Monday, November 16, 2009

THE EQUATION

And now, let's delve into something far more sinister: GRADES.



Up to this point, general discussion on grades in the climbing blogosphere has been fairly limited. Jamie has opened the conversation up several times here and here, and while many interesting opinions were voiced, little consensus was reached. I have two things to contribute to this discussion that will hopefully add a little more objectivity (not to be confused with OBJECTIVISM) to the topic of bouldering grades. READ ON.

For the past 7 and half years, I have kept an 8a.nu scorecard. Like many others, I have found the tracking of my progression to be quite fascinating. In a sport where there are no winners and losers (comps excluded), grades provide a paradigm for measuring one's success and seem to feed the innate competitiveness in many of us. While for most of us grades are not the biggest motivating factor, it's hard to deny the enjoyment one gets from breaking into the next grade or flashing a problem near your limit. Whether or not they dilute the climbing experience seems to be an irrelevant discussion to me; you may choose to care about them or not, but grades are certainly here to stay.

Recently my personal view on grades has evolved quite a bit. In the past I often chose to take the higher of two proposed grades when logging a boulder problem. Over the past two years i have more and more found myself downrating old ascents and taking the lower of two grades. This has been a crusade for accuracy; I realized that 8a is more of a tool for bringing about grade consensus rather than for advancing one's name in the ranking. This seems like a general trend in the 8a world. Many boulder problems that stood at a certain grade for years (think Eternia @ RMNP or The Flame @ Hueco), have since settled at lower grades after more people climbed them.

Since I've been paying closer attention to such trends, I've become fascinated by why certain problems are harder than others and how this might be better understood by breaking a boulder problem's grade down into the sum of its parts. For a while now, I have had a strange inkling that a problem's V grade could be accurately pinpointed by somehow adding up the difficulty of its individual moves. Then, several weeks ago it hit me; there appears to be an actual equation for this!


But only if:



Basically, X and Y are the V grades of two parts of a sequence. So, for instance, one section of V5 and one section of V7. Z is then the resultant V grade of combining those two sequences. In this example, 5 plus 7 is 12, divided by 2 is 6, plus 2 is 8. According to this theorem, combining a section of V5 with a section of V7 results in an overall section of V8 climbing. The limit is that the two sections that are being combined cannot be more than 4 grades apart. When the two grades are 4 apart, the resultant grade turns out to be the higher of the two numbers (i.e. a section of V9 and V5 ends up being V9, which makes sense). But, if they are more than 4 apart, the resultant grade would end up being LOWER than the highest of the two numbers (i.e. a section of V11 and V5 would add up to V10, this clearly does NOT make sense), hence the limit. In essence, sections that are 5 grades apart or more will always add up to the higher of the two numbers.

Here are some simple examples of this equation breaking down well known boulder problems:

Tommy's Arete, RMNP (CO): Generally regarded as two sections of V5; no single crux, very sustained. Fairly standard long V7, maybe on the slightly easier end.


Riddles in the Park, RMNP (CO): Also, easily broken down into two sections of about V10 minus, resulting in about V12 minus.



Eckagrata, Hueco Tanks (TX): Two set up moves of V5 to a V10 jump move. Still only V10.


Let's look at some more complicated examples:

Mo' Mojo, Hueco Tanks (TX): A V9 traverse, into a powerful three move V10 minus, into a tall V5 section to top the boulder out. The V5 is left out since it will not add anything, and we are left with 11.5, showing how this boulder problem seems to be consensus V11/12. It is important to note that a solid number such as 11 indicates the standard for the grade while 11.5 indicates a slash grade.



The Book of Bitter Aspects, Bradley (CT): Two set up moves of V8 or V9, into very unique transition move crux that comes in around V12, into a 5 move sequence of V8 to finish. 8.5 and 12 add up to 12.25. Here, since 12.5 and 8 are more than 4 apart, the last section is dropped and we are left with 12.25 or hard V12. .25 indicates a boulder problem that is on the harder side of the grade, while .75 actually indicates a low end problem of the next grade (i.e. 12.75 would be low end V13). While I was reluctant to downgrade The Book two grades, perhaps subsequent ascents will bring about consensus.



Left and Right El Jorge: An intro section of V6 splits in two lines, the Left turning into a V10 lock-off/bump crux and concluding with a two move V8 finish, while the Right coincidentally exits with a V7 lock-off/bump crux as well. 6 + 10 + 8 = 11.5 for Left El Jorge and 6 + 7 = 8.5 for Right El Jorge. This seems to explain why the Left was considered V12 for years before consensus settled at hard V11, and why people are reticent to downrate what could pass for an easy V9.





Furthermore, here are two upper-end examples that SEEM to support the theory:


Paul

Terremer, Hueco Tanks (TX): Starts by climbing Diaphanous Sea using the left hand holds for the right hand and vice versa (about V13), into Terre De Seine (a V13- move, into a V10 move, into a V4 finish). The 4 is eliminated, 13 + 10 = 13.5 or V13/14, 13.5 + 13 = 15.25 or mid range to hard end V15. Paul's estimate was slightly more conservative than Fred's and seems to be right around this. The fact that the crux move comes near the end of the problem seems to put it at around V15d.


The Story of Two Worlds, Cresciano (Switzerland): In one of the aforementioned posts, Jamie quotes Dave as describing this problem as the "New 8C Standard". Dave proposes that this problem combines a V13/14 start into an existing V14. Based on the equation, 13.5 + 14 = 15.75 or bottom end V16. Wow! Is it possible that Dave over-corrected and went too much in the other direction as he has done in the past with problems like Freshly Squeezed (originally V11) and Suspension of Disbelief (originally V13)? This problem has not seen a second ascent since it was put up in 2005.



Here is where things get a little controversial:


Nalle

Jade, RMNP (CO): One lock off set-up move and moving of the foot (about V7) leads to what people have labeled THE MOVE (V13), to three more moves of about V10/11, to a highball V4 finish. Here, since the 7 is dropped in conjunction with the 13, and the 4 is dropped in conjunction with the 10.5, we are left with 10.5 + 13 = 13.75, or bottom end V14. Uh oh. Perhaps fuel for the fire?



Right now you probably have several questions about the limitations of this theorem; let's examine the possible shortcomings of pigeon-holing the inexact science of bouldering grades into a single equation. First off, how do you account for the difference between a crux coming in the BEGINNING of a problem, as opposed to at the END? With all other factors being equal, the statistical difference between two such problems seems to weigh in at less than .25. To expand upon this, it appears that the key lies in the fact that pump-factor rarely comes into major play in bouldering; the vast majority of pure boulder problems do not cover more than 30 feet of climbing. Sure there are long roofs such as the Wheel of Life (Grampians, Australia) that have been given bouldering grades, but many have voiced that 5.15a seems to be a much more appropriate estimate of its difficulty than V16 (10 + 12 + 14 = 15.5 or hard V15 based on the equation). Many can attest that there are times when everything clicks and you get to the crux of your long-time project, and you execute the move as if you have not done any climbing up to that point. A personal example of this would be the Automator in RMNP (CO); while the crux move itself (V11), which comes at the end of the problem, took about 5 days to complete, when I got to it on the send I felt fresh and the move felt just like any other even though i had climbed a long V11 sequence into it.

How come adding a section that is four grades easier than another section does not increase the overall difficulty of the problem? It appears as if a four grade difference is a bit of a magic number. The reasoning is that if you are strong enough to climb a V10 section of a boulder problem, that a V6 section in the same problem will feel significantly easier and will not affect your perception of the grade. One prime example would be Ode to the Modern Man (Mt. Evans, CO); I am speculating, but it seems like this problem is one V11 move, into another V11 move, perhaps into a V10/11 move, into a V9 exit. 11 + 11 + 10.5 = 13.75 or bottom end V14. Very seldom do people fall off the V9 exit moves of Ode; if you can climb crimpy V14, crimpy V9 will not feel like much of a challenge.

Toshi

What about intangibles such as a move being not terribly hard but maybe just low percentage? After much thought, it seems that examples of this are fairly hard to find and are mostly negligible. In one notable example, the crux on Something from Nothing (Great Barrington, MA) after the break is an accuracy move weighing in at around V8. While it is easy to fall on this move, the problem feels closer to V10 than V11 when dialed. Perhaps a low percentage move like the one on Eden (Joe's Valley, UT), that cannot be tried without climbing the moves leading into it is more of a floating variable, but not by much (thanks JJ).

One definite limitation is the equation's possible inaccuracy with regard to the lower end of the V scale. While V3 and V3 do seem to add up to V5, anything lower doesn't seem to quite work. For instance, V2 + V2 should be easier than V4, and V0 + V0 cannot possibly add up to V2. Or can it? I will admit that it is easier for me to tell the difference between V9 and V10 than between V0 and V1 so it is hard for me to offer convincing arguments for or against this discrepancy. Also, I was not a math major, perhaps there is a way of scaling the equation to account for this?

Before you decry me as a grade-harping charlatan out to ruin the pureness of the sport, hear me out...

This is my attempt to bring about some impartiality and perhaps some accountability to assigning grades to boulder problems. Though as a whole climbers seem to have gotten considerably more "brave and humble" to quote the indomitable Jens Larssen of 8a, brazen examples of over-grading still exist. In the end, my hope is that this idea will actually REDUCE the amount of time we spend squabbling over grades, and allow us to focus on what is really important: THE QUALITY OF THE PROBLEM.


P.S. Though I have not been able to think of examples that strongly contradict the equation in my mental repertoire of boulder problems, it certainly does not mean that they don't exist. Give it your best shot! And I initially mentioned that I had TWO points to contribute to the debate...that was the FIRST one.

24 comments:

  1. http://www.australianbouldering.com/table.html

    ReplyDelete
  2. Correct me if im wrong, but i think shaken not stirred is supposed to be considered a boulder problem of 2 parts, a full v10(to the hueco) followed by a v7(big iron on his hip).

    (10+7)/2 + 2 = 10.5

    i couldnt really see that problem being downrated, much less to v10/11.

    None the less, it is a fascinating idea!

    ReplyDelete
  3. (thanks JJ) does not get you out of copyright infringement.

    you can pay me in vacation time. since this is not possible, you can pay me in beer?

    i'm thinking about installing a transamerican pipeline so that folks from the east and pacific northwest can simply tap my straw into their kegs for a constant flow.

    ReplyDelete
  4. @anon: ha! Amazing. Shades of Darwin and Lamarck, haha. I'm psyched to see that I'm not crazy! Well it's not groundbreaking I suppose, even though taking it down to more decimal places clears the picture a bit more. Why is this not talked about more?

    DISCUSS!

    @BA: I think saying it is V10 to get into Big Iron is a stretch...seems more like hard V11? And Big Iron always felt pretty hard for 7 points as well. Regardless, this seems comparable to wheel of life and more deserved of it's 5.14 grade rather than straight up V12.

    ReplyDelete
  5. Intriguing and well thought out. I wonder what percentage of the time this formula works. It seems like with the incredible ammount of information contained on 8a.nu, that we could glean more than who's where in the ranking, and actually start to make sense of the intangible grading system. Utilizing some of these principles, in conjunction with the consensus interests me...What about more detailed ways to log problems, other than grade, ascent style, stars, etc? Finally, a little more insight into the grading game, rather than, uh, feels like v8. Oh yeah? Why does it feel like v8...

    ReplyDelete
  6. I hope part two isn't as long winded. I want 12 minutes of my life back

    ReplyDelete
  7. Interesting to say that this will REDUCE the amount of time squabbling about grades, because what you have effectively done is split boulder problems into grades of 2 or more. This can only lead to 2 or more times the amount of squabbling...

    ReplyDelete
  8. And as far as Buzz is concerned...he told friends he didnt think it was v13, more like 10 or 11 if he really had to guess. However, he did not want to be the one to "officially" downgrade though in his head he already did. The only ones responsible for the "brazen" upgrade were those at Urban Climber wanting a hot story...

    ReplyDelete
  9. UC mag (or Buzz?) forgot to mention that he climbed a different varation of the problem, something that is obvious if you've seen both dosage 5 and this video. Buzz's variation has recently been repeated and is about V9/10. I'm not placing blame with one party or the other, but there should maybe be a little more fact checking when reporting something like this.

    ReplyDelete
  10. first of all, i think this is a very interesting idea and had fun thinking of examples that it works with (nearly everything)the only problem (or i guess question) i can think of is how can you take into account the nature (power vs endurance) of each section.

    looking at right martini, after climbing the v11 first part, one is going to be pumped and this v7 will feel very different from a v7 exit to a single move v11. the formula says an 11 into a 7 is still 11 (which in the case of a single move v11 i would agree with). however, in the case of something very pumpy like the v11 in right martini, that v7 is going to feel more like v9. if we take the grade of what big iron feels like after the first section then we end up with v12 but if take what big iron feels like by itself, then we end up with v11. so with that said, how do you decide when to take what it feels like in isolation and what it feels like on link?

    another example would be dark waters. if dark waters is 12 or 12/13 then we should end up with a number like 12.25 or 12.5. the first part (muddy waters) is considered 11/12. the stand to dark waters is considered 8. put that together ((11.5+8)/2)+2=11.5. so the question remains... do you take what dark water stand feels like on link (harder than 8) or what it feels like in isolation when computing the grade using your formula.

    ReplyDelete
  11. @Hayden: DW actually adds up to 11.75 or easy v12. Like I mentioned, longer boulder problems do confute the theory a bit, and most (like Right Martini) are easier to assign a route grade to. Dark Waters does seem to fit in that gray area where it's not quite long enough to warrant a route grade but it's just long enough where pump becomes a factor. Perhaps adding another variable for something like every 10 moves climbed? Hard to say...

    ReplyDelete
  12. Also, I think problems are generally graded based on how difficult the problem feels when you have it WIRED. It is easier to get one hard move wired than a long section of the same grade. Chances are you are not climbing a long section as efficiently as possible and more energy is expended.

    ReplyDelete
  13. Max, interesting post. Good work. My name is spelled Jamie.

    ReplyDelete
  14. FAIL. My bad, will fix as soon as soon as I have proper internet access. Thanks.

    ReplyDelete
  15. lets just say pump doesnt matter. long boulders/mini routes will just have mega stout ratings.
    how many v0 intro moves would it take to change the grade of a v9? doesn't matter. can i update my 8a now so that super slider gets 9 points?
    p.s. i'm liking the blog max. i'm glad we're facebook friends.

    ReplyDelete
  16. my equation is much easier.

    have i sent it yet?

    no = "v-not possible by humans" or, numerically, "v18"

    yes = "v7"

    why do you all try to complicate this formula?

    ReplyDelete
  17. Very interesting concept. The next big question is, of course, how do you determine the grade of a given section to begin with. I don't know if you can really remove subjectivity from the process completely--but perhaps i'm presaging your next post...?

    ReplyDelete
  18. I think the system still works with longer boulder problems. But takes a small amount of modification. For example, a boulder problem consisting of four V8 cruxes could be considered as follows. V8+V8=V10, then take that and V10+V8=V11 and then V11+V8= 11.5 (or soft V12). Notice that if you keep going adding V8's the limit approaches V12. Hence, I believe there is something a little more to the 4 grade in repsect to the easiest section of a boulder problem that can still advance the grade. Nonetheless something to consider.

    ReplyDelete
  19. "it is easier to get one hard move wired than a long section of the same grade"

    this is up to the climber's abilities really. A sport climber getting the occasional boulder day would probably get that long section wired to perfection in two or three tries, and possibly never do the single hard move...


    another thought.
    imho formulas like this are good for the very upper end of the scale where you don't have enough ascents to judge from.
    But for boulder problems having, say, 50+ ascents, it is more interesting to just look at the repeaters. Things like "who's the strongest who seriously tried the problem and failed", "who's the weakest who's done it", "how strong are those who can do this in a day", etc...

    the "look at the repeaters" kind of approach, in my opinion, tells us what we really want to know.
    Measuring "personal" progression doesn't need grades, because one could have an equally challenging time on an anti-style low-grade problem, and on a high-grade problem fitting her/his skills perfectly.
    To measure personal progression, well, the most accurate way is having your own classics and long-term projects and compare the experience over the years!
    We rather use grades to measure our progression thru the community, so the grade should tell us how of an hard/easy time the community has had on a piece of rock.

    ReplyDelete
  20. Maxim.

    It's been three months.

    More please.

    ReplyDelete
  21. My opinion is similar Matt Overhardt's.

    How do you grade a "section" of a boulder problem? Grades are assigned to problems and reflect how difficult it is to complete the problem from start to finish. You do not grade _moves_ and then combine those graded moves to conclude the grade for the problem. If you remove the top-out it is no longer a boulder problem and has no grade. If you don't do the start it is no longer a boulder problem and has no grade. You aren't climbing anything. You're doing moves - which do not have a grade.

    I would venture to say that your equation may only work for two already established boulder problems that can be climbed, start to finish, into each other. How often does that happen? For extensions, the bottom sections cannot be graded because they are not boulder problems and accordingly cannot receive a grade.


    Ingoring all I've said, has anyone tried to solve for X with Y and an already suggested Z? Redrum sit is great at V10. Stand is great at V7. This means the slap plus the matchy switchy parts is V9?

    Rave sit moves are V5? That one sounds about right. Velcro in Squamish puts the low extension at V12hard. Have to ask Tim or Sean.

    Also, I think it is silly to suggest Shaken not Stirred receive a route grade.

    ReplyDelete
  22. Some of these newer super-bright LED light products will light up on
    voltages ranging from 10 volts DC to 30 volts DC, while
    others will run on as little as 8 volts. Implementation of solar energy has
    environmental costs that may hamper its application on large-scale operations to reduce its impact.
    There's a color screen option which works the same way as the light bulb but has less of a harsh glare.

    Review my webpage: LED Deckenleuchten

    ReplyDelete