Congratulations on your purchase of Cubiprocle for Android by Aniviza! With Cubiprocle you can evaluate the outcome of an experiment, game, or just about any situation where you know the probability of an event, or think you know it, and have repeated observations of the behavior under multiple tries of the experiment.

For example, say you want to know if a coin toss is from a fairly weighted coin. So your null hypothesis is that the coin is fair and has a 50/50 chance of being heads. One way of writing that is ½ the time, it will come up heads, or 0.5 probability. Let's say you flip the coin 30 times, you'd expect by common sense anything around the mean value of 15 would probably be fair. But what about 12 heads? What if you saw 12, is that within fair, or is this coin bogus? Cubiprocle it! Enter .5 in the top field, 30 in the next, 12 next, and click Calculate, it's that easy! Cubprocle will sum up all the probabilities for less than 12 observations, greater than 12 and 12 itself, and lets you know which side of 12 is more or less likely and by how much. In this case, Cubiprocle says you could expect to see 12 heads about 8% of the time, and more than 12 about 82% of the time, not the most usual outcome, but is still within fair 95% confidence.

How about dice, say your friend has 2 dice and rolls craps 10 times out of 20. Are the dice full of it? The probability of a 7 with 2 dice is 1/6 of the time, or .166667. This Cubiprocle tells you that you should be suspicious, as you would expect to see 10 sevens of 20 less than 1% of the time, this is a significant event and something highly unusual, because it falls outside of two standard deviations. You should probably go play dice elsewhere.

Let's say in the county elections, 10000 voted total, 1000 in your district. Watson got 4800 votes to Sherlock's 5200. In your district, Sherlock got 900 votes. Was it rigged in your district? Of 9000 votes not in your district, Sherlock got 5200-900 = 4300, or 47.78 %. Let's use that for the fairness test, if your district is similar to the rest. Cubiprocle says something is definitely fishy, the odds of getting 900 or more votes in your district is hogwash.

Note that in this case, a normal distribution was used to approximate the cumulative binomial summation, after about 500 calculations, it could take a long time otherwise, this means that under some conditions the experiment may be ill conditioned, however, for most purposes you should be able to get meaningful results. So aside from this disclaimer, please don't use Cubiprocle for anything critical, it is for infotainment purposes only.

We hope you enjoy Cubiprocle!