand scout cards weekly. Change is hard for people.
That last part is not a criticism. It is just reality. Most people do not change, regardless of the upside. Change requires acknowledging that the old way is no longer the best way.
What AI Cannot Replace
AI will not replace coaches. It won’t replace learning. It will not be able to program better than someone with decades of experience who understands their team culture, athletes, and knows how to utilize the resources they have available.
A coach who cannot think critically, evaluate an athlete’s movement, or build genuine relationships will use AI as a crutch. Their outputs will get slightly more presentable while their core deficiency grows. Laziness and incompetence do not become competence because the tool changed.
A coach who can think, evaluate, and build will use AI to do in one hour what used to take a week. The ceiling goes up, not down.
The skill that matters most right now is not coding, data science, or software development. It is knowing what you are trying to build well enough to describe it clearly. What questions are you trying to answer? How do I utilize the data and build interventions? That requires real requires specific knowledge. It did not make a sports scientist any less valuable. It should simply save them time and money.
The Bottom Line
Nobody is coming to take your expertise. They are coming with a gas powered vehicle instead of a horse and buggy.
Don’t be scared to use it.
The coaches who spent years mastering systems, building programs, and learning how performance actually works are the ones best positioned to use AI effectively.
Stop protecting what you built. Build something better.
The Boeing 737 Excel file hit too close to home — at VFIT we built our own scoring sheets for combine data, finally killed them last year when a SaaS did 80% of it cleaner. The hardest part wasn't the tool switch, it was admitting that the years I spent making the file slick were a moat I'd built around myself, not the business.