Warning: You’re Not As Smart As You Think

Before you read any further, rest assured that this isn’t another post bashing Netflix for their recent price hike. Instead, since they seem to be in the news I thought I’d use a little bit of data from their business to make another point. And what point is that?

You’re not as smart as you think.

Or you might be too smart.

Or maybe it was the first one.

Now, don’t assume that I’m not smart enough to know I shouldn’t insult all my readers for no reason. I have a point and I’m as guilty of what I’m about to explain as each of you. I realized very recently thanks to a great book called “Everything is Obvious* (*once you know the answer),” that so much of what we devise in our minds as fact is quite the opposite. What our commonsense tells us is probably all wrong when it comes to predicting what’s going to happen in the future. Both the book and a number of challenges we help our clients with each day has helped me see this. So, the bad news is that what you’re predicting (or “forecasting” if you prefer) for the future is almost certainly wrong. The good news is that you can be aware of this and stop placing so much confidence in your judgement and prepare yourself with some contingencies when you’re ultimately wrong.

Don’t get down on yourself. You’ve got good company. People make terrible predictions about the future every day based on what appears to be commonsense logic. But commonsense is just that and nothing more. It’s impossible for a human (even with the help of the computer you’re reading this on) to account for every variable that will ultimately affect a given outcome. Consider the case of trying to predict what a single person might do in a controlled environment. For example, let’s just say it’s one person in a room with a table, a chair, and a glass of water on the table. Your job is to figure out if the person will drink the water in the next 10 minutes. Go.

How would you begin to figure that out? There are millions of variables that might affect this. How hot is the room? Has the person drank anything in the past 10 minutes? Does he need to go to the bathroom really badly? Likely, you’d come up with every possible fact you could about the situation and determine the answer. You might say that he will drink it, but you end up being wrong because you forgot to factor in the fact that there’s a sign on the wall that reads: “Don’t drink the water. It’s contaminated.” Oops.

That’s just one person in a very simple situation. Let’s say I made it five people in the room and asked you to figure out who would drink the water first. What if it were ten people? Now think about trying to figure out what people are going to do in a complex system (i.e., the real world) with billions of variables. That’s why you can never be certain about an outcome and why you shouldn’t trust your commonsense and bet the company on it.

Here’s where Netflix comes in. As Netflix attempts to wean us away from mailed DVDs and towards using streaming only, now’s a good time to consider what impact that might have. So, here’s your question: which has a bigger carbon footprint: shipping a single movie on DVD or watching a movie via streaming? Consider that the DVD needs to be stored in a warehouse (that presumably has heat and lights), put on a truck and sent to the post office, put on another post office truck or plane, sorted in another post office warehouse, and then put on another truck to finally be delivered to your house. For the streaming movie, you find it online and press play.

Commonsense tells you that shipping the DVD must have a bigger carbon footprint (i.e., all the steps used to ship it emit more carbon into the atmosphere). This is why commonsense isn’t reliable. It turns out that streaming a movie has a bigger carbon footprint than shipping one (study here). In fact, streaming’s carbon footprint is double what shipping is. But how is that possible with all the gas for the trucks and heat and electricity for the warehouse? Well, all those servers that need to run to stream your movie aren’t the most energy efficient, so they use a ton of electricity. And it’s electricity that isn’t that all that “clean” to produce.

Unpredictability, 1. Commonsense, 0.

One more example just to make the point: laws prohibiting texting while driving. Surely, banning texting and driving will have a positive impact on the number of accidents. That’s what commonsense would tell us anyway. Well, since this is a post about how commonsense isn’t reliable, I think you know where this is going. A recent study from the Highway Loss Data Institute (HDLI), which tracks traffic incidents in US, showed that bans on texting and driving not only didn’t reduce the number of incidents, the number of incidents actually increased. In the chart below, you see collision claims per 100 insured vehicle years, by month before and after texting law for all drivers, compared with Arizona, Nevada, and Oregon. In this case, California enacted a new law banning texting while driving as compared to three nearby states that didn’t implement any laws.

How is this possible? Banning a big distraction for drivers has to make it safer for everyone, which should mean fewer incidents. According the president of the HDLI, “[Lawmakers are] focusing on a single manifestation of distracted driving and banning it. This ignores the endless sources of distraction and relies on banning one source or another to solve the whole problem.”

Oops.

Unpredictability, 2. Commonsense, 0.

If you think about it, creating a policy that only impacts one of the billions of factors that affect whether or not there will be a traffic incident is pretty arrogant. What is says is that we’ve figured out which of those billions of factors is most influential and are eliminating it. Of course, figuring out that one factor with some degree of certainty is probably impossible. It’s impossible because you can’t even figure out what all the factors are much less which are most critical and even less, how you can impact those factors. Focus on the wrong factors and people pay attention only to those, which risks them no longer paying attention to other factors. In this case, it may be that since police officers and drivers are focused on stopping texting while driving, they pay less attention to drunk driving, which leads to a corresponding increase in incidents caused by this factor. You never know until it happens. Hence the title of the book: “Everything is Obvious* (*once you know the answer).”

So, what are you to do? First, don’t arrogantly assume that you know, understand, and can affect every factor that might impact a future situation. Do your best to understand what the factors might be and which are most important. You can make your plans based on this, but you should focus more on developing a process that allows you to quickly adjust your thinking and strategy, as real world data comes in and proves your commonsense to be wrong. If your plan in completely inflexible, you’ll fail. Instead, focus on developing plans and processes that allow you to quickly adjust based on the unpredictable nature of our world. And finally, just admit that your commonsense and “gut” is probably wrong (but also the best you’ve got) and don’t bet everything based on what your logical thinking might always suggest.

Let’s score one for commonsense for once.

Tags: , , , , , , ,

4 Responses to “Warning: You’re Not As Smart As You Think”

  1. Laura Creekmore July 20, 2011 at 1:40 am #

    Great post, great points. As much as we can measure our environment and even our own behavior today, we often screw up because of the things we assume — the things we don’t take into account, the things we believe to be unimportant, etc. etc. 

    It’s hard to get many complex questions down to the bare bones, but so critical if we want to improve our accuracy in predictions.

  2. Guzzo August 12, 2011 at 7:52 pm #

    “What our commonsense tells us is probably all wrong when it comes to predicting what’s going to happen in the future.”

    That’s the bane of every individual investor out there, and why it’s best to appropriately allocate one’s investments, and invest in passive, low-cost, index mutual funds or ETFs for the long term.

  3. Paulgottschalk November 22, 2011 at 1:19 pm #

    Good article but I do not agree with your first example and would like to share my thoughts…
    Argument 1: Two things (only having read the abstract of the link to your source), i) does the comparison include actual production of the DVD itself, ii) “the energy consumption and carbon footprint of streaming can be reduced to approximately 30% and 65% respectively of that of shipping” was written in 2010 and your article was written almost a year later. How about an update from U of Mass.
    Argument 2: Although in agreement I noticed your article about QR codes on billboards (as a suggestion vs. Facebook & Twitter logos; great article by the way) suggests that taking a picture while driving would be dangerous. I’m interested to know if you have come across any data suggesting texting is less dangerous than picture taking?

    • Jonathan Richman November 22, 2011 at 1:39 pm #

      Thanks for the comment. I love the challenging of assumptions.

      Argument 1: the study doesn’t account for the production of the DVD. It only seeks to compare the carbon footprint of receiving an actual DVD in the mail from Nexflix versus streaming the same movie from Netflix. The DVD (or some other storage media) has to be created in either scenario. If it’s a streamed movie, it had to be ripped or encoded and transfered electronically (which we know isn’t all that carbon footprint friendly), but that’s not considered either. That’s not really the point though. The entire point is that the natural assumption that people would make isn’t always correct. I have no idea if there’s an update from UMass (couldn’t find one), but based on what I know personally about the “greening of IT”, I doubt this change has taken place yet. Again, this was just an example to illustrate the point that the answer isn’t always what you think it is.

      Argument 2: I know of no data on the dangers of taking pictures while driving much less this activity compared to texting. I can only assume that people considered this so outrageously and obviously life-threatening to oneself and others that no one even considered studying it. Importantly, I’m not suggesting to text and drive either. I don’t think asking people to immediately do anything in a billboard is really safe whether it be texting or calling. That’s why most billboards are simply reminder ads. As far as the relative safety of texting for information or snapping a picture of a QR code while driving, I’m not sure. Both are unsafe and I’d guess that pictures are unsafest…but that’s a guess. I probably should have simply noted in the post that you shouldn’t risk people’s lives at all by using billboards to ask them to do things while driving.