You're looking at Less Wrong's discussion board. This includes all posts, including those that haven't been promoted to the front page yet. For more information, see About Less Wrong.

Alejandro1 comments on Harry Potter and the Methods of Rationality discussion thread, part 16, chapter 85 - Less Wrong Discussion

9 Post author: FAWS 18 April 2012 02:30AM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (1106)

You are viewing a single comment's thread. Show more comments above.

Comment author: 75th 09 May 2012 09:34:32PM *  3 points [-]

Some predictions for the next arc and beyond:

The climax where Quirrell's identity and/or motives are revealed will be in the next two arcs (p ≈ 0.8), and possibly in the next arc (p ≈ 0.3).

This last arc ended ominously; I think we're perilously close to seeing some serious shit. I assigned low probability to this happening in the next arc because Eliezer said the next arc picks up immediately after this one. We're still in April, and I have this hunch that maybe Harry's "What do I get if I can make it happen on the last day of school?" line to McGonagall was foreshadowing. I also think we'll get to see Quirrell execute his Christmas Wish plot. And speaking of Quirrell plots,

Quirrell will execute his Final Solution to the Granger Problem in the next arc (p ≈ 0.75) and very possibly succeed (p ≈ 0.5)

The next arc is going to pick up immediately following the last one. I assume that "immediately following" means "the day after". Quirrell might have given her more than one night to consider if he weren't planning on getting rid of her immediately. If seers all over the world have nightmares one night, then presumably something bad is going to happen very soon thereafter. Hermione meeting (or maybe even almost meeting) a horrible demise would send Harry straight to his Dark Side, and who knows what he'd do then.

The worldwide seer activity is due to Harry being about to kill a lot of people. (p ≈ 0.35)

Harry Potter's Dark Side just figured out a reliable way to kill large numbers of people in a small amount of time. Maybe he's about to use it.

The Trigger Warnings are next going to be updated for a chapter called Gur Olfgnaqre Rssrpg. Vs n ohapu bs crbcyr jngpu Urezvbar trg uheg naq yrg vg unccra, Uneel zvtug qrpvqr gurl nyy arrq gb qvr.

Or, if you connect Harry's two unwittingly ominous resolutions that directly preceded the two Trelawney nightmares, he might decide that everyone in Britain who supports Azkaban's use of Dementors needs to die. All he needs is a broadcast medium to disable the country's Patronuses, and he can send a Dementor to create a wizarding holocaust.

I assign low probability to this not because I think the clues don't point reasonably strongly in this direction, but because I fail to see how Harry and the story could recover from it. Perhaps Dumbledore would subdue Harry and take him into hiding? But then we wouldn't see the end-of-year stuff at Hogwarts, unless it happens two or three arcs from now.

At some point, Harry will break his Time-Turner to get out of a sticky situation. (p ≈ 0.85)

Twice in early chapters we were told that strange things happen when Time-Turners are broken. Once would be an offhand reference, but twice indicates to me that Eliezer has something in mind. If Harry successfully rescues Hermione from whatever Quirrell tries to do to her next, I think this might be how.


Those probabilities are mostly pulled out of thin air, but I've seen other people use them, so apparently it's expected. Is there some systematic method people use to arrive at them, or do you just sort of look out a window and see what number feels right?

Comment author: Alejandro1 14 May 2012 11:50:41PM 5 points [-]

A common method to get an idea what is the "subjectively correct" number to use as your probability is to imagine yourself betting (a moderate amount of money you would be willing to risk) on the claim, and deciding which odds would you accept. For example, if you would accept betting up to $40 against $10 on your claim, but not more, then the probability you assign to it is 0.8. If you would be willing to bet only up to $10 on a chance of winning $90, then your probability is 0.1.

Comment author: 75th 15 May 2012 12:08:03AM *  3 points [-]

I actually considered revising all my estimates using the rubric "What would I pay for ten shares of this prediction on Intrade?" But I decided that that method would likely introduce a strong bias based on my financial situation, even if I tried to imagine myself to be in a financial situation closer to the median.