Resource RBY Prediction Guide

[Hipmonlee]

Here is the Prediction chapter of my RBY Guide. This is version 0.2.

As I have mentioned to a few people, the goal was more or less to share everything I know, but it turns out after playing this game for 20 years I know a fuckton, and this whole thing got a little out of hand.

I want to thank ziloXX, Amaranth, Earthworm, MetalGro$$, ErPeris, Lusch, Peasounay and Serpi for giving me permission to use their battles as examples. And Earthworm and vapicuno for giving feedback.

I hope people find this useful. I tried to make sure it would be useful for players who are relatively new to competitive Pokemon as well as the top tier guys. It should also be useful for other Gens, although some of it is a little bit different. In particular, entry hazards, leftovers and u-turn like moves shift the balance of things considerably. But the fundamentals are all there.

One of my goals was for it to be better than Borat's guide, and, in my opinion, I fucking nailed it. I kinda rushed finishing it, because my partner is really close to having our baby, so I am not certain that it is the single greatest video game guide ever written, but I am perfectly happy to assume that it is.

Changelog:
0.2
- Minor grammar fixes
- Corrected values in equations for 2x2 game (thanks to Melanie for pointing that out).

 

[Amaranth]

"Your opponent’s Tauros will have the moves Body Slam, Earthquake, Blizzard and Hyper Beam or it is a bad Tauros."

This is heresy.

The rest of the guide is fascinating, because it's /so/ far from how I think about the game. I appreciate the different perspective.

 

[Hipmonlee]

tbh I really had fireblast in mind when I wrote that rather than thunderbolt, that's a move I can forgive. But, you know... I am not gonna walk away from an opportunity to stoke controversy.

Rest assured, if I write the full guide, I will cover in excruciate detail the relative merits of Thunderbolt, Stomp, Rest, Fire Blast, Tail Whip and possibly even Horn Drill, Fissure and Double Team.

 

[Ortheore]

Hmm I found it to be a really dense read- it's the kind of thing that I think I could only go over by skimming through it several times. Still I think it offers quality insight into the subject matter and should definitely be helpful to almost everyone, though in varying capacities for each player

Still, some very interesting stuff. I broadly think of "optimal" play as being based around evaluations of risk and reward, which I think is a much more simplified version of what you're getting at in the mathematically oriented section, which is pretty simply an attempt to solve the game as much as possible. It's when I use the kind of rationale in the wishy-washy section that I consider myself to be making a "predicting", since that kind of reasoning, though it can theoretically be incorporated back into the risk-reward calculations in the first half of your chapter, still can often leave you still feeling like you're making a sub-optimal play. But I looooove it when such plays work out, because then you're beating your opponent rather than optimising a game.

I also noticed that uncertainty of outcome (i.e. hax) wasn't mentioned much, though I suspect I overlooked it. Although it can be assumed to be incorporated into an analysis of payoffs for a given choice, it still feels weird to discuss solving the game without mentioning it since it's one of the main reasons the game is unsolvable (alongside type matchups and hidden information in terms of both team composition and double blind choice of actions)

 

[Hipmonlee]

Oh yeah, I had intended to dedicate a section to rng. I guess that got forgotten somewhere along the way. Basically to say that hax is just included into your payoff calculations, so it needs to be accounted for before you start thinking about prediction. And when I talk about risk in the guide, I am talking explicitly about payoff values and not about hax.