Ratio Goal Type?

What would be the right goal type to Beemind a ratio or average?

Let’s say I make spontaneous commitments to myself throughout the day that I usually fulfill in the next two minutes to a couple of hours. Whenever I make such a commitment, I create a data point with a value of zero. If I fulfill the commitment, I change the data point to one.

After a couple of days, I would have many data points with values zero and one, respectively, from which I can calculate the ratio of met commitments by dividing the sum of ones by the total number of entries. This will result in a value between 0 (I failed every single mini-commitment) and 1 (I succeeded for every single mini-commitment).

Now, what I would like to do is Beemind that this ratio stays above a certain ratio (0.5 let’s say) meaning I fulfill at least half of my mini-commitments. Is there a way to accomplish this (without external tools or scripting)?

3 Likes

Here’s the hypernerd approach:

For the case of keeping a success/fail ratio above 50% you could have a do-more goal with a flat red line at zero and then enter a -1 for every failure and a +1 for every success.

If you wanted a 2/3 success ratio, you could enter a -2 for failures and +1 for successes. You need 2 successes for every failure to stay above zero.

In general, for a success ratio r, you’d enter +1 for success and -\frac{r}{1-r} for failure.


Keywords: ratiominding, beeminding a ratio

5 Likes

Or you could try:

  • enter each new commitment with a data value of 0
  • update it to +1 if you do it, -1 if you don’t
  • set a do-more goal with a flat line set by your target ratio, so for 1:1 success:failure it would be 0 (because +1-1=0), for 2:1 it would be a flat line at +1 (because +1+1-1=0), and so forth
3 Likes

You should even be able to preload some safety buffer with an initial value.

1 Like

Thank you! Both suggestions make perfect thanks, and I should have come up with them myself.

I might report back here about how it goes with my “mini-commitments”.

2 Likes

I have several ratio goals, thanks to the truemean aggregation method. It’s a custom goal setting that averages the values of that day’s datapoints instead of adding them together. If you enter only zeroes and ones, that gives you a ratio.

n.b. you don’t want the badly-named ‘mean’ method, because that deduplicates the datapoints before averaging them. I even have trouble imagining a use case for that approach, but there must have been one for the first user who wanted to aggregate things that way.

2 Likes