What are the real odds of being killed by lightning, if you are outside in a field during a storm?

July 17, 2009 7:54 AM

What are the real odds of being killed by lightning, if you are outside in a field during a storm (i.e. factoring OUT all the people smart enough to stay inside)?

Hi folks,

I realize this is unanswerable with any precision, and perhaps even in ballpark figures, but thought I'd give it a go. I have a hard time convincing some of my teammates to stop playing when a thunderstorm approaches (or is overhead!). Invariably somebody points out that statistically you're very safe, with only a handful of people killed each year out of 300M in the US. I looked up the stats...

http://www.lightningsafety.noaa.gov/more.htm

... and the 30-year average is 62 deaths per year. So, sure, 62 out of 300 million are pretty good odds.

I looked at the locations for all the people killed in 2006-2009 at the site above, and sure enough they were all outside.

So really, the odds are not "62 in 300 million" they are "62 in X number of people unfortunate or stupid enough to be outside during a storm."

A long shot, I know, but is there any good way to ballpark X in a way that makes the "definite fun" vs. "possible death" tradeoff more ominous? Again, the conditions are standing out in a field, storm a mile away or closer (sometimes much closer).

Thanks!

Jim

posted by Jim Biancolo to science & nature (19 answers total)

I'd be interested in hearing answers to this, but as an added point: It's possible to be struck by lightning even if a storm is further than a mile. Indeed, it's possible to be struck even if the sky around you is sunny.

So your sample size may be a bit larger than just "people outside in a storm," if that helps at all. Maybe "people outside within N miles of a storm front".
posted by kaseijin at 8:00 AM on July 17, 2009

kaseijin, that's true, thanks, and I considered that. I'd be curious to hear that factored into the answer, but also, for my group in particular, folks have played with the storm literally overhead, so the size of X changes with each group's danger threshold (or lack thereof). So X is probably a much bigger number if you include people who stay outside on a sunny day with thunder just barely audible over the cloudless horizon, but much smaller for people who are willing to be outside, in a field, with the thunder booming and arcs of lightning visibly going sky to ground. Not that they're seeing the actual ground contact point, though. I think THAT might be enough to scare even this group indoors. :-)
posted by Jim Biancolo at 8:10 AM on July 17, 2009

If you are going to be logical about the odds, you should never drive to the ballpark. Even walking there would be riskier than playing in a thunderstorm. And never, never eat those hot dogs.
posted by weapons-grade pandemonium at 8:24 AM on July 17, 2009

I would be very surprised if you would be able to get anything approaching reliable data on a topic such as this - as kaseikin the boundaries of what counts as 'being in a storm' are hazy. You may need to think around this problem. It might help to ask, what size area of a storm has a high risk of being hit by lightening - I have no idea what the answer would be, however.
posted by munchbunch at 8:28 AM on July 17, 2009

You might get something approaching the answer by looking at the annual livestock deaths due to lightning, as generally cows, sheep, etc are not sheltered during thunderstorms. I couldn't find any good aggregated data, though, only stories of mass deaths due to cows being huddled up against wire fences that were then struck by lightning.
posted by jedicus at 8:49 AM on July 17, 2009

Check your local (and not so local) online weather outlets for lightning strike maps, like
this one for San Luis Obispo and the surrounding region. Lightning strikes are mapped
in real time, and you can see trends and directions.

Then it won't be a matter of conjecture. You can get indoors when your odds start increasing.
posted by the Real Dan at 8:51 AM on July 17, 2009

Pretty insightful question. Never thought about it, but you're right - there's a selection problem when trying to analyze the overall risk of electrocution since only certain types of people are going to be outside in the first place, and secondly, the relevant sample needs to be the exposure to treatment of lightning, which already assumes if nothing else people outside. Swimming pools, sports games, parties - they usually close up shop during an electrical storm, so it doesn't seem like you can use those events as your sample.

What about using a proxy first as a baseline? Maybe lightning rods? They are different from humans since they attract lightning, but maybe you could first think about the odds of a lightning strike on a lightning rod since they remain outside during a storm even when others are running inside. A second idea might be to use trees as the proxy, since they too are going to stay fixed in location and thus you eliminate the selection problem. You still have the problem of getting data, though.

Probably, a physicist is the person to ask. There's most likely some physical property associated with the storm that generates a strike on a certain location, and you simply extrapolate that to human beings. You definitely should not, though, use the naive estimate that you are criticizing, because you're right about the selection problems.
posted by scunning at 8:51 AM on July 17, 2009

Betcha a lot of those people killed are golfers. Metal pole, spikes in the ground...
posted by weapons-grade pandemonium at 8:55 AM on July 17, 2009

It only makes sense to decrease your risk if a storm is near. Sure the odds may be against getting struck, but betting wrong = death.

Statistics don't matter if you're the victim, or the victim's friends and family.
posted by The Deej at 9:11 AM on July 17, 2009

(BTW the above linked story just happened a few miles from me.)
posted by The Deej at 9:12 AM on July 17, 2009

Not everyone hit by lightening will be killed by it. I drove a pea combine as a summer job when I was a kid and witnessed ~6 direct lightening strikes, which generally threw the combine operators ~30 feet & knocked them out. They all woke up unharmed.
posted by torquemaniac at 9:13 AM on July 17, 2009

There was actually a Mythbusters that busted the metal spikes myth - the likelihood of being hit is equal with metal or plastic spikes:

http://www.tv.com/mythbusters/blind-driving/episode/1214874/summary.html

Now, holding a metal rod in the air (golf club, baseball bat, etc.) is another story...

Out of curiosity, which sport is this? Golf, baseball, soccer, something else?
posted by GJSchaller at 9:17 AM on July 17, 2009

From that NOAA site, there's a page by the American Meteorolgical Society:

It is common for people to refer to the chance of being struck by lightning as an improbable or unlikely event. Yet actual statistics say otherwise. Lightning strikes the ground approximately 25 million times each year in the United States (Orville and Huffines, 2001). Most people greatly underestimate the probability of being involved in a lightning strike. According to the National Weather Service, the chance of an individual in the United States being killed or inured during a given year is one in 240,000. Assuming an average life span of 80 years, a person's odds over their lifetime becomes one in 3000. Assuming the average person has ten family members and others with whom they are close, then the chances are one in 300 that a lightning strike will closely affect a person during their lifetime.

Although absolute personal protection from lightning cannot reasonably be achieved, following a set of simple guidelines can substantially reduce lightning casualties. The vast majority of lightning casualties can be avoided through improved public education. The public needs to be made aware of the magnitude of the lightning hazard and motivated to practice lightning safety.

In addition to the people killed by lightning, people get injured, sometimes permanently. There are some risks in life that are easy to avoid. Seriously, not worth the risk.
posted by theora55 at 9:20 AM on July 17, 2009 [3 favorites]

Start this from a completely different angle...

* Area of the earth inhabited by people that is not inside a building that meets your criteria -- outside in open field, etc. (X).
* Number of people within X at any given time.
* Area of X within lightning strike range at any given time (Y).
* Odds of a person within Y getting hit by lightning.
posted by Cool Papa Bell at 10:29 AM on July 17, 2009

The question seems kind of backwards to me, since it's not a purely random event. It's like saying, if only 1/100 people die a year because of gunshots in my town, I can stand in front of 99 bullets before I'll die. You're better served by determining the number of people struck by bullets who survive, much as your friends would be better served by determining the number of people who've tried to get hit by lightning, and their survival stats.
posted by nomisxid at 11:05 AM on July 17, 2009 [1 favorite]

@theora55, that's great, can't believe it was on the same site I got the stats from! So you start from a number of 1 in 3,000 rather than 60 in 300 million, and then it only gets worse when you filter out the people who stay inside. Yikes.
posted by Jim Biancolo at 12:06 PM on July 17, 2009

@GJSchaller: Ultimate Frisbee. This is just our local pickup group though, not reflective of a problem with the sport at large. The UPA has good lightning guidelines in place for tournaments, and in my experience they are rigorously enforced.
posted by Jim Biancolo at 12:25 PM on July 17, 2009

The NOAA link by theora55 calculates the probability of being struck by lightning. The original question was about being killed by lightning. The probability of being killed is much lower -- about one in 50,000 of dying from lightning vs some other cause.

The book above, Struck by Lightning" has some good stuff about varying lightning risks in different states and which is compared to other risks in life.
posted by JackFlash at 1:05 PM on July 17, 2009

@JackFlash, right, sorry, 1 in 3K for dying, 1 in 50K for getting hit. I assume though, that neither of those numbers correct for people who stay inside during storms. The lifetime odds for somebody who always goes to the middle of their house when the storm warnings start are presumably better, while those who run around outside are worse. I'd also like to amend my original question about "killed" to "killed or seriously injured."
posted by Jim Biancolo at 3:47 PM on July 17, 2009