Here we test AT ski bindings for backcountry skiing. All of their design and function is built for human-powered, climb-up-and-ski-down ski adventuring. Call us crazy, but we chose to do the vast majority of our testing in that environment. Any mechanical support was incidental to our backcountry skiing. Read other reviews on the internet carefully; most seem to have conducted the testing of backcountry ski equipment in ski resorts. Sure, some things can be deduced there, but it is also easy to lose perspective on what is important.
We employed a large team of accomplished skiers, spanning multiple years now, to test backcountry ski bindings. In the last year alone, our lead test editor himself has tested bindings in the Tetons, Wind River Range, Alaska Range, Chile, and Argentina. Previous seasons and other testers have taken things even further afield. Of course, some things need to be measured on the workbench. We note how we tested for each important binding criterium.
We tested touring performance by going touring. We skinned tens of thousands of feet with every single binding in our test. Some of our tested bindings are approaching 100,000 vertical feet of human-powered skiing. While doing so, we collected observations and took notes on propensity for icing, heel riser options, and adjustment range, and in toe pivot range of motion.
Again, our primary means of testing was in actual skiing. While skiing on a wide variety of surfaces (perfect pow to soul-crushing breakable crust), we noted mainly binding hold. Our team is pretty handy at noting, anecdotally, force transmission, chatter damping, and binding geometry, all while skiing. In just a few instances could we test release function. Mainly, our deductions on release performance defer to manufacturers and third party formal tests. When we can defer to the TUV organization and their DIN/ISO certification process, we note so. When we are citing manufacturer's descriptions, we make that clear as well. Finally, some aspects of downhill performance are a function of binding geometry. Using a millimeter ruler and a flat workbench, we measured each binding's stack and delta.
First, we measured the distance from ski top sheet to center of toe pins and center of heel pins (or deduced location of virtual heel pins in the case of bindings that do not use heel pins). We performed such measurements with a boot in the binding, in downhill mode. Binding delta is reported as the difference between heel and toe measurements, in millimeters. Binding stack height is the average of the two measured values.
Ease of Use
We define ease of use as the ability to get in and out, make transitions, and make adjustments to length and release value. To test, we performed all of those operations, comparing one binding to the others.
Before mounting, we weighed each binding, including mounting screws on a digital gram-sensitive postal scale. We have calibrated this scale recently against two other scales. In scoring, we made some minor adjustments to reflect presence or absence of optional features.
In some blessedly rare cases, we have actual durability failures to report on. In other cases, we have literal decades of experience with a certain product. Beyond that direct experience, we are forced to report our durability opinion based on second-hand reports and speculation based on a sound understanding of basic binding construction and materials.
We care about the testing we do and the conclusions we reach. We first ponder what you want and need to know, and then design our entire testing procedure backwards from there. The result, we are confident, is review content that informs without pushing. We have our biases, but we work real hard to guard against them. We have our preferences, and we honor them and understand when they come from experience and a sound understanding and when they come from more intangible sources. As it pertains to AT ski bindings, the information we present is based on decades of experience, a highly varied test team, and thorough processing of the information we all generate.