I'm calling it now: within the next two weeks, someone else is going to start a thread saying, "OMG. I had a killer board, my opponent cast DM and hit ALL THREE of my huge minions, turning them into trash!!! RIGGED!" Then someone else is going to respond saying, "Yup, happened to me, too. Bli$$ard needs to make money," while a third talks about ZTG and Blizzard's microtransaction patent and how that proves the whole game is a scam. It'll end with another person declaring "Blizzard's trying to keep everyone at a 50% winrate," while bragging about how they have created a deck with a 65% winrate.

I've played against mage a few times where they used this, and it tended to hit the same minion 2 or 3 times, even when I had 4 or 5 minions on the board. I was pretty thankful for that lol.

Again, people fail to get that Blizzard has no reason to rig this. If they're "sticking it to you," they're also benefiting your opponent. For no logical reason. Please understand a very simple fact: Blizzard could not care less who wins the game you're playing. You are not special (at least not to them). All they care about is that you ARE playing (and hopefully spending money).

Also understand that the posters on this site are not a random sample of all players: they are players who are more likely to discuss (and complain about) the game. So seeing a sizable number of "yeah, that happened to me, too" offers no real data about the randomness of the card.

If you really want to test this, start collecting REAL data, as in several hundred observations, tracking how many enemy minions were on the board and which ones got hit. Do that and I guarantee you the data will show that the targeting is random.

I guess its just probability. It's like rolling a single dice three times in a row. The probability of throwing a given number of pips (the dots on a dice) on each consecutive throw is always 1/6, so in theory throwing let's say three "sixes" in a row is equally probable as throwing any other combination of pips (e.g. three different ones, or two same ones and one different). If the logic is that on each shot the target is chosen randomly from all viable targets, all combinations are possible. As TallStranger rightly stated, the only way to prove this is not behaving in an expected manner is to track it across a large sample of data...

Personally I don't see any promises of absolute randomness from Blizzard. I mean they have to report odds with regards to pack purchases by law and do so. But I don't see any reason for them to be bound to absolute randomness. When cards don't quite work for a meta or game I could see them doing small tweaks to how the card plays without disclosing rather then remove the card or nurf it.

I am not saying that this is occurring here. But blind trust is a corporation is arguably drinking more koolaid then questioning and checking occasionally.

Obsessive questioning is a waste of time but obsessive trust and mocking of questioning is meh.

Probably confirmation bias, but ye, I feel like it's been happening more than it would with Arcane Missiles as well - Probably it is much more memorable because the effect is "bigger" and more visual than just doing a single point of damage.

However, to the arguments against this being "Why would blizzard do this", I would counter that argument with "by accident" - bugs happen after all, and it's not like we haven't seen oddball stuff before. It is perfectly reasonable to think that there could be a bug in the system, e.g. that after the first transformation, the minion being hit first would be considered by the calculation as 2 different minions, and thus the chance to hit it again would be 66% (2/3) instead of 50% (1/2).

Probably confirmation bias, but ye, I feel like it's been happening more than it would with Arcane Missiles as well - Probably it is much more memorable because the effect is "bigger" and more visual than just doing a single point of damage.

However, to the arguments against this being "Why would blizzard do this", I would counter that argument with "by accident" - bugs happen after all, and it's not like we haven't seen oddball stuff before. It is perfectly reasonable to think that there could be a bug in the system, e.g. that after the first transformation, the minion being hit first would be considered by the calculation as 2 different minions, and thus the chance to hit it again would be 66% (2/3) instead of 50% (1/2).

Okay, I guess in theory that could happen. My counter would be that (and, again I'm not a coder) this has to be a remarkably easy code to write. I'd be stunned if Blizzard managed to screw that up.

The good news is that such an effect should be incredibly easy to test: the difference between a 50% chance and a 66% chance is significant enough that a reasonably-sized test would detect it very quickly. So to those who are convinced this is happening, get a friend to agree to help you. Create a junk deck and do some testing. Always make sure that you have only two minions on the board when your friend plays DM, and track the results. You could probably get a sizable number of tests done in an hour. Post the results. Anything's better than the haphazard, anecdotal stuff we're getting right now.

Personally I don't see any promises of absolute randomness from Blizzard. I mean they have to report odds with regards to pack purchases by law and do so. But I don't see any reason for them to be bound to absolute randomness. When cards don't quite work for a meta or game I could see them doing small tweaks to how the card plays without disclosing rather then remove the card or nurf it.

I am not saying that this is occurring here. But blind trust is a corporation is arguably drinking more koolaid then questioning and checking occasionally.

Obsessive questioning is a waste of time but obsessive trust and mocking of questioning is meh.

First off, I'm not mocking anyone.

Second, It's got nothing to do with "trusting a corporation" (though any expert in corporate behavior will tell you that the vast majority of them are extraordinarily cautious). It's got to do with basic incentive: there simply isn't one. Making the missiles non-random necessarily favors one player (the target) over another (the caster). Why? Again, Blizzard has exactly zero interest in who wins your game.

Third, the card DOES say "random enemy minions." Blizzard is, in fact, stating the effect is random. Any "tweak" you describe would make it, BY DEFINITION, non-random. So why would Blizzard risk discovery that they were lying about this, especially when they have other, more above-board ways of addressing any problem? If it were too powerful, they could change mana cost and/or number of missiles to balance it. Doing so is 1000x better than lying to the fanbase.

Fourth, there's no way to call this alleged behavior a "small tweak." If the examples given here are any indication (and again, they don't even come close to a statistically-meaningful sample), it is a massive (and easily detectable) one.

Probably confirmation bias, but ye, I feel like it's been happening more than it would with Arcane Missiles as well - Probably it is much more memorable because the effect is "bigger" and more visual than just doing a single point of damage.

However, to the arguments against this being "Why would blizzard do this", I would counter that argument with "by accident" - bugs happen after all, and it's not like we haven't seen oddball stuff before. It is perfectly reasonable to think that there could be a bug in the system, e.g. that after the first transformation, the minion being hit first would be considered by the calculation as 2 different minions, and thus the chance to hit it again would be 66% (2/3) instead of 50% (1/2).

Okay, I guess in theory that could happen. My counter would be that (and, again I'm not a coder) this has to be a remarkably easy code to write. I'd be stunned if Blizzard managed to screw that up.

The good news is that such an effect should be incredibly easy to test: the difference between a 50% chance and a 66% chance is significant enough that a reasonably-sized test would detect it very quickly. So to those who are convinced this is happening, get a friend to agree to help you. Create a junk deck and do some testing. Always make sure that you have only two minions on the board when your friend plays DM, and track the results. You could probably get a sizable number of tests done in an hour. Post the results. Anything's better than the haphazard, anecdotal stuff we're getting right now.

Probably confirmation bias, but ye, I feel like it's been happening more than it would with Arcane Missiles as well - Probably it is much more memorable because the effect is "bigger" and more visual than just doing a single point of damage.

However, to the arguments against this being "Why would blizzard do this", I would counter that argument with "by accident" - bugs happen after all, and it's not like we haven't seen oddball stuff before. It is perfectly reasonable to think that there could be a bug in the system, e.g. that after the first transformation, the minion being hit first would be considered by the calculation as 2 different minions, and thus the chance to hit it again would be 66% (2/3) instead of 50% (1/2).

Okay, I guess in theory that could happen. My counter would be that (and, again I'm not a coder) this has to be a remarkably easy code to write. I'd be stunned if Blizzard managed to screw that up.

The good news is that such an effect should be incredibly easy to test: the difference between a 50% chance and a 66% chance is significant enough that a reasonably-sized test would detect it very quickly. So to those who are convinced this is happening, get a friend to agree to help you. Create a junk deck and do some testing. Always make sure that you have only two minions on the board when your friend plays DM, and track the results. You could probably get a sizable number of tests done in an hour. Post the results. Anything's better than the haphazard, anecdotal stuff we're getting right now.

Hey, I got bored and decided to do this on lunch.

I only got up to 56 casts, which is not really a big enough sample size to be fully accurate, but I got pretty close to what you'd expect.

So basically, our theory is that it should be 25%, based on normal probability. Acolyte's theory is that it could be counting it as a new minion, giving it a 66% chance to be hit. That would make the total chance 43.5% if he's correct.

So I chose only 2-drop or higher minions so that they could devolve twice, giving them the maximum chance to be hit using his theory both times.

Out of the 56 casts:

4 had to be discarded because they devolved into a Dormant minion on the first roll, meaning it was impossible to hit it twice.

37 casts hit both minions.

15 casts hit only 1 minion.

So, that comes out to 15/52, or 28.8%. That makes it slightly higher than the expected average, but since I only did 50 casts, I'd say a range of 3-4% points either way falls within acceptable variance.

What doesn't fall within that variance is the other theory which would be 15.5% higher than the tested result.

Conclusion: Devolving Missiles is not bugged, if there are two minions it has a 25% chance to hit one of them 3x and the rest of the thread doesn't understand how RNG works.

If someone wants to see my replays I have them uploaded. I'm just not going to spam up my post with a bunch of links.

Probably confirmation bias, but ye, I feel like it's been happening more than it would with Arcane Missiles as well - Probably it is much more memorable because the effect is "bigger" and more visual than just doing a single point of damage.

However, to the arguments against this being "Why would blizzard do this", I would counter that argument with "by accident" - bugs happen after all, and it's not like we haven't seen oddball stuff before. It is perfectly reasonable to think that there could be a bug in the system, e.g. that after the first transformation, the minion being hit first would be considered by the calculation as 2 different minions, and thus the chance to hit it again would be 66% (2/3) instead of 50% (1/2).

Okay, I guess in theory that could happen. My counter would be that (and, again I'm not a coder) this has to be a remarkably easy code to write. I'd be stunned if Blizzard managed to screw that up.

The good news is that such an effect should be incredibly easy to test: the difference between a 50% chance and a 66% chance is significant enough that a reasonably-sized test would detect it very quickly. So to those who are convinced this is happening, get a friend to agree to help you. Create a junk deck and do some testing. Always make sure that you have only two minions on the board when your friend plays DM, and track the results. You could probably get a sizable number of tests done in an hour. Post the results. Anything's better than the haphazard, anecdotal stuff we're getting right now.

Hey, I got bored and decided to do this on lunch.

I only got up to 56 casts, which is not really a big enough sample size to be fully accurate, but I got pretty close to what you'd expect.

So basically, our theory is that it should be 25%, based on normal probability. Acolyte's theory is that it could be counting it as a new minion, giving it a 66% chance to be hit. That would make the total chance 43.5% if he's correct.

So I chose only 2-drop or higher minions so that they could devolve twice, giving them the maximum chance to be hit using his theory both times.

Out of the 56 casts:

4 had to be discarded because they devolved into a Dormant minion on the first roll, meaning it was impossible to hit it twice.

37 casts hit both minions.

15 casts hit only 1 minion.

So, that comes out to 15/52, or 28.8%. That makes it slightly higher than the expected average, but since I only did 50 casts, I'd say a range of 3-4% points either way falls within acceptable variance.

What doesn't fall within that variance is the other theory which would be 15.5% higher than the tested result.

Conclusion: Devolving Missiles is not bugged, if there are two minions it has a 25% chance to hit one of them 3x and the rest of the thread doesn't understand how RNG works.

If someone wants to see my replays I have them uploaded. I'm just not going to spam up my post with a bunch of links.

Yep, certainly within a pretty expected level randomness. Variance of a binomial equation is N*p*q, where p is the probability of success, and q is the probability of failure. Here, that'll lead to 9.75, which means there's a standard deviation of about 3.12. Within this small number of trials, you're within 1 standard deviation.

//

Here's the other thing: the probability of having 15 or fewer out of 52 trials hit the same target if the chance of hitting the same target 3 times is 25%, is 0.791. If the real chance is 43.5%? The probability of 15 or fewer successes drops down to 0.022. There's pretty much no way you get this kind of result if Devolving Missiles is bugged.

Which most of us already knew. There's a kind of pleasure in conducting an experiment, debunking some bogus conspiracy theory.

Probably confirmation bias, but ye, I feel like it's been happening more than it would with Arcane Missiles as well - Probably it is much more memorable because the effect is "bigger" and more visual than just doing a single point of damage.

However, to the arguments against this being "Why would blizzard do this", I would counter that argument with "by accident" - bugs happen after all, and it's not like we haven't seen oddball stuff before. It is perfectly reasonable to think that there could be a bug in the system, e.g. that after the first transformation, the minion being hit first would be considered by the calculation as 2 different minions, and thus the chance to hit it again would be 66% (2/3) instead of 50% (1/2).

Okay, I guess in theory that could happen. My counter would be that (and, again I'm not a coder) this has to be a remarkably easy code to write. I'd be stunned if Blizzard managed to screw that up.

The good news is that such an effect should be incredibly easy to test: the difference between a 50% chance and a 66% chance is significant enough that a reasonably-sized test would detect it very quickly. So to those who are convinced this is happening, get a friend to agree to help you. Create a junk deck and do some testing. Always make sure that you have only two minions on the board when your friend plays DM, and track the results. You could probably get a sizable number of tests done in an hour. Post the results. Anything's better than the haphazard, anecdotal stuff we're getting right now.

Hey, I got bored and decided to do this on lunch.

I only got up to 56 casts, which is not really a big enough sample size to be fully accurate, but I got pretty close to what you'd expect.

So basically, our theory is that it should be 25%, based on normal probability. Acolyte's theory is that it could be counting it as a new minion, giving it a 66% chance to be hit. That would make the total chance 43.5% if he's correct.

So I chose only 2-drop or higher minions so that they could devolve twice, giving them the maximum chance to be hit using his theory both times.

Out of the 56 casts:

4 had to be discarded because they devolved into a Dormant minion on the first roll, meaning it was impossible to hit it twice.

37 casts hit both minions.

15 casts hit only 1 minion.

So, that comes out to 15/52, or 28.8%. That makes it slightly higher than the expected average, but since I only did 50 casts, I'd say a range of 3-4% points either way falls within acceptable variance.

What doesn't fall within that variance is the other theory which would be 15.5% higher than the tested result.

Conclusion: Devolving Missiles is not bugged, if there are two minions it has a 25% chance to hit one of them 3x and the rest of the thread doesn't understand how RNG works.

If someone wants to see my replays I have them uploaded. I'm just not going to spam up my post with a bunch of links.

Great job. Nice to see people do actual theory testing, rather than mindless conspiracy-theorizing!

Not only does this game bring the best out of people (btw: always accept a friend request from a face hunter you just won against) but it also helps you progress towards acute paranoia. It’s a wonderful achievement, blizz can be proud.

Not only does this game bring the best out of people (btw: always accept a friend request from a face hunter you just won against) but it also helps you progress towards acute paranoia. It’s a wonderful achievement, blizz can be proud.

True, true - I'm usually accepting those requests just for fun and to get a feel of the opponent's salty tears :) Last one wished me to "die of Covid"...

Oh, sorry dude. Soooo many people ACTUALLY make that argument that I wrongly assumed you were doing so as well.

I'm calling it now: within the next two weeks, someone else is going to start a thread saying, "OMG. I had a killer board, my opponent cast DM and hit ALL THREE of my huge minions, turning them into trash!!! RIGGED!" Then someone else is going to respond saying, "Yup, happened to me, too. Bli$$ard needs to make money," while a third talks about ZTG and Blizzard's microtransaction patent and how that proves the whole game is a scam. It'll end with another person declaring "Blizzard's trying to keep everyone at a 50% winrate," while bragging about how they have created a deck with a 65% winrate.

This forum is catnip for lunatics.

I've played against mage a few times where they used this, and it tended to hit the same minion 2 or 3 times, even when I had 4 or 5 minions on the board. I was pretty thankful for that lol.

Data sample size of just 3 so far.

First event: cast missiles with 2 minions on board. All 3 hit same minion.

Second event: cast missiles with 3 minions on board. All 3 hit same minion.

Event 3: 2 minions on board, hit 1 twice and the other once.

Event 4: 2 minions on board, hit 1 twice and the other once.

Event 5: 5 minions on board, hit 3 different minions.

Not playing much lately but will report back with more data as available. But it seems so far like there's nothing to see here.

Just had, with 4 targets, 6 missiles hit the same target.

Lol people are really trying to collect "the data" :D

it happened to me plenty of times even when opponent had 5 cards on board

Again, people fail to get that Blizzard has no reason to rig this. If they're "sticking it to you," they're also benefiting your opponent. For no logical reason. Please understand a very simple fact: Blizzard could not care less who wins the game you're playing. You are not special (at least not to them). All they care about is that you ARE playing (and hopefully spending money).

Also understand that the posters on this site are not a random sample of all players: they are players who are more likely to discuss (and complain about) the game. So seeing a sizable number of "yeah, that happened to me, too" offers no real data about the randomness of the card.

If you really want to test this, start collecting REAL data, as in several hundred observations, tracking how many enemy minions were on the board and which ones got hit. Do that and I guarantee you the data will show that the targeting is random.

I guess its just probability. It's like rolling a single dice three times in a row. The probability of throwing a given number of pips (the dots on a dice) on each consecutive throw is always 1/6, so in theory throwing let's say three "sixes" in a row is equally probable as throwing any other combination of pips (e.g. three different ones, or two same ones and one different). If the logic is that on each shot the target is chosen randomly from all viable targets, all combinations are possible. As TallStranger rightly stated, the only way to prove this is not behaving in an expected manner is to track it across a large sample of data...

Personally I don't see any promises of absolute randomness from Blizzard. I mean they have to report odds with regards to pack purchases by law and do so. But I don't see any reason for them to be bound to absolute randomness. When cards don't quite work for a meta or game I could see them doing small tweaks to how the card plays without disclosing rather then remove the card or nurf it.

I am not saying that this is occurring here. But blind trust is a corporation is arguably drinking more koolaid then questioning and checking occasionally.

Obsessive questioning is a waste of time but obsessive trust and mocking of questioning is meh.

Probably confirmation bias, but ye, I feel like it's been happening more than it would with Arcane Missiles as well - Probably it is much more memorable because the effect is "bigger" and more visual than just doing a single point of damage.

However, to the arguments against this being "Why would blizzard do this", I would counter that argument with "by accident" - bugs happen after all, and it's not like we haven't seen oddball stuff before. It is perfectly reasonable to think that there could be a bug in the system, e.g. that after the first transformation, the minion being hit first would be considered by the calculation as 2 different minions, and thus the chance to hit it again would be 66% (2/3) instead of 50% (1/2).

Anyway, what do I know...

Okay, I guess in theory that could happen. My counter would be that (and, again I'm not a coder) this has to be a remarkably easy code to write. I'd be stunned if Blizzard managed to screw that up.

The good news is that such an effect should be incredibly easy to test: the difference between a 50% chance and a 66% chance is significant enough that a reasonably-sized test would detect it very quickly. So to those who are convinced this is happening, get a friend to agree to help you. Create a junk deck and do some testing. Always make sure that you have only two minions on the board when your friend plays DM, and track the results. You could probably get a sizable number of tests done in an hour. Post the results. Anything's better than the haphazard, anecdotal stuff we're getting right now.

First off, I'm not mocking anyone.

Second, It's got nothing to do with "trusting a corporation" (though any expert in corporate behavior will tell you that the vast majority of them are extraordinarily cautious). It's got to do with basic incentive: there simply isn't one. Making the missiles non-random necessarily favors one player (the target) over another (the caster). Why? Again, Blizzard has exactly zero interest in who wins your game.

Third, the card DOES say "random enemy minions." Blizzard is, in fact, stating the effect is random. Any "tweak" you describe would make it, BY DEFINITION, non-random. So why would Blizzard risk discovery that they were lying about this, especially when they have other, more above-board ways of addressing any problem? If it were too powerful, they could change mana cost and/or number of missiles to balance it. Doing so is 1000x better than lying to the fanbase.

Fourth, there's no way to call this alleged behavior a "small tweak." If the examples given here are any indication (and again, they don't even come close to a statistically-meaningful sample), it is a massive (and easily detectable) one.

I could do that, if someone is interested

#kitoxme1576

Hey, I got bored and decided to do this on lunch.

I only got up to 56 casts, which is not really a big enough sample size to be fully accurate, but I got pretty close to what you'd expect.

So basically, our theory is that it should be 25%, based on normal probability. Acolyte's theory is that it could be counting it as a new minion, giving it a 66% chance to be hit. That would make the total chance 43.5% if he's correct.

So I chose only 2-drop or higher minions so that they could devolve twice, giving them the maximum chance to be hit using his theory both times.

Out of the 56 casts:

4 had to be discarded because they devolved into a Dormant minion on the first roll, meaning it was impossible to hit it twice.

37 casts hit both minions.

15 casts hit only 1 minion.

So, that comes out to 15/52, or 28.8%. That makes it slightly higher than the expected average, but since I only did 50 casts, I'd say a range of 3-4% points either way falls within acceptable variance.

What doesn't fall within that variance is the other theory which would be 15.5% higher than the tested result.

Conclusion: Devolving Missiles is not bugged, if there are two minions it has a 25% chance to hit one of them 3x and the rest of the thread doesn't understand how RNG works.

If someone wants to see my replays I have them uploaded. I'm just not going to spam up my post with a bunch of links.

I have never had missiles hitting a specific minion more than 2 times. This is nothing more than confirmation bias.

Always expect the unexpectable!

Yep, certainly within a pretty expected level randomness. Variance of a binomial equation is N*p*q, where p is the probability of success, and q is the probability of failure. Here, that'll lead to 9.75, which means there's a standard deviation of about 3.12. Within this small number of trials, you're within 1 standard deviation.

//

Here's the other thing: the probability of having 15 or fewer out of 52 trials hit the same target if the chance of hitting the same target 3 times is 25%, is 0.791. If the real chance is 43.5%? The probability of 15 or fewer successes drops down to 0.022. There's pretty much no way you get this kind of result if Devolving Missiles is bugged.

Which most of us already knew. There's a kind of pleasure in conducting an experiment, debunking some bogus conspiracy theory.

Great job. Nice to see people do actual theory testing, rather than mindless conspiracy-theorizing!

Not only does this game bring the best out of people (btw: always accept a friend request from a face hunter you just won against) but it also helps you progress towards acute paranoia. It’s a wonderful achievement, blizz can be proud.

Take a walk on the wild side...

True, true - I'm usually accepting those requests just for fun and to get a feel of the opponent's salty tears :) Last one wished me to "die of Covid"...