Is it just me, or does the "75% chance" or "80% chance" not actually mean that or am I just unlucky?
Practically every single time I forged something that wasn't 100% chance, it failed. Am I just unlucky or is this an actual thing?
Forging question
Well darn.
Guess I'll keep grinding to 90 shinings per level per piece of equipment then!
:'(
Unless you go 2x-ing all the time, dont do it.
It only pays out if you consistently do it, no matter what.
If you do it occasionally you will just waste your crystals.
On the long run ( im talking about 50-100 items fully heated ) it will be successful, but you just need to bear with it even if you fail big time from time to time. No forge box chance tho'.
Several of us have collected data on forging rates. Most of the data is for 100% chances, to see how often we get forge bonuses. I have never seen a data set refute the notion that the 70%-80% chances are what they say. From my knowledge of programming, I also have no reason to suspect that they would have programmed it incorrectly.
Assuming that the chances are what they say, and ignoring how they might affect the forge bonuses, your best option is to use these medium (70%-80%) chances. Heating a 5-star item then requires an average of 420 Radiants instead of 453. But you have to be willing to put up with the occasional failure, which is psychologically tough for some players.
The way to put up with the occasional failure is: Remind yourself that you've saved 1/3 of your fire crystals on each other forging. That's not a little. That's a lot.
Just use the 100%. I've had the 2x chance fail around 5-10 times in a row and the set back in terms of resources just isn't worth it. We're talking like 500 radiant fire crystals, which is like a month of grinding for mortals. Even if it pays out consistently better around 2x, steady progress and guaranteed results are the only way to insure you don't just waste time.
I've had the 2x chance fail around 5-10 times in a row...
That will sometimes happen, but only rarely (if the chances are as claimed). The medium chances are good for players who can adopt a long view. If you instead have a short view --- perhaps because you need to heat your Ash Tail Coat quickly, for a coming Shadow Lair trip --- then you might want to minimize your risk and use 100% chances.
Even if it pays out consistently better around 2x, steady progress and guaranteed results are the only way to insure you don't just waste time.
If you use the 100% chances to heat a 5-star item, then you are wasting about (453 - 420) / 453 = 7% of your wealth on average. You are paying this premium to avoid taking on risk.
The same principles operate in the wider financial system. 25-year-olds saving for retirement are told to focus on high-risk, high-reward investments such as stocks, because they have a long time for their luck to average out. 65-year-olds are told to focus on low-risk, low-reward investments such as insured accounts with guaranteed rates, because they need some dependable money in the short term.
Everyone has different tolerances for risk, and that's fine. I just like to make sure everyone is making informed decisions.
Normally I prefer to keep such things to myself in order to frustrate certain individuals since after all, I'm under no promise to appease them in this way but today I'll humor you with a legitimate argument. Wow. A legitimate argument from Fehzor. Oh boy. Here we go.
Lets define a "soul crush" as being 3 or more losses in a row, resulting in a total of 4 90+ payments for the same "70%" level up being made.
Odds of soul crush occurring: .3 ^ 3 = .027 or 2.7%
Odds of 1+ soul crush per item: 7.8%, as there are 3 70% levels to get through.
Average loss from a soul crush of this order, assuming either 3 losses or 4 losses + 1 level up = 275.2
This can be calculated as 64 (average loss) * 4 * .7 + 64 * 5 * .3 = 275.2
Subtract 64 from that loss, since you're paying that last level either way, and you get 211.2 wasted rads from your average soul crushing experience.
Lets say the average gain is what happens when you don't get your soul crushed, subtracted from the 100% odds...
Avg. gain per level = 96 - (.7*64) + (.7*.3*128) + (.7*.3*.3*192) = 96 - 83.776 = 12.224
(There is a small margin of error here, as this doesn't account for the 2.7% that ARE soul crushing)
"Wins" required to pay for a soul crush = 211.2 / 12.224 = 17.27 = about 6 heated items
Over the course of 6 heated items, you would expect to have:
Odds of 0 soul crushes = 61% = you make a good profit off of this
Odds of 1 soul crush = 30.51% = you break even, gaining slightly
Odds of 2+ soul crushes = 8.3% = you break even or fail.
So how many items do you have to craft before you end up getting the 6.2ish% increase from these last 3 levels? At 18 items crafted you're still around a 12% chance of breaking even/losing. Double that and to 36 items and you still have about a 7% chance, but it's at least above a 90% chance of you succeeding. You're unlikely to lose out but it isn't out but progress is still going to be slow and frustrating, and unpredicted setbacks are a nightmare. I'd much rather know how many items I can head, and have a chance of being pleasantly surprised by a forge box than have to go through this gauntlet.
Most players are only heating a few items to begin with. Lets say they heat 12 items total, and take your advice to use the 2x. 25% of the time they're going to have a drastically harder time getting through the game. It's just not worth it, especially for your first few items. 25% is kind of a lot of people.
I haven't checked your calculations. Your motivation is reasonable. The problem boils down to managing expectations.
Naive stock market investors get freaked out when the market declines. They sell off their stocks and move their money into something safer. Then, when the stock market picks up again, they get back into it. The consequence is that they're selling low and buying high, which is the worst investment strategy. So a financial adviser will tell them to expect downturns, not freak out, and take a long view based on the best available statistics.
Similarly, Spiral Knights players should understand basic probability enough to know that "soul crushes" will sometimes happen. They should consider them the price of saving an amazing 33% on each successful forging. Is the price worth it, numerically speaking? On average, yes.
But you are right, that many players might not understand or trust this. Or maybe they have good reasons to avoid risk, as I explained above. So I try not to beat them over the head with it.
I just worry that if we advise players to go for this we're going to end up hurting a good 8-40% of players and in that way the advice may as well be 8-40% wrong. These losses are the kind of thing that makes you want to quit the game.
The gains on the other hand aren't even much of a game changer at 7%.
Just wanna point out an example here.
A friend of mine actually 2x'd every 5 star items he ever had. We are talking about more than 50 items here.
He often ended up with no fails with an item at all.
Of course sometimes awful fails came in for him too.
Overall it was worth it for him if we strictly count only the radiant numbers (forge box possibility = 0.00%).
If i can remember correctly he ended up avering 380/390 radiants per item.
But let me point out for everyone who reads this: it is only working if you are able to handle the psychological breakdowns this method causes. Only worth it in the long run.
Myself? I would advise _everyone_ to not use it. I never used it once. 3x is much better for your soul, not to mention the thrill that maybe you get a forge box :)
Alright, now I'm convinced it's rigged.
So I started a new knight, and I wanted to forge my Zapper. I decided to test out the 90% chance option.
It failed.
T_T
You are not along on this one, just play XCOM for long enough and you will too begin to appreciate the pain of failing a 80% shot on the Muton who then naded your whole squad the next turn. Although in XCOM you can at least recuperate without the unnecessary grinding.