Goodhart’s Climbing Grade? Fine But How Long Could He Deadhang?

You may have heard the phrase ‘it’s a speed limit, not a target’ and they’re kinda right. The limit is set as a maximum safe speed you could travel, rather than as some arbitrary number to match our speedometer. There’s a lot of numbers you could choose between, the idea is simply not to pick a speed above it.

Now I’m not one to judge anyone on their driving style or choice of speed, nor is that my point. What I find interesting about the phrase is that it actually demonstrates one of my favourite laws called Goodhart’s Law.

But how? And why does it matter so much? Surely we like having targets in climbing! Well, yes we do but the problem comes when we conflate the two.

Goodhart’s Law Originally

Okay I’ll level with you on this one: I tried to read the original paper to cite Goodhart’s Law but, being from Monetary Theory and Practice, I struggled to get through it. However, this is where the concept (if not the phrase specifically) seems to stem from. These days, Goodhart’s Law has become so commonplace that we can even look it up on Wikipedia.

Charles Goodhart was a British economist but it didn’t take long before other sectors began utilising the same concept; or indeed falling foul. In the 1980s/90s, schools and universities began using test scores, graduation rates, and league tables as performance targets (as anyone who’s been through the education system can surely attest). Oh the irony that a little bit of research might have prevented them forcing countless youngsters into what are effectively simply memory tests. Elton (2008) looked performance indicators in higher education, while Fire and Guestrin (2019) even equated how Goodhart’s has infilitrated the success ratings of academic institutions; I wonder if Goodhart’s alma mata does the same…

A little later and healthcare did likewise and I’m sure many Brits will be aware of the government obsession with NHS waiting times. The issue? By focusing on waiting times, it becomes a target and the standard of care can fail. I could go on: business, marketing, even policing have found themselves with various performance targets leading to a reduction in the quality of service (that one comes not from me but from the Radio 4 programme It’s a Fair Cop).

To put it simply, the more strongly people are judged by a number, the more likely they are to manipulate behaviour around that number—making the number less trustworthy. And it works in sport too.

Goodhart’s In Sport

Stuart McMillan from Altis wrote The Shuffle Start & Goodhart’s Law: when good enough is good enough” with a brilliant comment:

Runners often set split times for each mile or kilometer. However, if a runner becomes overly fixated on these splits, they might not adjust their pace based on the race’s dynamics, such as changes in elevation, weather conditions, competitor’s tactics, or their own physical state. By rigidly sticking to predetermined splits, they risk not capitalizing on segments where they could push harder, or not conserving energy when needed — potentially affecting their overall performance.

If we go more academic, Impellizzeri, Marcora and Coutts (2019) shows how training load metrics can be useful but misleading if treated as the goal rather than a guide. I could go on but let’s see how this relates to climbing specifically.

Goodhart’s Law In Climbing Context

British Mountain Guide and Strong Mind coach Angus Killie describes how Goodhart’s consicely traslates into the climbing sphere while Dave McLeod (who I’m sure needs no introduction) warns on “Focusing too hard on subgoals like hangboard numbers can distract from keeping track of other, harder to measure goals”.

There are several measures that we take in climbing and with climbing coaching heading in the direction it appears to be (more competitive with standings, league tables, podium positions and so on) it seems likely that the issue is likely to get worse instead of better. I’m going to bench the competitive element but instead will focus on recreational climbing. It’s still as relelvant there too.

Grades As a Measure of Progress

The most simple and common measure we use in climbing are climbing grades. I’ve written about them a lot on here in the past (here discussing what it can tell us about our style and here investigating how different grades tell us about our tactics, analytical skills and perseverance) but in the context of Goodhart, let’s consider them merely as a measure of our progress.

Many beginner climbers aren’t as aware of climbing grades, certainly not as something to aim for. Instead, they come in to climb, trying a variety of projects, occasionally finding a win when they complete a climb of a new grade. For them, the grades become a measure of how well they are performing.

However, for most climbers, there comes a point where there is a shift. As progress slows, new grades fail to fall as easily and soon they become a target. One of the most common goals I hear from clients is “to climb 6c” or some such number.

This shift, applying Goodhart, shows that the grade no longer acts as a good measure of the climber’s progress. Instead, it has shifted to become something different. This shift isn’t necessarily a bad thing (more on this below) but we must then become aware that something has changed and we must adjust our approach accordingly.

Measures and Targets in Training

Imagine this as a scenario:

• You want to climb a local boulder problem but your fingers keep uncurling from the holds. You decide you need to improve your finger strength to succeed
• Your mate has a Tindeq (fancy force measuring device) so you arrange a test to measure how strong your fingers are for the right pull, hold type, hold size and so on. With this, you collect your data on your current finger strength, which tells you that you can pull 41kg in the desired pull, great!
• Now you come up with some training exercises that will get your fingers stronger, over time. You begin to crack on with the exercises
• Every now and again, you go see your mate and repeat the same tests with his Tindeq. By the third test, you find you can now pull 45kg! Progress!
• But you don’t know how hard you need to get your fingers to complete your climb… So you decide to aim for 50kg
• You continue to train and test but find that progress starts to come in smaller increments. After the same amount of training, you find you’ve added 2kg, not 4kg, to your max pull strength
• You continue to train, hard, sure that if you can only get to that magic 50kg pull, you’ll be ready to go try your project again

Now let’s analyse this little hypothetical story. What’s happened here?

1. The initial pull test is initially a measure to check progress for the desired move/problem
2. As the required strength is not known, metrically, increasing the result of the test quickly becomes the new goal or, in other words, becomes a target
3. As such, the initial measure no longer acts as a good way of checking progress. We may still be getting stronger, yes, but we have lost sight of the initial purpose of the whole exercise and may then miss an opportunity to achieve what we set out to in the first place

Behavioural Changes as a Consequence of Goodhart

We can see this time and time again in other sports. This article by Jim Stone describes Goodhart’s Law in volleyball and ice hockey and how turning measures into targets changes behaviours, distracts players from their original goals and sees them prioritise poorly in order to achieve the new goal; often moving them further away from ‘scoring the goal’ that will win them the match.

One big consequence of a shift from measure to target can be a behavioural change. In team sports, asking a player to get more shots on goal may see them shoot too early, rather than wait for the right opportunity or to pass to a teammate in a better position.

We’ve seen this sort of thing in climbing, too; and it’s not a recent thing! For decades, climbers have saved projects in order to be able to onsight the climb. Instead of the onsight having a place on a hierarchy of style, so to speak, it becomes a barrier to an attempt. I’ve spoken to so many older climbers that left classic climbs for the onsight only to have aged, declined and now have no opportunity to complete the climb in any style.

This combination of ascent stlye and climb can often exemplify Goodhart’s in climbing. Resonate with you? Consider whether the style is your target or the route. If the style is your measure of your abilities (on the condition it is safe to do so) ask yourself: why not give it a go? Because then at least I get to climb the route.

Ah, now you’ve hit on something different. It may be the case that your goals change as a consequence of beginning the initial training.

I often cite two stories at this point. One was a local climber in North Wales whose goal was to campus 1-4-7 on the Campus Board. He supplemented his training for his goal by going climbing. Now I can’t recall exactly but I would surmise that he didn’t initially enter the climbing wall with campusing as his goal. At some point, while campusing to improve his climbing, he found out about this difficult challenge and his priorities changed.

Who am I to tell him he’s wrong in his priorities?! Not that I would. Crucially, though, this climber wasn’t kidding himself that he wanted to get better at climbing. His target was clear, it was merely slightly more unusual than most.

Another story relates to a conversation about the above video from Hooper’s Beta. I was discussing the video with a junior coach and avid climber and whether Hooper was right. My argument was that actually, band at the feet was closer to climbing on steep ground and probably more appropriate if you were training front levers to improve your climbing.

After a bit of back and forth, he stopped, looked at me and said “yeah, but front levers are cool”. And that’s when it hit me: it depends on the target goal. Hooper makes complete sense if the goal is to complete a front lever but not if you’re using front levers to be better at climbing. The key is not to conflate the target and the measures used to get there.

Do I Even Need Metrics?

Metric testing – gathering hard data that we can measure somehow – has become very popular in climbing in the last few years. It is easy to interpret, it involves numbers that we can gather and see trends over time. And that makes a lot of sense.

But not everything is that simple. You’ve probably heard the expression ‘it’s not the size that counts, it’s what you do with it’ and it’s all to do with two concepts: quantity and quality.

If you go a little more academic, research works in a similar way. You effectively have two types of research: quantitative or qualitative. Other than being incredibly hard for me to say (it’s much easier to write), these two types of data are two sides of the same coin. Which we choose depends on which is more appropriate.

Some things simply can’t be measured very easily. Sure, you can apply a simple 1-10 scale but often, it’s not actually that necessary. Moreover, it might just make things much worse for yourself.

A Cautionary Tale

I’d like to finish with a warning. I’ve written this article very soon after two previous pieces to be published on the Resource Centre. Excited at my newfound enthusiasm for writing again, I checked to find I’ve written 32 articles to date.

“Wow,” I thought, “That’s more than I realised! I wonder if I could make it to 50…”

Therein lies the irony: in writing a piece on Goodhart’s Law, I fell foul of Goodhart’s Law. It is very easily done. So be warned.

Summary

Measures and Targets are both completely valid ways of progressing but we need to be careful. Conflating these two can lead to issues.

A conscious shift is not necessarily a bad thing but often, a switch from a measure to a target can be subtle enough that we fail to notice; thereby no longer measuring our progress as we initially intended.

We’re certainly not alone in this issue and many other sports have various different reasons for Goodhart being so significant. Still, with us being climbers it is worth considering how it affects us specifically.

So the next time you’re considering your climbing grade or partaking in some metric training to get stronger, consider whether you are measuring as a metric or a target. The difference could matter more than you may think.

References

Goodhart, C. A. E. (1975). Problems of monetary management: The UK experience. Sydney: Reserve Bank of Australia.

Elton, L. (2004). Goodhart’s law and performance indicators in higher education. Evaluation & Research in Education, 18(1–2), 120–128. https://doi.org/10.1080/09500790408668312

Fire, M., & Guestrin, C. (2019). Over-optimization of academic publishing metrics: Observing Goodhart’s law in action. GigaScience, 8(6), giz053. https://doi.org/10.1093/gigascience/giz053

Impellizzeri, F. M., Marcora, S. M., & Coutts, A. J. (2019). Internal and external training load: 15 years on. International Journal of Sports Physiology and Performance, 14(2), 270–273

Kille, A. (2025, July 10). Goodhart’s law. Angus Kille. https://anguskille.com/2025/07/10/goodharts-law/

MacLeod, D. (n.d.). What to measure with hangboards [Online forum post]. Reddit. https://www.reddit.com/r/climbharder/comments/b0p0gp/dave_macleod_what_to_measure_with_hangboards/

McMillan, S. (2023, September 13). The shuffle start & Goodhart’s law: when good enough is good enough. ALTIS. https://altis.world/center-news/the-shuffle-start-goodharts-law-when-good-enough-is-good-enough/

Moore, A. (n.d.). It’s a Fair Cop [Radio programme]. BBC Radio 4. BBC Sounds. Retrieved April 25, 2026, from https://www.bbc.co.uk/sounds/brand/b060fj66

Wikipedia contributors. (n.d.). Goodhart’s law. In Wikipedia. Retrieved April 25, 2026, from https://en.wikipedia.org/wiki/Goodhart%27s_law