Max/Average Grade of Ticks for Crag by Climber: Example the Gunks

Disclaimer: I'm not good at stats, I didn't thoroughly test my code, there's probably a whole host wrong with how I went about this.  I was just curious, not looking to start a fight over how shitty my data is. 

tl;dr
Less than 7 percent of climbers who report ticks on MP average leading cleanly at 5.9 or above in the Gunks.
Less than 4 percent of climbers who report ticks on MP average leading cleanly at 10a or above in the Gunks. 
Less than 2 percent of climbers who report ticks on MP average leading cleanly at 10b or above in the Gunks. 

Roughly 20 percent of climbers who report ticks on MP have lead a 5.9 clean at the Gunks.
Less than 18 percent of climbers who report ticks on MP have lead a 10a clean at the Gunks
Less than 10 percent of climbers who report ticks on MP have lead a 10b clean at the Gunks

Link to raw-ish data/graphs

--------------------------------------------------

So this started with a conversation where I was musing what percent of climbers are climbing 5.9/5.10 in the Gunks (part of a larger conversation on breaking into said grades).  Is the notion that the gunks are sandbagged reflected in the tick data?

But I was curious: what percentage of climbers at the Gunks are at each level?  More specifically, how many are climbing at or above 5.9? 

The MP tick data is imperfect, but it's the best we got. So I pulled the data.  I first just pulled all ticks for every route.  The graph below shows the average grade ticked and max grade ticked.  The average was calculated by giving 5.0-5.14d a number value, multiplying this by number of routes at that grade ticked, summing this for all grades, and dividing that total by total number of ticks.

Plots: All Ticks at Gunks

I showed this to a friend, and they then wondered what would happen if you limited ticks to redpoints/flashes/onsights. There is a decent amount of toperoping and following going on.  The data is not standard, so I made some assumptions in separating ticks into categories:

- Clean leads: all ticks with "Lead" in them and do not include "Hung" (i.e. lead / onsight, lead / flash, lead / redpoint)
- Leads: all ticks with "Lead" and "Hung" in them (i.e. Lead / Hung/Fell)
- TR: everything else.

I took only those ticks in the "clean lead" bucket and replotted.

Plots: Clean Leads at Gunks

I'm going to expand on this eventually when I have freetime to compare across crags. But thought someone else might be interested in the meantime.

Some other interesting things:
- There are 4899 distinct profiles that have ticked a route in the Gunks
- There are 95445 ticks in the Gunks (averaging to 19.5 ticks/profile)
- There are 2391 distinct profiles that have ticked a clean lead in the Gunks
- There are 25114 clean lead ticks in the Gunks (averaging to 10.5 ticks/profile)

Edit: I realized that while ignoring slash grades is fine, ignoring plus grades is not. Updated to reflect.

Edit 2: Here's the Grade distribution at the gunks

Interesting work!

Here's what the Red looks like (I know this isn't a direct comparison, mixed sport/trad vs trad areas, but it was the next place I was interested in).

I would be very curious to see Yosemite Valley

The 12a Effect is real, and even bigger than I'd thought it would be!

Other scattered thoughts: 

• The other "a" grades at the Red get a boost relative to the surrounding grades, but it seems less pronounced than 12a. Lesson: if you don't want people to climb your route, rate it 11d.
• I bet 80% of the 11a ticks at the Gunks are Carbs and Caffeine.
• Same for Retribution and Nosedive for the 10b ticks
  • ETA: I took a look, and there are almost twice as many 3+ star 10bs than 10as at the Gunks. I think this accounts for the bump in the number of 10bs as compared to 10as for peoples' max grade. 
• ETA: Contrasted with the Red, where the "a" grades get a huge bump, the Gunks has a massive peak at 5.9 and then a big dropoff to 10a, and a slight recovery for 10b. Some guesses:
  • there's a mental barrier to climbing 5.10 on gear that many folks never get past. This is why you rarely have to wait in line for a 10. But once people get over the hump, they don't get stuck at 10a.
  • The lack of good 10as makes people feel like the gap to the next grade with lots of options is too great.
  • There are just a ton of awesome 5.9s at the Gunks (which is true). So people aren't focused on moving up to the next grade.
• Interesting dip at the 5.7 grade for max lead, all ticks in the Gunks. Is that just because there are relatively fewer classics at that grade (other than V3, Something Interesting, and CCK, there really aren't as many as compared to 5.6 and 5.8)?
• Can you pull the same graphs but for all ticks, not just max and average? I think that would capture the overall popularity of those grades a little more clearly. I wonder what percentage of traffic is on the hard versus easy routes at those crags.

Great work! Do the rest of the major areas!

Yosemite Valley

What's a tick?

mark kerns wrote: What's a tick?

A tick is a small, parasitic arachnid that can host viruses and bacteria that are harmful to humans. When recreating in an area known for having ticks, it is worth performing a complete check on one's body, to ensure nothing snuck on. The Gunks are more known for chiggers than ticks though, so I have no idea what the hell everybody is talking about.

@Nkane is this what you're talking about? This is the number of total ticks at each grade. Makes sense if I'm trying to gauge traffic

Would love to see one for the Bow Valley as well as Squamish for us Canadians!

Kalli Schumacher wrote: @Nkane is this what you're talking about? This is the number of total ticks at each grade. Makes sense if I'm trying to gauge traffic

Yes! And the data supports my hypotheses about 5.7 and 10a!

Thanks!

Bump for Smith or Index

Bow Valley

Squamish



Smith Rock



Pretty sure this is all correct? did it all pretty quickly... need to automate the middle jump or better yet just post the source code and y'all can run it   

Super cool, thank you very much. Now if there was only some magic way to know what percentile of the general climbing population one is in, not just within those that track climbs on MP. I feel like MP users who track ticks are going to skew to stronger climbers where your average 5.6 - 5.9 casual climber won't even know what MP is.

...But that's probably because I'm butthurt to know that I'm only in the top 15% for max grade of 11c in the Bow Valley :p

Do rumney and red rocks!

Kalli Schumacher wrote: Yosemite Valley

Look at all those people not sending Serenity/Sons!

wow this is really great. did you collect the data manually by scraping ticks? i didn't think their APIs had area ticks, though i would love that feature for this kind of analysis.

As a math & stats nerd, this makes me happy. Nice! 

Chris Blatchley wrote: wow this is really great. did you collect the data manually by scraping ticks? i didn't think their APIs had area ticks, though i would love that feature for this kind of analysis.

Yes, their API isn’t super useful. 

Kalli Schumacher wrote: Here's what the Red looks like (I know this isn't a direct comparison, mixed sport/trad vs trad areas, but it was the next place I was interested in).

Data might be skewed - RRG has online guide that allows to keep track of routes climbed. It is more complete than mproj, the interface is better as well. As a result, people who mostly climb at RRG might be more inclined to use RRG guide.
RRG grades in the online guide, and, most likely mproj, are consensus grades, as opposed to the typical FA grades that most areas have.

Luka Bogdanovic wrote: Super cool, thank you very much. Now if there was only some magic way to know what percentile of the general climbing population one is in, not just within those that track climbs on MP. I feel like MP users who track ticks are going to skew to stronger climbers where your average 5.6 - 5.9 casual climber won't even know what MP is.

Here's the alternative hypothesis:
The new climbers who are all excited and stoked are the ones eagerly ticking every route they do on MP whereas the more experienced, stronger climbers don't particularly care about ticking their routes on social media (though they likely keep track on their own). Hence we see lots of 5.6-5.9 ticks and significantly fewer above 10d.