Good Enough Bets
I took a class in uni where we had problem sets due every week. I typed out the proofs for the first three problem sets in LaTeX. (For those of you do don’t know what LaTeX is, it’s essentially a markup language for writing fancy mathematical text.) It took me around eight hours to finish each problem set. On the day the third assignment was handed back to us, I saw the kid next to me get a single piece of notebook paper with a 93 on the top. For reference, I probably turned in some 8–10 page LaTeX document and got something like a 98. I asked to see his piece of notebook paper and upon further inspection discovered that he had only bothered to scribble two or three sentences for each proof. From then on out I only wrote my proofs on notebook paper and got a lot more hand-wavy about the details. I ended up getting lower grades on my homework assignments on the whole but I didn’t care because I was saving so much time and energy.
Scribbling on notebook paper instead of typing out a solution is LaTeX is what I call a “Good Enough Bet”. I understand that many people already do this all the time without being able to describe exactly what they’re doing or why it works. I’m going to try and break down Good Enough Bets as a way of helping myself, and hopefully you, use these sorts of bets consciously and effectively.
The first interesting thing about Good Enough Bets is that they involve withholding instead of staking. Bets in the colloquial sense require the bettor to put forward something — time, money, cigarettes, IOU — in order to have the opportunity to win more. Good Enough Bets require the bettor to withhold something that they usually spend in order to have the opportunity to get away with the withholding.
Another way to think about a Good Enough Bet is as an active expression of the question, “Can I get away with doing less?” In my case, it was, “Can I get away with scribbling several sentences for each proof?”. I would’ve never taken the bet had I not seen that kid get away with it. I would’ve been too afraid of getting a terrible grade and consequences of that down the line.
Good Enough Bets are scary because you can lose more than what you bet (in this case, what you withhold). If you place a Good Enough Bet that you can get away with doing twenty hours of work a week instead of forty, you could lose your job, which is probably worth a lot more to you than twenty hours of your time. But sometimes you have to make a Good Enough Bet because you can’t really afford to not withhold your time or energy. This brings us to the reasons why people make Good Enough Bets.
The first reason why someone would make a Good Enough Bet is simply that they don’t know they are. If they came into the job thinking that twenty hours a week was the standard, they are inadvertently making a Good Enough Bet. The good thing about making accidental Good Enough Bets is that you don’t get any of the nerves while waiting for the results to come back.
The second reason for making a Good Enough Bet is that you’re in a situation where you have to. For this kid who scribbled two-sentence proofs, I’m relatively sure he always did his problem sets the night before it was due and was playing video games for the rest of the time. Once you have five hours left before the deadline you have to cut corners or just not turn it in. It could be procrastination, addiction, or bad luck that puts you in this position. The only question here is the magnitude of the bet (e.g. do I scribble two sentences for every proof or not turn in the homework at all?).
The third reason for making a Good Enough bet is that you can’t tolerate the status quo anymore. This is the situation where you’d rather cut corners and suffer the possible consequences than continue doing what you’ve been doing. Instead of time or energy constraints forcing you to make the bet, it’s your tolerance (or lack thereof) that’s forcing you to make the bet. If you can’t stand working eighty hour weeks, you might start only working forty hour weeks as a way of keeping yourself sane, and if you get fired, so be it. The alternative (burning out) was worse anyway.
None of these three reasons for making Good Enough Bets is proactive. The first reason is passive because you don’t even know you’re doing it; the second and third are reactive because the circumstances give you no other choice but to make the bet. If you can proactively make good Good Enough Bets, you’re in a position to save a lot of time and energy over the long run. So how do you identify possible Good Enough Bets and decide which ones are worth taking?
I’ll start with identification. Whenever you’re routinely spending resources on something, there’s an opportunity for a Good Enough Bet. Do you get $60 haircuts each month? Try going to the $20 place and see if it’s good enough. Do you spend an hour every night cleaning the kitchen? Try only cleaning for twenty minutes a night and see if that’s good enough for you. Do you obsess over every detail of trip planning? Try being more loosey-goosey with your next trip and see if that’s good enough.
Now you have to vet which Good Enough Bets are worth taking. The best place to start here is with past data points — in other words, what you’ve gotten away with / not gotten away with, and what other people have gotten away with / not gotten away with. When I saw that kid’s sheet of notebook in class, I was looking at what he had gotten away with. The more data points that show that you can get away with something, the more relevant these data points are to your situation, and the more consistent the data points are, the better. In this case, I knew that this kid had gotten away with two-sentence proofs for three different problem sets, was doing the exact same problem sets as I was that were being graded by the same TAs, and had gotten away with it every week. That’s about as good as it gets. If you’re trying to proactively make a Good Enough Bet you have to proactively seek out these sorts of data points. In my case I stumbled upon them, but if I had been more proactive, I would’ve gone around and talked to people about the problem sets when they were handed back and see what sort of effort got what sort of score.
If you can’t find any past data points, you’ll have to start looking for adjacent data points. For example, if people say that your professor is an easy grader, you might be willing to try two sentence proofs. That sort of stuff — basically, extrapolating from data points that are further away or less concrete.
If you can’t find any past or adjacent data points, the Good Enough Bet becomes a real shot in the dark. Not being able to find any relevant data points is pretty rare, but if you are in this situation, it comes down to whether you’re able or willing to lose and your risk tolerance. It’s just like any other bet. I can’t really recommend taking Good Enough Bets with no data because you can usually find some data if you spend a little bit of time and energy searching.