Math never lies, but assumptions are frequently wrong
Few professionals are impactful enough on their chosen field to merit having something named after them. Werner Munter, an iconic Swiss mountain guide, has done it twice. He is credited with introducing the “‘Halbmastwurf’ — a belay hitch for climbing — which for English-speaking climbers and guides has become ubiquitously known as a Munter hitch.
The Munter hitch is an elegant, adjustable, reversible means of controlling friction on a rope while belaying or rappelling and can be used as a load-releasable tie-off. He also developed a standard methodology to forecast the time requirement to complete a given route in the mountains. The model is versatile enough that it can be used across a diversity of conditions, going uphill or downhill, from walking to alpine climbing and skiing. His model has come to be known as the Munter Pacing Model. It has become so well accepted and deployed that there is now (of course) an app for it (Link).
Like all models, the Munter Pacing Model is only as good as the user inputs and the assumptions that it relies upon. The critical assumption in this example is the Munter Pace at which a skier or climber can travel up or down a given leg of the route. With a bit of experience, a backcountry skier can quickly learn to forecast an accurate Munter Pace for a given set of conditions and in turn, set achievable time expectations for a desired route in the mountains.
When used correctly, the model has proven to be a robust tool for trip planning. One way things can go sideways is when the actual conditions differ from the forecast conditions and the backcountry skier fails to adapt plans to conditions in the moment. For example, creating a time plan that forecasts easy movement but encountering deep snow and heavy trail breaking. Unless plans are adapted to the new conditions, a skier could easily find themselves digging an emergency shelter instead of enjoying après-ski back at the local pub.
All models require user inputs and assumptions. For any model, especially those used in a predictive fashion, it is critical that model assumptions are periodically tested and, if warranted, revised. When a model fails to predict a certain outcome, you will often hear the lamentation that “the model was wrong.” A model at its core is simply math. Chances are the model (i.e., the math) is fine, but the user assumptions were wrong. Blaming the model is an intellectually lazy attempt to avoid accountability for inappropriate and/or outdated assumptions. This is equally true for a monthly household budget, a company financial model, a technoeconomic model, or a time plan for a day in the mountains.
