This commentary is by Simon Testa, a retired geologist and self-described data nerd who lives in Boulder. He’s an avid Flatiron scrambler and the author of “Best Flatiron Scrambles.”
A neighborhood friend took me scrambling in the Flatirons for the first time in June 2016.
I was hooked. The fast and efficient nature of moving unroped on the Flatirons provides a freedom that is captivating and allows one to link up multiple routes in a single outing.
Scrambling exists in the gray area between walking (Class 1) and technical rock climbing (Class 5). Traditionally, it refers to movement over terrain rated Class 2, 3 or 4. But in the Flatirons, the term takes on a unique meaning. Here, “scrambling” often refers to ropeless ascents of Class 5 routes, which, by definition, is free soloing. Despite this distinction, I’ll continue using the term “scrambling” in keeping with local vernacular.
The week of Dec. 16, 2024, was a grim one in the Flatirons. Two scramblers lost their lives.
Whenever a bad accident or fatality from scrambling occurs in the Flatirons, I find myself reflecting on the risk of this activity I enjoy so much. It’s important to distinguish risk — the probability of a negative outcome — from consequence, the actual outcome. The potential consequences of Flatiron scrambling are extreme, including serious injury or death. But how risky is it really? In other words, what is the probability of a fatal outcome while scrambling in the Flatirons?
It’s a question I’ve often pondered. I was intrigued to see if understanding the risk better would have an effect on my relationship with Flatiron scrambling.
A century of scrambling in the Flatirons
People have been scrambling in the Flatirons for over 100 years, but its popularity has surged in recent years. As of December 2024, I’ve sold 2,410 copies of my “Best Flatiron Scrambles” guidebook, and it’s not uncommon to see 10 or more people scrambling Freeway (Class 4) on the Second Flatiron at the same time.
How often do Flatiron scrambling fatalities occur? I identified 16 scrambling-related fatalities — eight since 2002 and eight earlier:
- First Flatiron: 4 fatalities
- Second Flatiron: 4 fatalities
- Third Flatiron: 7 fatalities
- Unroped descent after climbing the Third Flatiron: 1 fatality
All but one were men. Their ages ranged from 17 to 49, with an average of 28.
How risky is scrambling compared to other activities?
To put Flatiron scrambling in perspective, I examined the mortality rates of two other activities: driving and BASE jumping.
- Driving (2022, U.S.): The risk of death per trip is 1 in 5 million.
- BASE jumping (1995–2005, Norway): The risk of death per jump is about 1 in 2,000.
Driving is an everyday activity most people hardly think about, even though it carries inherent risk. BASE jumping, on the other hand, is widely regarded as one of the most dangerous recreational activities.
Estimating the mortality risk of scrambling in the Flatirons requires two key data points:
- The estimated number of scrambles per year
- Using Strava data, I estimated between 21,000 and 52,000 scrambles annually.
- The number of fatalities per year
- Since 2002, there have been 8 identified fatalities (0.35 per year).
- Since 2020, there have been 3 fatalities (0.6 per year) — suggesting an increase as the activity has become more popular.
Using these figures, I calculated an estimated mortality risk per scramble:
- Lower estimate: 1 in about 150,000
- Upper estimate: 1 in about 35,000
How does Flatiron scrambling compare?
- Scrambling is 33 to 143 times riskier than driving in the U.S.
- Scrambling is 18 to 75 times less risky than BASE jumping.
Cumulative risk: How much have I gambled?
I’ve been scrambling for eight and a half years, covering 2,253 routes and 1.1 million feet “on rock.” This is a lot of scrambling. Although each individual scramble carries the same individual risk, the cumulative probability adds up over time.
By my calculations, my cumulative probability of death over this period was:
- Lower estimate: 1.5%
- Upper estimate: 6.3%
Put another way, if I continue at my current rate — 265 scrambles per year — my annual probability of death from scrambling is between 0.18% and 0.77%.
To compare, the average mortality rate for a 53-year-old U.S. male (my age) is 0.7% per year (CDC WONDER database, 2018–2023). That means:
- Flatiron scrambling alone adds 24% to 110% to my baseline mortality risk.
- Scrambling is riskier than heart disease (0.15% annual mortality) and cancer (0.12%) for a man my age.
Reflections on risk and perception
Psychometric research has shown that people assess risk more by experience and emotion than by probability. This analysis was an attempt to quantify what I intuitively knew but never calculated.
Before doing this, I underestimated the risk of Flatiron scrambling. Like many who do high-risk activities, I may have been experiencing a form of cognitive dissonance — believing Flatiron scrambling was safer than it was because I enjoy it so much and have done it so many times without incident. But now, I see the numbers differently.
I initially assumed that fatalities would be skewed toward less-experienced scramblers. But seeing the historical data, I’ve begun to think more about cumulative risk and the complacency factor among experienced scramblers.
Will I stop scrambling? No.
Like most high-consequence activities, it’s a risk-reward calculation. For me, the benefits of Flatiron scrambling still outweigh the risks. However, this process has changed my perspective. I plan to take two key precautions going forward:
- Avoid complacency. Before starting any scramble, I will pause and remind myself that experience does not make me immune to risk.
- Start wearing a helmet. While it won’t prevent a fatal fall, it could protect me in situations where a head injury would otherwise turn a recoverable mistake into a tragedy.
Note: These are estimates, and the actual mortality rate is unknowable based on the available data. Many assumptions were made using incomplete information.
Well done piece. One thought on it: the fatality data you’ve unearthed includes a number of people who are / were inexperienced non-climbers. Seven of the 16 fatalities have unknown or no climbing experience. It’s possible (maybe likely) that all seven of these had no experience. I’ve been soloing the Flatirons since 1998 and recall several deaths of people who were not climbers at all, but just kids out messing around.
If you were to to run these numbers using only data from experienced climbers you might see that risk of death from scrambling among that group (who likely take many more trips up the Flatirons each year than inexperienced or no experience people) is actually a good bit lower.
This would be an interesting bit of data. In other risk sports (BASE, backcountry skiing and high-altitude mountaineering, for example) the risk of death seems to increase with experience. Could it be that scrambling the Flatirons is different?
Before looking at the data I thought that likely both incidents and fatalities would be strongly skewed towards the inexperienced (the cognitive dissonance I mention). However, as Phil Berggren mentions, the more experienced one becomes the higher the cumulative probability of an incident.
https://www.climbing.com/places/boulder-flatirons-scrambling-risk-assessment/ Breaks fatalities down by experience and notes that 75% involved people with at least scrambling experience. (You could argue that scrambling experience isn’t meaningful, but then you go down a rabbit hole of where to draw the line… 5.10+ with regular outdoor experience? Etc)
The argument that the risk in certain activities is reduced by a higher experience level in the participant is tempered by the fact that the more one engages in that activity, the more one is exposed to risk.
When scrambling the Flatirons I’ve looked around at the base of the climbs and noticed a lot of items that appear to have been dropped hidden in the bushes, like water bottles, etc. Very disconcerting to be sure. A helmet might help but that depends. Of course we’ll probably never know how many people met their demise by having something dropped on them.
I have personally witnessed a phone being dropped (on the Second Flatiron) and have almost been hit in the head by a softball sized rock dislodged from a ledge by a crow (on the Southern Goose Egg). A helmet would be very beneficial in similar scenarios. One of the great benefits of scrambling is that there is a low financial cost of entry (just need footwear with good grip (ideally climbing shoes or approach shoes with sticky rubber)). A helmet is an additional cost — but IMO worth it. A helmet would also be useful if one fell in the first 20 feet or so, or if one fell onto a ledge. Like most things, wearing a helmet, is a personal choice. I have started wearing one and truly hardly notice it on my head.
Fantastic analysis thanks for crunching the numbers on this.
Being an inexperienced climber, I often go alone, though it’s been a few years. My favorite scramble is the 2nd flatiron, so I appreciate your article and its warning as well as the reminder to get back out there and go slowly.
The 2nd is one of my favorites too. It’s an absolute classic. It is great that the trail goes right by the start and the end of the scramble, it’s open all year round, there are many options on where to go without getting too technical, and of course the jump is exhilarating/fun.
Yes, sooo much fun!! And the trail is convenient, especially when traveling with people who don’t want to scramble.
I’ve been a member of Rocky Mountain Rescue group since1965, and I know of at least 3 fatalities not in your list. RMR would have rather complete data on this.
Thanks Chuck, the Boulder community is extremely luck to have the Rocky Mountain Rescue Group (RMRG) and people like yourself that volunteer to be part of such an incredible organization. All my royalties from my scrambling guide book go to RMRG. I attempted to limit the data to “scrambling” (un-roped) related incidents. Do you happen to know if the additional fatalities you know of were related to un-roped ascents. My research was based on publicly available data (that I could find). Agree RMRG must has the gold standard when it comes to data. I am happy to analyze the RMRG data if they are willing to share it.
I’m gonna agree with Jeb about a more experienced participant having a far lower risk (at least until/unless they begin to push their risk tolerance and decide climb harder lines unroped).
As for greater frequency leading to a higher cumulative risk, that’s obviously true to a certain extent, but you also gain both familiarity and confidence, likely have rehearsed any crux moves (or determined a route to avoid them), and very likely have more specialized footwear than an inexperienced scrambler, so I think the advantages to an experienced scrambler are many. And if you’ve come from any sort of technical climbing background, I’d also guess those movement and footwork skills transfer in such a way that broad statistical estimates of risk probabilities are, idk, somewhat inaccurate… (hopefully =)
How do you statistically quantify the probability of an inexperienced person choosing a poor route? Or of relying on a rotten or fractured hold? And r of their stress level or anxiousness impacting their movement or their decision making? I’d expect that such difficult to control for factors have a huge impact in the actual risks people individually experience out there, but I also tell myself good stories…
See you out there, and Simon let’s get out!
Matt – pre-research this was definitely also my point of view. The more I contemplate it though, the more I see complacency (of an experience scrambler) being the biggest risk. Without a much larger and more rich dataset however, it is difficult to truly know. And to your last point……yes, let’s get out soon……
The more you scramble/free solo; the more likely you are to have a hold break. I know that not counting falls on projects; a good share of falls I have had were due to broken holds.
The all volunteer Rocky Mountain Rescue Group is an invaluable service to the Flatiron Scrambling community (as well as to the broader community of course). Please refer to the link below to their donation page.
https://www.rockymountainrescue.org/contribute
Nice read! What was your method for the volume estimates with Strava out of curiosity. In the flatties it feels like experience/familiarity with a route is also valuable for safety, maybe more than climbing strength or general roped climbing experience. Any thoughts on risk correlation with route difficulty too? For the differences in scramble volume, Freeway (fortunately) feels very under represented in the data set
This is a great question. I looked at the Strava segment data for what I believe to be the 10 most frequently scrambled routes. Strava indicated 67,543 recorded ascents (note: Freeway on the Second, Direct East Face on the First, and the East Face of the Third, not surprisingly made up 76.5% of the ascents). Then I made a table (lots of copy and pasting) of all the Strava users’ (n= 717) fastest times on the Direct East Face and plotted ascents per year. The data stabilized in 2020. The data from 2020 onwards represented 70% of the users. I used this 70% as a proxy for the 9 other popular routes. This reduced the 67,543 to 47,280. As there are 1,834 days from January 1, 2020 (the beginning of the number stabilization of scrambles) through January 8, 2025 (the latest date in the Direct East Face dataset) this calculates to 25.8 scrambles per day. As these data capture the most popular routes scrambled in the Flatirons but not all the routes, I estimated the above selected Strava segments represent 90% of the scrambles (there about another 10 flatiron scrambles that are routinely scrambled, but each of them are likely only around 1% or less of the total). Therefore, I adjusted 25.8 scrambles per day to 28.7 scrambles per day. Additionally, of course, not all scramble ascents are recorded on Strava. The percentage of Flatiron scrambles recorded on Strava is the largest unknown in this analysis. As such, I used an estimated lower (20%) and upper (50%) limit. At 20% the 28.7 scrambles per day would factor to 143.5 scrambles per day (or 52,377.5 per year – I rounded to 52,000). At 50% the 28.7 scrambles per day would factor to 57.4 scrambles per day (or 20,951 per year – I rounded to 21,000). Intuitively, an average of 143.5 scrambles a day (over 365 days) seems too high, but it is included in the analysis to provide a low estimate of Flatiron scrambling mortality risk.
Hi Simon,
Great writeup, thank you for the effort, time and thought you’ve poured into this project, it shows.
On the issue of extrapolating population statistics from a skewed sample, it is problematic to guess how representative Strava ascents are of total solo attempts. Your estimated 20-50% limits are sensible, I would argue it might be closer to 5-30% (seasonality, CU population depending). Based on anecdotal evidence, I think 143 scrambles a day, averaged over the year is not too high; especially across all terrain features people attempt to solo in the Flatirons.
As other commenters mentioned, climbing proficiency, route selection, route reading, situation awareness and general preparedness all correlate to experience and personal risk appetite in different ways. The average Boulderite Strava user tracking their pre/post work romp up the Flatirons is likely very different in these observed and unobservable characteristics, compared to the average individual soloing in non Strava-logged attempts.
I would love to expand upon this work by looking at all accidents, as life-changing, non-fatal injuries should be appropriated weighed in risk analysis of soloing.
Your conclusions on complacency and cumulative risk are well taken, gravity always wins. Thank you for reminding me of this fact.
Ben – I appreciate your comments. I have tabular data on the 65 incidents I found using public sources since 2001 that I am happy to share. 50% of these involved a serious fall or fatality. Of course the 65 incidents, like the Strava data, only represent a fraction of the actual incidents. My analysis would have been greatly improved if I had access to RMRG incident data and tabular data for all Strava individual outings (rather than just users’ fastest times outings). I went back and forth on the percent range to use for Strava data. I am hoping your suggested 5% lower limit is more representative as this would lower the risk. Before using the Strava data my guess was that on average for 365 days there would be 50 ascents per day. I was just on Freeway and there were a lot of folks on the rock on this beautiful evening! It would be very interesting to know the true annualized number.
Of note: 50% of the recent fatalities (since 2002) where on the Second Flatiron. Freeway represented 51% of the Strava ascents for the routes I selected as the 10 most popular. I also have data on the 65 incidents (stuck scramble, fallen scrambler, fatality) since 2001 (this was the most time consuming part of the research). Of the 65 incidents, 34 (52%) were on the Second Flatiron. These numbers make intuitive sense to me.
Agree that familiarity is a risk mitigant. We must try our best, however, to not let familiarity breed complacency. Regarding route difficulty, there is just not enough data. Of course, folks “scrambling” 5.6 are likely also experienced. There are for sure many more ascents of Freeway (Class 4/5.0) by inexperienced individuals, but 75% of recent Second Flatiron fatalities involved individuals with know rock climbing and/or Flatiron scrambling experience.
Great work!
I would estimate that “complacency” is the greatest risk factor for the experienced climber or scrambler, while “getting in over your head” is the biggest risk for the inexperienced, these being two different groups.
And “driving to the trailhead” is always up there! Every five years more of us are killed in our own cars than died in WW l and WW ll combined.
100% Agree.
And yet, your data seems to imply that scrambling IS more dangerous than driving. I wonder if you have a sense of how many scrambles are by “experienced” climbers versus how many by those with under say 5 years of climbing. And then, how many deaths are in each group.
Yes, correct —- more dangerous than driving.
That would be great data to have. Unfortunately, I don’t have it.
The more you free solo/ scramble; the more likely you are to fall because of a broken holds. I have had multiple holds break that I had used a hundred times before in “scrambles/ solos”. That to me is one of the larger risks of complacency.
RMRG Instagram post today: The RMRG team evacuated a fallen scrambler from the sunset flatironette behind the 1st flatiron last night. Coming into summer, we love to see climbers continuing to test their holds and wear a helmet.
Ihttps://www.instagram.com/p/DIU8U2ZSKsl/?img_index=1
Your table of fatalities includes Franklin Lankford in 1968. I was an undergraduate at CU at the time. Frank was a classmate and acquaintance of mine. While I have been unable to verify this with a short search of contemporary on-line records, I believe his death on the 3rd Flatiron was reported as a suicide at the time. If true, does this distort your statistics?
Thanks for the comment Trace. Yes — this would distort the results a bit (reduce the risk). My source for this incident is an American Alpine Club Accident report from 1968 (see link below). It does not mention suicide but does mention was he was depressed. As suicide was not definitive I included Franklin’s passing.
Unfortunately, there are many unknowns and assumptions in the analysis.
https://publications.americanalpineclub.org/articles/13196800602/Colorado-Third-Flatiron
The more you free solo/ scramble; the more likely you are to fall because of a broken holds. I have had multiple holds break that I had used a hundred times before in “scrambles/ solos”. That to me is one of the larger risks of complacency.
Yep – cumulative probability applies to all risks, and agree broken holds is a common one. I have also had hand and footholds I have previously used break while scrambling. I wrote a different variant of this analysis (for Climbing.com) where I mentioned some common risks:
Below are some of the major risks of scrambling, as well as a possible solution for at least partially mitigating that risk:
Hand or footholds breaking: Test holds before committing weight
Rock or other objects falling: Wear a helmet
Weather changes. Check forecasts; abort or call 911 for rescue if weather worsens.
Scrambling off-route into more difficult terrain: Study route descriptions; go with or follow experienced scramblers.
Complacency: Pause before each outing to recognize this risk.
I’ve done the third Flatiron roped up, age30. At my level, the trip up and free rappelle took around 2-3 hours, no rush. Due to possible factors like rain and the possibility of other climber distractions, I felt unroped climbing was too dangerous for my skill level. Although I was not muscle bound, neither was I extremely dextrous. If you are experienced and in good health, its likely a reasonable risk.
Working the stats and numbers has little to do with this high risk activity.. gravity will win always. No matter how experienced or how focused you are it will not compensate for that brief moment when for no reason you simply lose your stance.. on the most simple roped climb a hold pops or a foot slips and away I would go… Thxs to my belayer I still can climb tomorrow
As a roped, occasional third flatiron climber I have had many scramblers pass me. Most have been polite.
Two of them climbed only a few feet directly above me and my simul-climbing child. That particular experience was terrifying until they got far enough above and to the side of us to no longer be a free fall threat to us.
I wonder how long until a scrambling accident results in collision that injures others — scramblers or roped climbers.
Those terrifying moments have been enough to put me off of roped climbing the third ever again.
“But in the Flatirons, the term takes on a unique meaning. Here, ‘scrambling’ often refers to ropeless ascents of Class 5 routes, which, by definition, is free soloing. Despite this distinction, I’ll continue using the term ‘scrambling’ in keeping with local vernacular.”
Perhaps it is time to push back against this misuse of the terminology. Such choice of language reflects the widespread ego boost climbers and peakbaggers seem to get by minimizing fall exposure and other risks in their descriptions. This is endemic on SummitPost, for example.
It appears that around half of the deaths you include are inexperienced or never-ever climbers. “Scrambling” sounds like (and – used as usual – largely is) something you can do without experience, by relying on wits and agility, risking mainly scrapes or at worst a broken bone. To the extent that, for example, CU undergraduates and other recent transplants hear or read descriptions of these Flatiron ascents described as scrambling, they will underestimate them and put themselves at high risk. Arguably, then, some of the blame for these injuries and death falls to those who publicly minimize the risk by perpetuating this terminology.
I especially agree with DaveJs comment about your use of the word “scramble.” I climbed the Direct Route (5.6) on the First several times with one of the original Stonemasters from SoCal (Peter Steres), and he asked me what is the difference between 5.6 and 5.12? Then he said, “There is none.” “If you fall, UR dead.” I would not call the first two/three pitches of the First a scramble in any sense of the word. The truth is that some people don’t mind risking their own lives. Just don’t try getting it on me.
This is an incredibly well-written piece! I’m a high school student in Colorado currently writing a research paper (more specifically, an ethnography) on free-soloing culture, and until now, I didn’t realize how often free-soloists frequented the Flatirons. Mr. Testa, would it be possible for me to interview you either in person or virtually? In fact, I’d like to extend the offer to any knowledgeable Boulder climbers who see this as I’m looking to gain more personal perspectives on the sport. Please let me know, and thank you for writing such helpful pieces!
Hi Simon,
I wanted to let you know that the one female fatality on 5/10 1986 was Robin Bartel. I was with her that day. She was photographing me and my partner Eric Doub who were on Northwest Passage on the North side of the 3rd flatiron. She was in the gully to North of the flatiron. She lost her balance while helping a dog down the gully. She was not an experienced climber. Please add her name.