Climbing grades differ by region or even by crag within the same area. Some routes or boulders may be graded by the hardest move, while others feel like they factor in rests. Ultimately, the grade boils down to the perception of the technical difficulty, steepness, and effort needed to clip the chains.
Climbers have done their best to keep grading consistent by crowdsourcing the “final” grade. The first ascensionist assigns a grade, but repeats can move that grade down (downgrading) or up (upgrading). Essentially, this relies on larger sample sizes to create more validity.
To add to the inconsistency, there are several grading systems at play, and converting between them doesn’t always feel right. Americans use the Yosemite Decimal System (YDS) for routes. Europe subscribes to the French grading system. There are also Australian and British systems. Throw in differences in physique between climbers and the conditions, and it makes accepted consensus grades seem low percentage.
But in the last few years, an entirely new system has quietly gained momentum among climbers worldwide. And I’ve come to appreciate the data-driven focus behind it. French route setters and climbers in Grenoble, France, used their strong backgrounds in math and programming to devise a more systematic grading system in 2022: Darth Grader.
The goal was to break climbs and boulders into sections and rests. Then, an algorithm would assign the total route grade more accurately. While these section and rest ratings are still subjective, the Darth Grader calculator shrinks the subjectivity to smaller sections.
Proponents of the system claim it provides the most objective grade possible for something that cannot avoid subjectivity.
Issues With Climbing Grades
The creators of Darth Grader state that there are two main sources of cognitive biases when grading a route. These usually lead to overestimation or underestimation of difficulty.
The first is that when climbing at your absolute limit, any slight addition to that difficulty seems exponentially harder. The opposite can also be true; climbing routes well below your redpoint grade can seem overly easy.
For example, a 5.11 climber might feel a 5.12a is more difficult than the grade. Meanwhile, a 5.13 climber on the same route may feel that it’s easier than the grade. And the same climber may feel that route is harder or easier based on the current fitness level or conditions.
The second cognitive bias the Darth Grader founders discovered owes to the emotional investment climbers might attach to redpointing a project (leading a route without any artificial aids).
Maybe you’ve spent months or even years on a route, and you’ve traveled a long distance every season to do so. When you finally clip the anchors, it would be hard not to inflate the grade to justify the extreme efforts.
How Darth Grader Can Help
Although Darth Grader requires subjectivity, it aims to increase accuracy and consistency by splitting subjectivity into more manageable chunks.
Climbing just a section is easier than climbing the entire route. Darth Grader assumes you will be more accurate in grading each section and rest. The algorithm then takes your subjectivity out at that point and calculates an overall difficulty rating in a consistent way.
Any section that is between five and 10 moves is subjectively graded on V-grades (in the U.S.). Sections longer than 15 moves require the use of YDS. Both of these ratings have the option of “hard” or “soft.” Rests are rated as “good,” “medium,” “bad,” and no rest. Darth Grader can also rate multi-pitch routes.
To enhance accuracy, the Darth Grader “white paper” lists benchmark routes for each grade. This presents uniform standards on which to base the subjective grading of each section. The paper also includes a definitive guide on how to rate each rest.
The online calculator handles the remainder of the process of grading the entire route.
Darth Grader Caveats
Due to the cognitive biases, Darth Grader specifies routes that are reasonably below your limit but harder than “easy.”
“Medium” difficulty routes create the most accurate grading per section. And since you are grading retrospectively, you must know the climb and each move and rest very well. You must also have enough sends in the grades you are considering to know what each feels like. This requires a lot of climbing.
However, these are requirements for the current method of grading the entire climb subjectively.
Does Darth Grader Work?
Although I am accustomed to reading research papers about exercise physiology, the white paper about Darth Vader was more math than climbing. So, I asked a self-proclaimed climbing nerd to look it over.
I sent the white paper to my climbing buddy Sean Moorhead, who has spent his career in data science, statistical analysis, and algorithm design. Among a long list of achievements, Moorhead has published several astrophysics papers and a white paper for a new software algorithm for data storage and processing. This paper led to a patent.
In short, he was the exact right person for the job.
His response to my query about legitimacy started off with, “They’re trying to put a math function to reverse engineer grades. The part of the paper where they say, ‘If Y is the grade of the first section and X is the grade of the second section …’ is interesting and demonstrates that grading isn’t as objective as we think it is.”
Moorhead pointed out one glaring shortfall. “The sample size of data is really small and from a limited geographic area to make any real conclusions,” he said. “If they really are doing a scientific analysis of how humans grade climbing difficulty, they should broaden their study to international routes, a variety of first ascensionists across a variety of grades and styles, and let the data show how people perceive difficulty at aggregate.”
But Moorhead applauds the effort. “This is definitely an opportunity for really cool insights into the human perception of effort and difficulty. If they rolled this out to a sample size of a few thousand routes, I’d maybe trust it more,” he concluded.
Darth Grader Rating System: Final Thoughts
The creators of Darth Grader have a plethora of information, theories, and explanations of how and why their system provides more accuracy and consistency than the traditional way. I used two routes that I have done hundreds of times over 30 years of cragging in the same area.
To keep the variability lower, both routes are 5.12a. Each is popular for the cliff, and the grade has withstood the test of time. Neither route’s grade has been argued or changed for decades.
I know each move like the back of my hand. I can verbalize with accuracy how every handhold, foothold, or smear feels when it’s hot or cold, humid or dry.
My Results With Algorithmic Route Grading
I was super-deliberate, dutifully considering each section’s and rest’s grades before entering my numbers into the calculator. When I did, it produced surprising results. In both cases, the algorithm claimed two letter grades harder, delivering a little ego boost. I assumed rating each section would result in a downgrade by taking the power endurance factor out of grading.
But I recalled the countless times visiting friends got upset with me, claiming I had sandbagged them with a route that was harder than its listed grade. And it wasn’t a slight sandbagging to them; it was in-your-face ridicule — as in two-letter grades.
So, is this system perfect? Not at all. But theoretically, it makes sense to my algorithm-chewing friend and myself. The free-to-use effort to change the status quo with logic and math is refreshing. I may spend a lazy afternoon punching in all my local redpoints. I could always partake in a little ego-boosting!
