Virtue points

One great way to hold yourself accountable to your goals and resolutions is Beeminder. Beeminder is an app that works as such:

• You give them your credit card information.
• You set goals (exercise for 15 minutes today; finish the report by Monday; read Wuthering Heights by February) along with wagers ($5; $20; $50).
• When a deadline passes, the app asks you if you’ve succeeded at your goal. If you say you didn’t, they charge your credit card the amount you wagered.

This is a really neat idea, self-described as “reminders with a sting”. You set goals, and then you’re actually incentivized to follow through, because if you don’t, you’ll lose money (the sting). I know some people who got dramatically better at accomplishing their goals after setting up Beeminder or something similar.

I’ve never tried Beeminder — I’ve always felt a little averse to it. Upon introspecting, I think the main reason for this is the negativity: my brain really doesn’t like the notion of setting up a self-punishment. Beeminder essentially lets you make a threat: “I need to do this, or I’ll face consequences”. I don’t like threats, and I think I’d end up feeling bitter about having to give Beeminder my money.

At the same time, I want some sort of accountability system. There are various things that I would like to do, but don’t do because I lack the motivation. Earlier today I was thinking about all this and came up with a system that I think would work better for me. It’s called virtue points.

Similarly to Beeminder, the virtue points system works by setting goals. But instead of wagering money, you set a number of virtue points (VPs) that you will earn by reaching your goals. Here are some hypothetical numbers involving actual goals I might set:

• Exercise. Worth 1 VP for every five minutes of exercise.
• Read [paper] by end of week. Worth 15 VPs.
• Finish [blog post] by end of week. Worth 10 VPs.
• Go to [social event]. Worth 5 VPs.
• (This is really important for the system to work, and will always stay the same) Donate to charity. Worth 1 VP per dollar donated.

Additionally, every week, you set a VP goal. This goal is required! If you set a goal of 30 VPs in the coming week, you must satisfy that requirement. And because of the “donate to charity” option, there will always be a way for you to do so.¹

The thing I like most about this system is that it’s framed in a positive way. Instead of being “reminders with a string”, the framing is “here are the various ways that you can earn your virtue points this week, so that you can be the person you want to be!” And whereas in Beeminder the recourse is losing money, here it’s donating to charity — something I won’t end up feeling bitter about!

I think the positivity aspect is really important for me, and that means I should be somewhat careful when setting goals. For one thing, at least in the beginning, I’ll be setting positive goals (“accomplish x”) rather then negative goals (“don’t do y”). Then, if the system seems to be working, I’ll add in negative goals (such as “don’t spend more than [time] on my phone”). Also, I’ll probably set my weekly VP goal low at the beginning, so I don’t end up getting discouraged.

The other thing I like about this is that virtue points make goals commensurable. One could imagine a system that’s in between this and Beeminder, where for every goal I either accomplish it or donate to charity. But the VP system gives me more flexibility. I don’t feel like reading the paper this week? That’s okay — I can donate more to charity, or exercise more, or make progress on my blog post! Put another way, the system enforces a certain global virtue goal, but I have total flexibility in terms of how to achieve that goal. And if I find myself consistently earning my VPs through a particular goal, that means one of two things:

• I can make that goal worth fewer VPs (so I will need an alternative route to my weekly goal); or
• Mission accomplished! I now consistently accomplish one of my goals, to the point where it has become a habit.

There’s one last component, which is that the amount you donate to charity outside the VP system needs to be independent of what happens inside the system. If you end up donating more outside the system because you feel guilty about how little you donated inside the system, your incentives are messed up: now, when you earn a VP by donating a dollar to charity, the increase in how much you will have donated over the counterfactual is less than one dollar. (This is the same reason that donation matching isn’t as good as one would naïvely expect: often, the more the amount donated through the match, the less the matcher ends up donating on their own.)

For this reason, I’ll make my 2021 charitable contributions at the start of 2021 rather than at the end, and I’ll commit to not donating anymore in 2021 except through the VP system (and miscellaneous things like awards for solving puzzles). I’m excited to try this system and see how it works out!

(By the way, the concrete goal that inspired this system is reading one academic paper a week next year, as compared with my current rate of one every couple months. I’ll aim to hold myself accountable to an acceptably high standard of comprehension by writing a monthly blog post summarizing the papers I read. So if you’re interested in theoretical computer science, you have that to look forward to!)

^{1. If you want to use this system, note that you can adjust how many VPs your goals are worth so that your weekly goal won’t be too much of a financial burden even if you go with the charity option.↩}

Advertisement