Known future prices don't update possible past price range & small spike strangeness

[NeatNit]

I believe the code that generates small spike is this: https://gist.github.com/Treeki/85be14d297c80c8b3c0a76375743325b#file-turnipprices-cpp-L312-L344

This week I got this pattern up until yesterday: https://turnipprophet.io?prices=108.66.62.58.53.103.147......&pattern=1

Let's take a look at the 3-day peak. In the prediction, it looks like:

1. 151 to 215
2. 151 to 216
3. 151 to 215

But based on the code:

rate = randfloat(1.4, 2.0);
sellPrices[work++] = intceil(randfloat(1.4, rate) * basePrice) - 1;
sellPrices[work++] = intceil(rate * basePrice);
sellPrices[work++] = intceil(randfloat(1.4, rate) * basePrice) - 1;

First issue: The peak itself can be at minimum intceil(1.4 * 108) = 152 - not 151.

Now, today I didn't check in the morning but I did in the afternoon, and got:
https://turnipprophet.io/?prices=108.66.62.58.53.103.147..153....&pattern=1

This confirms that rate is about 1.416, which means that the range for the day before and day after is only 151-152. This is correctly shown for the day after, but not for the day before. Turnip Prophet still shows a possibility for 200+ on Thursday AM.

Probably the code never goes back to fill in the gaps of missing past information.

[peter50216]

Related past PR: #132.

It's a known limitation that the current code doesn't do backward propagation on possible ranges at all.

My take is that this issue is low priority, since the past values are not useful in any way (time travel resets everything).

[NeatNit]

My take is that this issue is low priority, since the past values are not useful in any way (time travel resets everything).

That's true, but a user who sees this and doesn't know that it's incorrect will be bummed that they missed a peak that never actually occurred.