• Enter your email address to follow this blog and receive notifications of new posts by email.

    Join 30 other followers

  • January 2016
    S M T W T F S
    « Dec   Feb »
     12
    3456789
    10111213141516
    17181920212223
    24252627282930
    31  
  • Recent Bookmarks:

  • Archives

  • Categories

Revisiting Cooper’s Jump

Much of the debate in the DB Cooper case has centered around whether he survived the jump. The weather wasn’t perfect, Cooper was jumping at night with borrowed equipment and without standard skydiving gear or clothing. I’ve previously looked at this problem and came to the conclusion that the odds were in Cooper’s favor even if he was an inexperienced skydiver. In my quest to understand the dynamics of Cooper jump, I looked at bailout situations faced by bomber crews during the Second World War. I believe WWII represents the best available population for analysis, as thousands of novice skydivers were forced to jump from stricken aircraft under extreme circumstances over unknown terrain, often at night, with unfamiliar equipment, in any weather situation and from a variety of aircraft speeds and configurations.

From the situations I read about, combat bailout jumps were almost never made under ideal conditions (stable aircraft going less than 150 mph at proper altitude). In fact, there are some amazing tales of survival, including people who survived being thrown out of exploding aircraft. Many jumps were made at very low altitudes, giving the flyers very little to no time to stabilize their bodies and pull the ripcord “by the book.” Training was very minimal and often, especially in the RAF, completely absent. Thus, there is a strong prima facie case for similarity between Cooper’s jump and bailout jumps done in WWII.

Re-examination of WWII parachuting

In my original paper, I back-engineered a method for estimating Cooper’s survival probability using the available WWII parachuting data, which amounted to comparing the number of airplanes lost (multiplied by the number of crew per aircraft) to the number of POWs held by the Axis powers. This gave an approximate survival rate of 80%. This process was done for several reasons, among them to save time, get a bigger sample size and because there wasn’t an easily accessible collection of complete data. Unfortunately, aggregation of data at this level allows for a lot of variance. Other confounds included not knowing the conditions surrounding every bailout situation, and the fact the number of POWs didn’t necessarily represent the true number of survivors.

What was really needed was raw data from a large pool of actual parachuting events aggregated by a third party. In the process of finishing the first paper, I found just such an index. It was nearly complete and would give a good comparison between night and day jumps. Most importantly, it was not based on survivor self-reports, the information was third party and independently researched. Overcoming survivor bias is, in fact, the main problem this data set solved. If you use only self-reported survival stories, you automatically skew the data toward survival as non-survivors can’t relate their experiences.

Further, the data is only from Denmark. Confining the events to a single geographic region helps control the number of variables; most importantly the civilian population of Denmark wasn’t inclined to murder allied airmen like those in Germany (there are some anecdotes about German civilians and SS troops executing downed allied airmen, however Freeman Dyson discounts this as a major survival factor in his book “Disturbing the Universe”). In essence, this data acts as a natural experiment to compare night and day jumps, and also gives us a healthy sample size from which to draw conclusions about Dan Cooper’s survival.

The results

In 66 nighttime bailout events involving 210 jumpers, there were 192 successful parachute deployments for a 91.4% survival rate. There were 6 possible no-pulls, i.e. bodies found with undeployed parachutes, and 4 apparent parachute mals. For the daytime airmen, we have very similar results. In 46 situations, there were 244 airmen leaving stricken aircraft with 235 survivors on the ground for a 96.3% survival rate. Again we find four apparent parachute mals, but only one possible no-pull. The 91.4% survival rate for RAF airmen jumping at night is significantly higher than the result from my initial study, and it gives more credence to the possibility of DB Cooper surviving his jump.

It needs to be stated that none of the no-pulls are confirmable. There was no record anywhere in this sample of a group of men leaving an aircraft together only to discover someone didn’t get a handle on their ripcord. There was a total of 454 jumpers between both groups, so the fact we have not one confirmable no-pull should be a significant indicator of how rare no-pulls must have been. There were tens of thousands of bailouts during WWII, so it’s a near-certainty that some no-pulls happened. But I can’t find enough information to get a baseline on just how often they occurred. Even using the number of possible no-pulls among the RAF in Denmark as a baseline for inexperienced night jumpers, we’re looking at a rate of less than a 3%.

I ran a simple t-test comparing the US daytime and RAF nighttime samples; at a 95% confidence interval, the two samples are significantly different, but by an extremely thin margin. The difference between the two datasets was 4.88%, which is just a little higher than the 4.36% needed for significance. We can’t be entirely sure of the cause of the difference between the RAF night jumps and the American daytime jumps. Not only did RAF bomber crews operate at night, they also had smaller escape hatches than American aircraft, and RAF crews often had to attach their parachutes before jumping, giving them less time to prepare for the actual bailout. Regardless, this data confirms the efficacy of the parachute as a life-saving device, and that amateurs and inexperienced persons can use this device with a high rate of success even under difficult circumstances.

Examination of No Pulls in Skydiving

Anecdotally, No-pulls are a common occurrence in sport skydiving. You can find self-reported occurences and even videos with a simple search engine query. Skydivers even have equipment that automatically deploys their parachute if they haven’t pulled their ripcord (or thrown their pilot chute) by a certain altitude. Skydivers take it as a fact of their sport and have equipment and protocols to prevent fatalities caused by this phenomena. My interest is whether a no-pull represents an inherent risk of parachuting in general, or if it is a byproduct of the sport of skydiving. To figure things out, I looked at the details behing all the no-pull fatalities recorded in the Skydiving Fatalities Database.

Results:

There were 833 lethal parachuting incidents catalogued in the Skydiving Fatalities Database between 1995 and 2009. Of these, 107 are classified as no-pulls. However, even among these, only 48 fatalities appear to be directly applicable to Cooper’s jump. The majority of the recorded no-pulls involved demo-jumps, group formations, camera work, health problems, suicides, or other situations that Cooper would not have been dealing with during his jump. Thus, we can safely assume that most of the no-pulls catalogued in the Skydiving Fatalities Database, and thus most no-pull occurences from the world of skydiving, are caused by either the stunts and activities innate to the sport of skydiving, or by factors related to the mental or physical health of the skydiver.

Assuming Cooper was not suicidal, and there’s no evidence he was, and assuming he didn’t have a heart attack during the jump (which would have been a real possibility for a middle-aged smoker in 1971), this greatly reduces the possibility of a no-pull situation during Norjak. When we combine this information with the WWII parachuting data, it becomes clear that DB Cooper had a very good chance to survive his jump. However, the database did show that six of the no-pull fatalities happened when borrowed or rented equipment was used, and one individual died while jumping at night; both of these risk factors were present during Cooper’s jump but the represent a small minority of the total fatalities profiled in the database.

It will always be difficult to infer the true survivability of Cooper’s jump since we do not know the extent of his skydiving experience, the state of his equipment, the configuration of the money and how it was attached to his harness, his mental state and his health. We can say with some certainty that the weather and the lack of daylight had very little effect on his survival. In all likelihood, from the broad perspective presented here, we can be about 90% confident Cooper pulled the ripcord on his parachute and it deployed properly. From there, we can only make educated guesses whether he was hurt on landing, or whether he lost his life by drowning in the Columbia.

Advertisements
%d bloggers like this: