The Scented Llama of Doom Best not to ask really...


Insufficient Pessimism

People fear worst-case scenarios. Or, more accurately, people fear the worst-case scenarios for the things that they fear. If you’re afraid of public speaking you’ll imagine being laughed at by your audience as you stumble haphazardly through a poorly written speech. If you’re afraid of sharks you’ll imagine being torn to pieces the instant you fall off a boat. These fears are strong enough to change our behaviour - we’ll decline to give that speech or we’ll refuse to get on the boat. This is because our brains, sadly, are really bad at calculating risk. As school children we’ve given terrible speeches in front of the cruelest audience possible (our peers) and the result was never as bad as we imagined it to be. Scuba divers regularly swim with sharks, in fact some go looking for sharks, and they don’t get eaten.

A worst-case scenario always depends on a lot of things going wrong at once. Consider that you first have to get on a boat and then fall off in the vicinity of a shark and that shark has to bite you and that bite would need to be fatal - you can see how many variables are at play in just one scenario. (In the case of a speech consider that out of all the mistakes you could make you’d need to find a really funny one to be laughed at and then consider how difficult it is to intentionally make people laugh at you). Mathematically speaking we need to multiply the probability of each event to get the overall probability of the worst-case occurring and each time you multiply the chances get smaller and smaller. This is why worst case scenarios almost never happen - they’re highly improbably by their very nature. To some degree we know this - we realise that people giving speeches are rarely laughed off stage and if sharks ate everyone who swam in the ocean chances are the seaside would stop being a popular holiday destination.

Our brains are terrible at calculating these probabilities so we assign more weight to the outlier (for example: getting eaten by a shark is so rare that cows kill more people per year than sharks. Cows).

Great White Shark
Not as dangerous as a cow.

The good news then is that we have much less to fear. The bad news is we think in exactly the same way about best-case scenarios.

The reason for the shark theme is that my family and I have recently returned from a seaside holiday. I’m happy to say that, despite trying quite hard, none of us were eaten by sharks. We did think of dangling Two Year Old off a pier to attract one but my wife wasn’t keen on the idea.

We decided we wanted to leave really early to avoid traffic so we woke up at 02:30 with the intention of leaving at 03:00. We had this intention despite the fact that in 8 years we have never managed to mobilise our family in half an hour and that was before Two Year Old wrecked the previous average. We ended up doing really well and leaving at approximately 03:30 which was probably a record.

Let that sink in - our initial goal was so laughably wrong as to be half of our record time.

To quote Harry Potter:

Well…” Harry said slowly. He tried to organise his thoughts. How could he explain himself to a Professor-witch, when she didn’t even know the basics? “Muggle researchers have found that people are always very optimistic, compared to reality. Like they say something will take two days and it takes ten days, or they say it’ll take two months and it takes over thirty-five years. For example, in one experiment, they asked students for times by which they were 50% sure, 75% sure, and 99% sure they’d complete their homework, and only 13%, 19%, and 45% of the students finished by those times. And they found that the reason was that when they asked one group for their best-case estimates if everything went as well as possible, and another group for their average-case estimates if everything went as usual, they got back answers that were statistically indistinguishable. See, if you ask someone what they expect in the normal case, they visualise what looks like the line of maximum probability at each step along the way - everything going according to plan, with no surprises. But actually, since more than half the students didn’t finish by the time they were 99% sure they’d be done, reality usually delivers results a little worse than the ‘worst-case scenario’. It’s called the planning fallacy, and the best way to fix it is to ask how long things took the last time you tried them. That’s called using the outside view instead of the inside view. But when you’re doing something new and can’t do that, you just have to be really, really, really pessimistic. Like, so pessimistic that reality actually comes out better than you expected around as often and as much as it comes out worse. It’s actually really hard to be so pessimistic that you stand a decent chance of undershooting real life. Like I make this big effort to be gloomy and I imagine one of my classmates getting bitten, but what actually happens is that the surviving Death Eaters attack the whole school to get at me.”
Harry Potter and the Methods of Rationality Chapter 6

I’d like to distinguish between Harry’s ‘worst-case scenario’ and the ‘worst-case scenario’ I started with. If we’re scared of something we imagine a real worst-case, if we’re planning for something we imagine something that doesn’t even qualify as an average case. If my wife and I had followed Harry’s suggestion and assumed that we would take as long to start our journey as we did for the last journey we would have been pretty much exactly right. We didn’t - we imagined that things would go according to plan.

This is where the concept of “Insufficient Pessimism” comes into play (I got the term from Methods of Rationality in a later chapter where Harry complains about not being “sufficiently pessimistic”). Pessimism is generally a derogatory term in our society - we’re constantly told to be ‘more optimistic’. This is a good thing in terms of worst-case scenario thinking - get on the boat, it’s fun and you won’t be eaten. In a planning sense, however, it’s better to follow Harry’s advice and assume it’ll take as long as last time and if there isn’t a last time we should be really pessimistic and then multiply the already pessimistic estimate by pi.

I’m a software developer and I’ve seen the planning fallacy hit hard in every company I’ve worked for. For every task in a given project there seems to be a certain amount of error and in a large project those error values just keep adding up. One possible solution is to try and do things in steps that are as small as possible - that keeps the error margin small.

In life to accomplish anything worthwhile takes a lot of effort and some luck as well. If we expect things to go smoothly we might get discouraged. My eldest daughter constantly imagines things going according to the best-case scenario and is constantly disappointed. However, if we accept that good things take time then we’ll be more likely to continue on in the face of perfectly normal bumps along the way. Viewed from that perspective having Sufficient Pessimism is a good way to keep grounded and avoid disappointment.

Sometimes we’re in a bad situation and think that things will get better. Maybe there’s a situation you’re in and it’s consistently been bad. You still think - it’ll get better, the best-case scenario will happen. Chances are it won’t. If you’re Insufficiently Pessimistic you can get trapped in a loop and never get out. Get out.

Will Season 4 of MLP be as good as Season 1? No, no it won’t.

This is the first post of 2013. I didn’t intend it to be. That’s the planning fallacy for you right there…



Comments