Reap, I know you have a very interesting method for looking at the rules of the game but I don't understand how you can argue against the literal definition of a word
Random - governed by or involving equal chances for each item.
With a random answer, you can't say a certain answer is likely. What you are describing is a Biased Selection which by definition isn't random.
You are going to come to a dead end somewhere. Where by your logic, there won't even be a method to use to make the random selection.
No you won't. With just a d6 and a d20, you can simulate a random outcome with equal chances for any number N of choices. For some numbers N you have to assign a NULL value to certain rolls and roll again if one comes up, but you will get a result eventually. (And the probability of hitting NULL value after NULL value after NULL value etc. approaches zero very fast.)
Throw in some other dice like a d8 or d12 and you have even better ways to achieve this.
Edit:
Of course, you since you can write any integer number in base N, using an N sided die will suffice. So you really only need one die of any kind. Heck, you could use coin throws (which really is just a d2) and solve the problem in binary!
The great Achilles'. Man, this was yet another poor name choice. Should have been like Akelets. They are getting too old for this.
Anyways, Haktos the Unscarred. Kind of a mind-boggling ability, huh? Well, not to despair. I'm going to teach you how to not only easily choose the value at random, but also how to do it in a way that puts the odds in your favor.
The jist is simple: You flip three coins. Label each one heads, two heads, or three heads result a number. The one you want most label as the two heads result, and your best backup label as the one heads result. These are the two most probable results. The one you want least, label as the three heads result.
And there you will easily be able to choose this effect at random and will have the best odds in your best favor.
I don't... I can't even begin to explain how little sense this makes.
You think that "probability" = "random"?
Sure. Why not. Fine.
What I don't get, though, is why you're bothering with coins at all? You seem to insist that the individual pieces of calculating that random value have to have even weight (such as dividing a d6 three ways) and stacking them in a way that creates uneven outcomes but that seems utterly unnecessary if your only guiding force is that "probability = random". Why not skip the "middle man" altogether, in that case?
"Let's roll a D20. If I roll a 1, it's *unwanted value 1*. If I roll a 2, it's *unwanted value 2*. Otherwise, it's *wanted value*"
Bam! Just upped the probability from 75% to 90%.
Why stop there, though. There are plenty of random number generators available online. Why not roll a digital d100? A d1,000?
Everyone MTG player on the face of the earth must be incredibly silly to see the word random and not immediately ignore it when a random number generator can give me a 99.9% chance (or better) of getting exactly what I want. I mean, it's a total mystery why Wizards even prints that word "random" on cards when anyone with a brain would just pull out an RNG app.
From where I am standing, the only difference between your coin-flipping method and my RNG method is that an inexperienced player may not notice what's happening with your method... which shouldn't bother you as you're not doing anything illegal. It's not like a judge will lay out ad-hoc rulings right there at the table based on their own feelings and disqualify you, right?
Is there any difference between the coin-flips and the RNG method that I'm not seeing? I mean, there are multiple possible options and one of them is chosen by random means. That meets all of the hallmarks of probability, which we all know = random.
Okay, well now you're not using a medium with 'equal sides' to the result. This now involves 'rules bending'.
Just like I said. This method is not securable and is going to come to a dead-end.
// Additionally, there are numbers that don't divide evenly at all.
That's why you assign a NULL value to any result that is higher than N. If you get a result that's higher than N you simply start over. This will end eventually, and the result you finally get does have equal probability to any other possible result.
Lets look at a simple example:
N=11
You only have a d6. The next number divisible by 6 is 12 = 2*6. Roll the d6 as a d2, 1-3 equals 0, 2-6 equals 1. Then roll the d6 again, this time as a d6. Multiply the first result by 6 and add the second. If you get 12, do the whole process again, d2 and d6.
You can apply this method to any number N. Of course, you will get a result faster if you reduce the number of NULL values as much as possible. In the example it is possible to get a string of 12s, yes, but the probability for continuing it will always be just 1/12. Getting three 12s in a row is already very unlikely (1/12^3 = 1/1728).
You could also simply pull out a phone and use https://www.random.org/, which produces true random numbers without the bias you'd get from rolling dice or flipping coins. Random.org uses atmospheric noise as a basis for its random numbers, and is often used to produce random numbers that have millions of dollars riding on their outcome, such as a lottery selecting its winning numbers. They are very heavily invested in having good randomness.
With a random answer, you can't say a certain answer is likely. What you are describing is a Biased Selection which by definition isn't random.
No you won't. With just a d6 and a d20, you can simulate a random outcome with equal chances for any number N of choices. For some numbers N you have to assign a NULL value to certain rolls and roll again if one comes up, but you will get a result eventually. (And the probability of hitting NULL value after NULL value after NULL value etc. approaches zero very fast.)
Throw in some other dice like a d8 or d12 and you have even better ways to achieve this.
Edit:
Of course, you since you can write any integer number in base N, using an N sided die will suffice. So you really only need one die of any kind. Heck, you could use coin throws (which really is just a d2) and solve the problem in binary!
Former Rules Advisor
"Everything's better with pirates." - Lodge
(The Gamers: Dorkness Rising)
"Any sufficiently analyzed magic is indistinguishable from science."
(Girl Genius - Fairy Tale Theater Break - Cinderella, end of volume 8)
Two Score, Minus Two or: A Stargate Tail
(Image by totallynotabrony)
Just like I said. This method is not securable and is going to come to a dead-end.
// Additionally, there are numbers that don't divide evenly at all.
I don't... I can't even begin to explain how little sense this makes.
You think that "probability" = "random"?
Sure. Why not. Fine.
What I don't get, though, is why you're bothering with coins at all? You seem to insist that the individual pieces of calculating that random value have to have even weight (such as dividing a d6 three ways) and stacking them in a way that creates uneven outcomes but that seems utterly unnecessary if your only guiding force is that "probability = random". Why not skip the "middle man" altogether, in that case?
"Let's roll a D20. If I roll a 1, it's *unwanted value 1*. If I roll a 2, it's *unwanted value 2*. Otherwise, it's *wanted value*"
Bam! Just upped the probability from 75% to 90%.
Why stop there, though. There are plenty of random number generators available online. Why not roll a digital d100? A d1,000?
Everyone MTG player on the face of the earth must be incredibly silly to see the word random and not immediately ignore it when a random number generator can give me a 99.9% chance (or better) of getting exactly what I want. I mean, it's a total mystery why Wizards even prints that word "random" on cards when anyone with a brain would just pull out an RNG app.
From where I am standing, the only difference between your coin-flipping method and my RNG method is that an inexperienced player may not notice what's happening with your method... which shouldn't bother you as you're not doing anything illegal. It's not like a judge will lay out ad-hoc rulings right there at the table based on their own feelings and disqualify you, right?
Is there any difference between the coin-flips and the RNG method that I'm not seeing? I mean, there are multiple possible options and one of them is chosen by random means. That meets all of the hallmarks of probability, which we all know = random.
That's why you assign a NULL value to any result that is higher than N. If you get a result that's higher than N you simply start over. This will end eventually, and the result you finally get does have equal probability to any other possible result.
Lets look at a simple example:
N=11
You only have a d6. The next number divisible by 6 is 12 = 2*6. Roll the d6 as a d2, 1-3 equals 0, 2-6 equals 1. Then roll the d6 again, this time as a d6. Multiply the first result by 6 and add the second. If you get 12, do the whole process again, d2 and d6.
result probability
1: 1/2 * 1/6 = 1/12
2: 1/2 * 1/6 = 1/12
3: 1/2 * 1/6 = 1/12
4: 1/2 * 1/6 = 1/12
5: 1/2 * 1/6 = 1/12
6: 1/2 * 1/6 = 1/12
7: 1/2 * 1/6 = 1/12
8: 1/2 * 1/6 = 1/12
9: 1/2 * 1/6 = 1/12
10: 1/2 * 1/6 = 1/12
11: 1/2 * 1/6 = 1/12
12: 1/2 * 1/6 = 1/12 (if you get this result, start again)
You can apply this method to any number N. Of course, you will get a result faster if you reduce the number of NULL values as much as possible. In the example it is possible to get a string of 12s, yes, but the probability for continuing it will always be just 1/12. Getting three 12s in a row is already very unlikely (1/12^3 = 1/1728).
Former Rules Advisor
"Everything's better with pirates." - Lodge
(The Gamers: Dorkness Rising)
"Any sufficiently analyzed magic is indistinguishable from science."
(Girl Genius - Fairy Tale Theater Break - Cinderella, end of volume 8)
Two Score, Minus Two or: A Stargate Tail
(Image by totallynotabrony)