Lattice Blog

Open or half crimp for performance profiling in climbing?

We use maximal finger strength (as measured by a 5 sec. hang on a 20mm Lattice Edge) as a key data point in our performance profiling and when combined with all the results from our testing it gives a good indicator of climbing potential. However, we have wondered for some time whether forcing a particular grip position during testing would make for a better measure of performance (effectively removing the ‘grip position’ variable  from the equation).

Existing research uses maximal tests in non-overhead positions and in our experience this has poor specificity and does not produce good results, so that line of enquiry was out!

What we decided to do next was run a specific testing event at one of our Lattice partner walls to try and determine whether open or half crimp grip position was a better measure of climbing performance. The sample size (n=13) was too small to draw any statistically significant conclusions but what we did notice was several cases where a climber’s open handed grip strength significantly exceeded our expectations given their reported grade.

From here we decided to investigate this pattern further by looking at our existing data set collected over the years of assessments. Fortunately we’ve consistently collected information about the grip type used in each assessment so we could easily compare the two.

What we found was a better correlation between finger strength in a half crimp position and boulder grade vs. finger strength in a an open grip position and boulder grade. The correlations with route grades were similar.

The supporting statistics for both grip types did not significantly alter this finding, so based on this data it is our opinion that it is better to measure grip strength in the half crimp position than an open grip position.

We are always looking to improve our processes, so we’ll continue to collect data on this issue and alter our processes when it makes sense.

Note: this post is based on a previously published facebook post.

8 comments

  • christopher foulkes says:

    hi remus
    when doing the crimp finger strength testing session i find that keeping my fingers at 90 degress on the 20mm edge is near impossible,they just open out ,yet if i use a drag type position can add upto 30kg.
    does this mean i need to work on this 90 degree position strictly! to improve finger strength.

    chris

    • Hey Chris, it sounds like you’ve got quite a big difference between your half crimp and drag grips, so working on your half crimp is likely to help improve your overall finger strength.

  • Juhanic says:

    Have you controlled for grip preference?

    I feel there’s a pretty good chance that if you measured the correlations only for the people regularly using that grip you may get a stronger correlation.

  • Georg says:

    Guys, I know, having fancy garphs always gives a scientific look but it is mandatory that the numbers and axis are readable!

    • Christian says:

      I think what he means is that the residuals (left plot) of the model they are using are distributed more evenly across the different grades for half-crimp. Let me try to explain.

      They probably did a statistical model where climbing performance as some linear model of finger strength. Say
      climbing_performance = a + b*finger_strength.
      This model says: give me your finger strength, and I will say what your max grade is.

      So when they use this model, they can compare what the model says and what they actually observe. The difference between what the model say and what they observe is called the residual (i.e. model error).
      In any statistical model (that is useful) there is residuals. Statisticians often compare the residuals between different models to asses which model works best. One way is to compare if the residuals have the same distribution across observations. This is to ensure that the accuracy of the model is the same for different observations (the residuals should on average be 0 and have the same variation around zero for all observations (i.e. for all grades)).

      When using the half crimp, the residual distribution (the gray and white vertical boxes) looks to be more equal for different grades. The accuracy of the half-crimp model is higher, hence half crimp is a more reliable measure for climbing performance (grade).

  • G says:

    Cheers Remus

  • G says:

    Hi, how do you define the grips?

    Is half crimp just crimp without the thumb, or more like chiselling?

Leave a Reply

Your email address will not be published.