> Introduction

> There is a lot of praise for weekly reviews. [...]

What are you talking about? Reviews of what? Predictions for what?

I finished reading the "Introduction" completely clueless what this article is about. I continued "Benefits", still not having a clue what the article is about. After another paragraph I closed the tab.

Sorry, this article seems to be targeted at some niche audience, and I'm not in it. It would be great to have some actual introduction to indicate what you are talking about.

I didn't close the tab outright, but I agree that the author should be more forthcoming about what exactly they are talking about.

The whole "weekly review" thing just absolutely lacks context. Like, are they talking about the "weekly review" in the GTD methodology? Or in something like Scrum? Or something else entirely?

So I left a comment on that blog and the author replied that the "weekly review" is the one in GTD.

you may then also like “Get Rid of the Performance Review!” by samuel culbert[0] which advocates performance previews (predictions) rather than backward-looking reviews. [0] https://www.goodreads.com/book/show/7348406-get-rid-of-the-p...

(edited author name)

Correction: Samuel Culbert, not Stephen Colbert.

ha, i think my subconscious was saying we need more colbert.

Yeah. My first thought was: yearly employee reviews... nice thought though!

This feels like the sort of thing economists would latch on to: when in doubt, either create an auction to find out more about preferences, or make it a betting market to get a decent estimation.

It's also something I've read about recently, that we frequently make decisions implicitly based on probabilities and expected values, only we very rarely make these numbers explicit. If we did, we might act more rationally.

I like the idea of this a lot. I have trouble seeing how I would "get better" at predicting stuff solely through this, but perhaps forcing myself to think it through would be sufficient.

> I have trouble seeing how I would "get better" at predicting stuff solely through this, but perhaps forcing myself to think it through would be sufficient.

I think the approach would work well in the "false positive" case, i.e. that you are always fairly certain some kind of activity is worth your time, but it ends up disappointing you every time.

On the flipside, if thinking through something makes you come up with a low expected value, you probably won't do it and won't know. But then again this can also be a pointer that you should maybe "dip your toes in" first and not commit too much time to something you don't see as promising.

> “we frequently make decisions implicitly based on probabilities and expected values”

that’s all intelligence is, the ability to make predictions about the future. our subconscious does most of the work, which is why it’s not “explicit”. feelings are the bubbled up (non-numeric and fuzzy) results you’re looking for.

it’s not mathematical in situ. math is something we overlay on it to bring it to a conscious, external realm of understanding.

> It's also something I've read about recently, that we frequently make decisions implicitly based on probabilities and expected values, only we very rarely make these numbers explicit. If we did, we might act more rationally.

Do you recall where? I'd be interested in reading it

Most recently in The Undercover Economist by Tim Harford.

I can't give a full recommendation because it's recent enough that I'm still only about halfway through it, but overall it has been worth a read so far. (I found some chapters dull -- in particular the bits about the 2008 financial crisis have just not interested me -- but I have high hopes it will regain my interest again in the rest of the chapters.)

If you decide against action because of your prediction, how can you evaluate your decision/prediction?

You can't.

I use something similar to fix( work in progress ) my problem of having hard time saying no to people and things.

So by having predictions you can cut down on your choices. Just seeing the list with chances you have attributed to them it's sometimes enough to clarify what will be best value for your time.

Great insight, this is a general issue with making decisions based on predictions. It comes up in Hanson's "Futarchy" / decision markets too.

For example, if you have a huge bias against recruiters and never take their calls because your predictions are < 3%, but the true probability is 50%, you'll never find out those predictions are wrong.

One way to mitigate this is to add random exploration. Whenever you decide not to do something, with probability say 1/20, just do it anyway.

Skip to the example section then read back up top. I couldn’t get anything till I skipped to it

On seeing the title my mind immediately jumped to prediction markets as a tool that should replace scientific papers' reviews. Which is a bad idea because of tiny amount of specialists in any scientific field and resulting minuscule size of potential prediction market.

This is an advertisement fluff piece that links to some "prediction book" URL.

The prediction book site mentioned in the article is free to sign up for (with no unusual info requested), and its code is open source (published on GitHub). The article explains how the author uses said website within his productivity-/time management habits.

What about it made you classify it as advertisement fluff?

You can just as well do it on paper or virtually anywhere. No point using such a service.

The value proposition of isn’t easily found or understood.