Having recently heard grumblings of those dissatisfied with the progress they've made while employing the new Destiny Cipher spell pattern, and having also personally been instructed to use this spell to bond with a tool I was given, I took it upon myself to design and perform a controlled experiment to test its efficacy.

I began by purchasing from the local despondent Elothean in the Middens an ornate mirror. I performed 100 predictions of the future, using a complete understanding of the celestial influence for each, without employing the Destiny Cipher spell. I recorded incidents of both the damaging blasts of heat, and the oozing feeding sensation of bonding. I will refer to this as sample one, or the control population. I recorded the final state of this mirror and my bond with it, before discarding it.

Next, I purchased an identical ornate mirror from the same Elothean. With this mirror I performed 100 predictions in the same way, except that I did employ the Destiny Cipher. I note for the sake of full disclosure that while taking this sample, I also performed 4 full pool predictions that did not employ the Destiny Cipher, by accident. Two of those four resulted in bonding, but more importantly none of them resulted in damage. These 4 trials were not included in analysis for either population. They were ignored. I will refer to the 100 predictions with Destiny Cipher as sample two, or the test population.

As further control, my charisma score was constant and not trained or buffed during these tests. I model the change in bonding probability due to damage as "very slow," and thus treat the bonding probability as constant for each sample. This implies that the random number of successful bonds in each sample will follow a binomial distribution.

The null hypothesis for my analysis is that the probability of bonding with the Destiny Cipher spell is double the probability of bonding without the Destiny Cipher, because that is what we were told when the spell left preview.

I evaluated the cumulative distribution functions to determine the probability of a result equally extreme or more extreme than those I observed as a function of the success probabilities. I used a golden section search over the success probabilities to find the value most likely to fulfill the null hypothesis given my data.

In sample one, I observed bonding 9 times out of 100. In sample two I observed bonding 5 times out of 100. This results in a p-value of ~.0038. Therefore, I reject the null hypothesis, and claim that Destiny Cipher is not working as intended.

---

TLDR: I tested Destiny Cipher, and I'm 99.62% sure it isn't working.

I'll admit I don't understand your stats, but the way I read this is you have n=1 in each group. You also threw out 2 bonding events which would have made it 9/100 versus 7/100. That said, something with an 8-10% effect size is going to require in the ballpark of thousands of mirrors to test appropriately. Maybe less, I just kind of estimated a standard deviation.

I am not great at math and really nobody has to be to figure out that bonding tools is very hard. IN fact I recall reading a post recently about two moon mages who were circle 200 or near who failed at capping tools recently doing every possible step to ensure the highest chance of bonding.

To me that spells out something is off in general with bonding. I think that because of the nature of the beast that is bonding, the results could swing really far either way in any sample, even larger ones. Perhaps the issue is at the core of the system and not with DC.

I feel the system ought to be made so there is a much higher base chance of bonding and to supplement that there should be higher quality tools available year round. Perhaps with those changes, DC as it stands will be great.

Also I would note that DC works good to increase the accuracy of my predictions when I don't use a tool. And with how good it does this for me, I don't even need a capped tool with all things considered. But its a kinda depressing aspect of Astrology that could be enjoyable or even fun.

"An' I'm learnin' 'ere in London what the ten-year soldier tells; If you've 'eard the East a-callin', you won't never 'eed naught else. No! you won't 'eed nothin' else"

THANK YOU for looking into this and fixing it so quickly, Javac! This is really good news!


- Navesi

Sweet, thank you.

I, um, don't suppose I could get the rare mirror that I wasted from this restored?