Hacker News new | past | comments | ask | show | jobs | submit login
Ask HN: Tools for rational decision making?
87 points by miguelrochefort 36 days ago | hide | past | web | favorite | 27 comments
Someone stole my phone 2 weeks ago.

I could have walked across the street and purchased any of the 50+ models the store had on display, in under an hour. But I had specific requirements (USB-C, NFC) and didn't want to spend over $200.

I spent over 10 hours searching for a replacement online. I found a good candidate, and a store in my city appeared to have it in stock. So I spent 15 minutes trying to memorize how to get there, walked for over an hour to get there (didn't want to pay $2 for a ride), waited 30 minutes until an employee became available, asked if they had the phone in stock, and learned they they didn't. I then walked back home, got lost, and eventually arrived 2 hours later. I then started my search over, slightly tweaking my criteria.

After 2 weeks without a phone, over 20 hours of research, and visiting 5 different stores, I finally bought a replacement. It's second hand, not exactly state of the art, and I just just found out the vibration motor is broken. But it's too late to return it, as I already left the country. I probably would still be searching for a phone if I didn't have a flight yesterday as a hard deadline.

I'm now looking to buy a case for it. I must have spent 4 hours on this already. Today, I'll visit a few different stores to see if they have it. I found a few online, but shipping makes it expensive and inconvenient.

I do this all the time. Every decision, big or small, is treated as if it's a matter of life or death. I can't seem to make a decision without a non-arbitrary external deadline. I'm doing 99% exploration and 1% exploitation. I value my time at 0.

How do you make rational decisions? What if they're complex? When do you stop? Do you use any tools?




I'd recommend to read up on "maximizing" vs "satisficing":

https://www.lesswrong.com/posts/5MW3DWbvcrztvKnMS/in-praise-...

https://www.psychologytoday.com/us/blog/science-choice/20150...

Personally, in your example with the phone, I think that maximizing is a good strategy. My phone is a critical part of my work envronment, so I'm prepared to invest time and money when chosing and buying one -- that is, I'm a maximizer when it comes to buying phones.

On the other hand, when the cost of error is small and the timespan of possible negative consequences is short (e.g. mintes or hours), I'd satisfice. Examples: buying a bottle of water or a quick snack.

As for deciding when to stop, there is the Secretary Problem (https://en.wikipedia.org/wiki/Secretary_problem), but using this strategy can be painful for maximizers.


I've been to lesswrong a few times and I just never really got it.

But that article was amazing, thanks for the link.


This sounds very familiar.

For example, the simple task of buying a computer from parts could easily take me anywhere between one to six weeks. And this is after the months it took me to move from the desire to upgrade to a commitment. Similarly, for upgrading my laptop, I have on more than one occasion waited for newer models to be announced or become available, only to finally pick a different one.

There are some things which helped me deal with this tendency to over-analyze such decisions:

1. Accept that this is who you are. Even though it feel as time lost, or spent unproductively, you are still benefiting from it. For me, though it still annoys me, doing the research is also a hobby.

2. You are requesting a tool for rational decision making in the title. But I doubt that is what you need or want, or even if such a tool would actually be helpful. For example, what you could do is write down all requirements, give them a weight and scoring and fill in some type of a matrix. If you are similar to me, you'd likely spend even more time analyzing, tweaking your weights, assigning your scores, etc.

3. Realize that the perfect decision does not exist. And no amount of analyzing will prevent that. Often though, making a 'bad' decision will either have a minimal impact on your life, or, can be easily corrected (return policy / low price compared to disposable income). Learn the difference between those types of decisions.

4. Delegate some decision making. For example, when I am looking for a new car, I'll call one of my friends whose opinion I trust implicitly.

I hope this helps


One common advice to help in decision making is the 10-10-10 rule. How much will this decision matter in 10 minutes, 10 months and 10 years.

When I have the right perspective, many decisions that I used to agonize over become easy.


Consider getting screened for ADHD by a professional.

While it's impossible to build a case from one anecdote, some aspects of your story are all warning bells – the hyperfocus on "unproductive" parts of the decisionmaking process, getting lost, decision fatigue, analysis paralysis, and lack of action unless there's external accountability.

Understanding these underlying problems are very helpful to be able to stick with the other tools provided in this thread.


Decision matrices can be very useful to take rational decisions:

https://en.wikipedia.org/wiki/Decision_matrix

However indecisiveness is usually due to anxiety and other emotions. If it becomes burdensome to the point of impacting one's life it needs professional help from psychology.



I think you're approaching this from the wrong angle.

The problem isn't that you've got some sort of decision-making paralysis; indeed, I'd argue that based on the criteria you're determining, you're being very rational.

Instead, the problem lies in the detailed --and maybe often unrealistic-- criteria that you're assigning to each and every life decision, small or large.

Take the phone example you gave. You wanted USB-C, NFC, and under $200. A quick Google confirms that achieving 2/3 would have been easy, but placing those constraints on your decision, and then having the sheer bloody-mindedness to stick to those constraints, even after the first disappointment... that's at the root of your issue.

Take the second example - the case. You have found cases which matched your requirements already, but then you placed the further constraint of not buying online because "shipping makes it expensive and inconvenient". (Passing over whether this is really true) again, it's your insistence on unrealistic criteria which complicates things.

(It reminds me an old adage from my car-modifying days: 'cheap, fast, and reliable; you can have any two'. In your two examples, you're demanding cheap, fast, and reliable.)

There is no easy solution to this. Something has to give. So instead, just be honest with yourself. If you want to continue to make life choices that are based on very stringent criteria, and also want to get those perfect items at a low price and even locally... well, you've got to accept that in most cases, the trade-off is that it's going to be difficult and time-consuming. That's where you are now.

Alternatively... try to see the your decision constructs for what they are, and in cases where a quicker resolution is ideal (I'd suggest that, for your own sake, this is actually most of the time, unless you're really gratified by the research process) try to flex somewhere along the line - either flex your stringent criteria, or accept that you're going to have to pay more, or use mail-order, or whatever. In particular, you appear to be very cost conscious - phone under $200, not wanting to pay for shipping, or a $2 ride. Is this really vital for your survival, or is this just learned frugality? If it's truly vital, then flex your other criteria. Otherwise, you'll find that spending a little extra money in the right places usually makes things go a lot lot smoother :).

(I've personally never found that the mental model of assigning a value to my time and using this as a lever to alter the way I spend my time works --as to me, my free time is, well, free-- but YMMV.)


How much money would you want to be paid to work for one hour? Doing something like phone research? For six hours? Surely more than 0. Work out your hourly rate. Then factor it in to your decision. Sounds like you spent hundreds trying to save $100 or something, a lot more than you could possibly have saved.

Treat yourself like you would treat someone you really love.[0] Don't do stuff you wouldn't ask someone else to, or pay them to do. Imagine if your spouse made you/asked you to do all that work over finding a phone for them. That would seem crazy. Like they had no respect for you whatsoever.

Or maybe you have nothing better to do with your time. I guess not. In which case, carry on.

[0] from Louise Hay's How to Love Yourself. I had to learn to do that. (I used to say the most horrible, abusive, self-defeating things to myself all the time. Now, never.) We can treat everyone in our lives nicely and considerately, while being so mean to ourselves, never having been taught that you are someone in your life too, who deserves respect, to be treated well, and loved.


Many times I've thought about writing a specialized tool for that but I think the complexity easily becomes ridiculous.

That said, Excel/LibreOffice works really well for me when lots of numbers are involved. Otherwise project management software, split decision into multiple tasks and do research as needed. (Writing in Wiki)

As others pointed out, don't overthink. On the other hand, who cares, if you think it's worth it.


I always tend to overthink. At the end, I almost always leave it upto math/science. Spreadsheets are a great way to visualize maths associated with things.


What your actions are missing is a closed feedback loop. You have a few reject signals("must have x/y/z feature") but you will benefit from thinking like a game designer and developing more of them to qualify your decisions and generate new feedback[0].

This process means "knowing yourself" more than it does "develop one technique" - you have preferences and biases. You can't know for sure if you got the best phone. But you can know if your process for "buy new tech gadget" is improving by looking at your usage rates, cost/longevity, etc. and these are things that are governed by how you run your life, overall.

You are not rational in will, but in practice; if you design your personal systems well and follow through on what they indicate, you increasingly make good decisions despite your worst inclinations saying otherwise.

[0] http://ludamix.com/dive/performance/


Well, I don't think it has to do with making rational decisions at all.

Imagine, You had $10 Million Cash in Bank, A House with no mortgage and a passive income of $200K per year.

Would you still have a hard time making this decision?

You probably wont. You will call Uber Black, go into Apple Store, pick up iPhone XS with Apple Care. Take Uber Black Home again because you cant be bother to drive your whatever car parked in Garage.

Likely you wanted the best for the $200, and you couldn't get what you want. One of the problem with low budget is their limited options. Which means you spend more time thinking about those trade off.

So has nothing to do with rational decision, you are just too worry to make the wrong decision because you have limited resources. So to solve your problem, is to forget about these research, buy it. And spend time thinking about how to make more income so you don't have to worry about stuff like these ever again.


Well, now I have something to point to if anybody asks if there's a tech bubble, I suppose.


I made this Google Sheet for rational decision making. Hope you like it: https://docs.google.com/spreadsheets/d/1b9Kjw9pW5cDwGJaJl561...


I think you might be describing analysis paralysis (https://en.wikipedia.org/wiki/Analysis_paralysis).

While I can't advise you how to deal with it I find that acknowledging the tendency makes it easier to override, particularly for relatively trivial and inconsequential decisions like your example.

Years ago I used to aim for 100% "information coverage" to base a decision on. These days I'm slowly starting to realize that the diminishing returns past a certain percentage of "information coverage" are rarely worth the time, though this depends on the gravity of the decision. In the case of the make/brand of my phone case I'd visit a site like thewirecutter.com and order their top pick.


I evaluate such decisions using three measures:

1. What is my time worth? Estimate how much time do you want to spent at most to buy a phone.

2. How much enjoyment do I get from this activity? Estimate how long before you get bored with the "chase" of finding phone.

3. Is it a need or want? Need gets handled quickly, want gets deferred as long as possible.

You spent 20 hours on research, assuming your time worth $20/hr, you could have spent $400 more and get a $600 phone right away.

Assuming you will enjoy the "chase" of finding new phone process for a few hours, let's say at most 3 hours, you could buy $540 phone after spending 3 hours.

It is a need and not a want, so I will default to buying as quickly as possible once the enjoyment of "chase" has waned.

In the end, I would have upped my budget to $540 of money and 3 hours of time to find a new phone.


I used to run a list to talk about neurological stuff with folks living with neurological stuff. We would post research articles and talk about how this impacted actual life to live with it.

You sound like you probably have low affect. People who are more emotional can make snap decisions. They can "god with their gut." People with low affect can't do that. They lack the file that says "I have a good (or bad) feeling about this."

It can help to decide to spend X amount of time on it or to set a deadline. "I need a new phone within X period of time. I will get the best phone I can reasonably find by X date." and then make your peace with the fact that most purchases won't be "perfect" decisions.


For starters, your time value cannot be 0. At minimum, your time value is equal to your living expenses.


You might want to investigate the book Superforecasting and the related goodjudgmentproject.com


Use a mental model. 10/10/10 seems a good fit.

Every time that you’re diluting yourself into a decision ask yourself it that thing would matter in 10 minutes 10 hours 10 days 10 months etc.

You can also use this same mental model to evaluate the importance of something in relation to other vectors. Will this impact your tomorrow, next month, next year.

Are your finances going to be impacted by this today, next month, next year.

Also think about how things will compound. Will this decision have a compounding effect or not. Decisions with compounding effects matter, decisions without compounding effects usually not.


One way to avoid agonizing over a phone is to either buy exactly what you previously had, a Google phone or walk into the Apple Store and get an iPhone. The other way to look at it is with a $200 budget get the best iphone you can and then at a later date or when you get insurance money for the theft or something trade it in because it retains it's value.


You don't make rational decisions. You make emotional decisions.

The tool is choosing to not sweat the small stuff sometimes.


Read this might help

Algorithms to Live By Brian Christian


Step 1. Don't over-think - somebody gave me this advice a few months ago in a very offhand and almost sneering way, but I couldn't thank them enough for having "don't overthink" finally seared into my head

(especially since a lot of these phones are returnable, the decision need not be wholly life-course altering!) - you can almost always guarantee that flipping back and forth between opposing decisions is pretty much the worst course of action (when dealing with potential actions that don't consequentially affect your life in the short term - if you are deciding between crossing the street while the bus is coming towards you or waiting, that would be a counterexample to this, for instance)

Step 2. Optimize for minimizing regret. Afterwards, getting stuff done.

Most day to day decisions are inconsequential, and if not, you can improve on your initial choice - doing so eventually you get an accurate picture of the landscape of potential outcomes. Basically, for a lot of things you don't have to consider all the outcomes at the start, you pick up information as you go along. This can considerably reduce one's decision anxiety. In more technical terms, look up ergodicity; check out this guy's twitter: https://twitter.com/ole_b_peters

Step 3. Solve your anxiety.

You seem like an anxious person. I was anxious like this too. It was totally exhausting, just totally exhausting. I've found that spending a little extra money is well worth reducing the headache of unwanted and exhausting contemplation. Your time is worth so much more than zero, and so much more than whatever your wage is.

You know what tends to work well enough, most of the time? Getting the long-term consistently-popular device, the one that works great for both techies and non-techies alike. Instead of looking down upon people who didn't know how to even begin doing their research, I know look up to them for being able to make a good choice with almost no effort.

A good example of this is that I myself also recently had to pick out a phone for myself. I spent a while researching different Android phones that had specs that would be at least as good as the top iPhone. But then the day before I made the decision, after years of being a committed Android guy, I just said fuck it. I'm getting an iPhone. And it vastly exceeded my expectations, and it made me realize my preconceived notions about what I actually want were pretty wrong. For a case, I just typed in the model of the iPhone + ' case' on Amazon, sorted by most popular, and got the Amazon choice one. Whatever, it worked great just fine with almost zero consideration.

Brain time is valuable! We're knowledge workers! Let's make it easy on our heads.


I’m not saying I’m right or better but I feel like I have the opposite problem. I don’t do much research. I do the bare minimum to confirm whatever bias I started with and then just start executing. Usually works out well enough.


I just got a new phone with a new plan. It is a 2 generation upgrade (s6 to a8). It was the default phone. And it's "great"! I have had enough phones to know that this will be "ok" and much better than what I had. Would I treat a car like this? Or a house? No. But a phone? Yes.

I think there's the process of deciding which decisions need more scrutiny is far more important than the specific process you would use to micro manage a phone purchase vs a house purchase, for example.




Applications are open for YC Summer 2019

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact

Search: