Sharks, Beds, and Death Statistics

Per this article, shark attacks on swimmers in California happen at the rate of 1 per 738 million beach visits. So when I go swimming at the beach, I have a 1 in 738 million chance of being attacked that day, regardless of how many beach visits I make. Right?

Per this article, the chances of dying by falling out of bed or over a piece of furniture are 1 in 4,238 deaths. So when I die, there is a one in 4,238 chance my death was caused by a furniture fall. Right?

So how to you compare these statistics set up in different ways? On the face of it, they both seem like deaths by activity, but it’s not deaths per furniture/bed visits, or deaths by shark attack, so is there a way to numerically compare these two ideas?

I’m not unaware of the fact that 738 million is actually a very big number on the human scale. Especially considering that your lifetime will probably be on the order of only 27 thousand days, so you would need to “visit” your furniture about 26 thousand times a day to get to shark attack numbers.

Are sharks 26 thousand times safer than beds?

So, math gurus, is there a way to reconcile these “chances of death” numbers. To my math challenged brain, it seems like fractions where I need to find a common denominator or something.

No. When you die, there is a 100% chance that your death was caused by your cause of death. That statement is not really so meaningful when phrased like that. As the article says, shark statistics changed because of more people going into the water, even though the number of attacks went up. Which are you? One of those divers who’s been going into the water daily since the days of Gidget, or someone who just moved to the west coast and goes to the beach only on Labor Day?

The statistic about furniture means that if you randomly pull 4,238 death certificates out of some theoretical collection of all death certificates in the country (or all the countries from which the data was gathered), one of them is highly likely to be from falling over furniture. It’s not the same thing as saying, “There is a one in 4,238 chance my death was caused by a furniture fall,” even though the media like to phrase it that way. That kind of phrasing is an abstraction. Only if your death certificate is being randomly chosen from all death certificates does it make any sense to say that.

Yeah, in general the two statistics in comparison mean that “One is more likely to die by tripping over furniture,” but that’s based on certain assumptions that you can make about most people. If you surf every day and live on boat with hardly any furniture the statistics might not be very meaningful for you as an individual.

I am math challenged as well, so this is how I figure these things out. It may seem convoluted, but it requires no math.

Statistics are typically recorded after the fact. Otherwise they are an estimation of potential in a case with every variable included. Seeing information about what has happened doesn’t necessarily mean it will happen. There are so many factors and a combination of factors that lead up to a death scenario, if you change one small factor by being vigilant about anything risky, for instance, using furniture anti-tip devices or swimming somewhere sharks are not known to frequent, you completely change the statistics entirely by removing a variable that multiplies the chance of a given outcome. If car accident statistics say 1 in xx,xxxx chance of death, you remove yourself from that group by having a car with better safety, driving a bit slower, wearing your safety belt, and put yourself in a larger/less risk group, of 1 in xxx,xxx or potentially greater. Little changes multiply to a lot. This is my reasoning by why I am skeptical of most statistics since they are all encompassing and blanketing and typically fail to include small variables.

Hope that made sense, I just ate a ton of food and I am ‘food drunk’/sleepy.

I’d be willing to bet that the majority of the furniture-related deaths are from elderly people, probably suffering from dementia of some sort, rolling out of bed. Beds in hospices are designed to minimize this. They don’t have rails on the sides anymore, but the middle of the mattresses are lower than the edges. Yet their occupants still manage to occasionally roll out of them.

Ye…es, but only if the characteristics of your visit represent the average of the sampled group - some beaches will be safer than others; some visitors will have exposed themselves to more risk than others - if your visit is similar to the average of all those, then I think it’s fair to say you have the 1 in 738 million risk.

If you visit the beach and your activity consists of eating ice cream and making sandcastles at a location where sharks have never been seen, your risk is lower than 1 in 738 million. If you spend the day surfing, at a beach where sharks are commonly seen, your risk is higher.

I am reminded of the French professor of statistics who was arrested, back last century, when he was found carrying a bomb onto a plane. When asked why, he explained that statistically, while the probability of there being a bomb on the plane was small; the probability of there being two bombs on the same plane was minuscule, so he took one to reduce his chances of being blown up

The best plan is to sleep with a shark, in that case. What could go wrong?

In order to find the shark data they put the number of shark deaths as the numerator and the number of beach visits as the denominator.
In order to make a similar ratio for furniture you need to know the number of deaths, the percentage of deaths only helps if you know the total number of deaths. Once you know the number of furniture deaths that becomes the numerator then the denominator is the number of people in the country time 365.

eburacum45 sleeps with the fishes.

Actually, the way it’s phrased, “beach visits” are both the numerator and denominator so each visit would increase your risk. The denominator is 738 million visits. The numerator increases by one for each individual visit, regardless of who makes it. So each of your visits increases the numerator as well. One visit = 1 in 738 million, 2 visits = 2 in 738 million, and so on.

Everything else you point out is absolutely on the mark.

It’s a silly stat.