Someone in the party just got killed, was flipping through Monster Manual to look for cool races for his new character. He mentions "Hey, X race looks cool" (I forget which one), "With a LA of x."
I mentioned, "That's probably a bad race since if a monster has more than 1 HD they count as their class levels."
The DM looked at me and told me, "No, you're dumb, you're wrong." (this was after I chased down and killed an enemy who he thought he was going to get away via flying away. It was a perfectly reasonable decision -- the enemies were intelligent, apparently slavers who worked for a government that would get mad at us if they found out we murdered his friends. Why the hell would we just let him get away? Even a good character would be willing to kill them to protect his party. Of course, not all good characters will, but mine had a decent int score, so... Anyways, he looked really ****** afterwards, especially since he had made character sheets for the enemies for some reason.)
I looked through the MM a bit, then said, "I'm going to play a dragon then, all right? This only has an LA of 5, and 12 hd, so it should be legal." The DM just told me 'no.' In retrospect, that was probably a bad move. Its not always good to argue against your DM. Instead, I should've suggested the player who was rerolling to use that.
Not really related to rulings, just dumb players. The DM agrees wholeheartedly with our BA Statistics guy that statistics will effect the probability of a dice roll/coin flip if you care about the previous results.
For example, flipping a coin (and assuming its fair) and getting heads three times in a row is a 1/8 chance. Given that you flip two heads in a row, your chance of getting another heads is 1/2. Pretty simple. He tried to argue that mathematically is was still 1/8 somehow. After me and another player convinced him that was wrong, he stated that the chances of getting a heads (assuming a fair coin) given that you got two heads before was less than 50%, because in a lot of coin flips the result tends to be somewhere around 50/50, so it would correct itself to achieve that...or something. I even stated we're assuming its a fair coin, and he stated the coin is still fair even if he assumes it has anything other than a 50/50 chance on the next rol,l due to what you know about its flips already affecting the coin flip.
The DM argued that given that you flipped two heads in a row on a fair coin, your chance of getting another heads is 1/8, because "Just because you flipped/rolled already doesn't change the probability that it will happen. Its the law of probability." (this is actually a summarization of several arguments I've had with him on this subject, he pretty much dismisses my arguments because the BA Statistics guy agrees with him and he took Calculus in high school 8-9 years ago) Jesus christ...then he told me to look up Conditional Probability on wikipedia. Did he even freaking read the article? I would be convinced this guy was just trolling me if I didn't know him so well.
I really think I'm just going to stop playing with this group. A lot of them are friends, and I'd rather just stay friends rather than get mad at them over DnD.
The DM argued that given that you flipped two heads in a row on a fair coin, your chance of getting another heads is 1/8, because "Just because you flipped/rolled already doesn't change the probability that it will happen. Its the law of probability." (this is actually a summarization of several arguments I've had with him on this subject, he pretty much dismisses my arguments because the BA Statistics guy agrees with him and he took Calculus in high school 8-9 years ago) Jesus christ...then he told me to look up Conditional Probability on wikipedia. Did he even freaking read the article? I would be convinced this guy was just trolling me if I didn't know him so well.
I'm quite alarmed that you have a professional statistician in your group who doesn't understand the principle of statistical independence. In particular, he really ought to understand that statistical independence is mathematically defined in terms of conditional probability; that is, p(A|B) = p(A). That's really a fairly critical point, and it's one that should have been covered in Stats 101 and used in every other course he took toward his major.
Your DM and his statistician cohort are referring, sort of, to the Law of Large Numbers, which is also known as the Law of Averages. That principle actually says that, given a sufficiently large number of trials, the sample distribution will come to resemble the "true" distribution of the event. Your DM's mistake is a common misconception--so common that it has its own name, the Monte Carlo fallacy.
Anyway, you're still correct here. On a fair coin (or a fair die), the likelihood of getting the same result 100 times in a row is extremely low. However, if you know your coin/die is fair, the outcomes of the next event (roll, flip, etc.) is not affected by the previous outcomes, no matter how unlikely those outcomes were. That's the plain-english version of the p(A|B) bit above.
I really think I'm just going to stop playing with this group. A lot of them are friends, and I'd rather just stay friends rather than get mad at them over DnD.
That's a good idea. It's better than losing friends over a game, anyway.
If so....never EVER go to the school that guy got his BA from...they don't know how to teach, nor tell if someone actually knows the diff.
Not really related to rulings, just dumb players. The DM agrees wholeheartedly with our BA Statistics guy that statistics will effect the probability of a dice roll/coin flip if you care about the previous results.
For example, flipping a coin (and assuming its fair) and getting heads three times in a row is a 1/8 chance. Given that you flip two heads in a row, your chance of getting another heads is 1/2. Pretty simple. He tried to argue that mathematically is was still 1/8 somehow. After me and another player convinced him that was wrong, he stated that the chances of getting a heads (assuming a fair coin) given that you got two heads before was less than 50%, because in a lot of coin flips the result tends to be somewhere around 50/50, so it would correct itself to achieve that...or something. I even stated we're assuming its a fair coin, and he stated the coin is still fair even if he assumes it has anything other than a 50/50 chance on the next rol,l due to what you know about its flips already affecting the coin flip.
The DM argued that given that you flipped two heads in a row on a fair coin, your chance of getting another heads is 1/8, because "Just because you flipped/rolled already doesn't change the probability that it will happen. Its the law of probability." (this is actually a summarization of several arguments I've had with him on this subject, he pretty much dismisses my arguments because the BA Statistics guy agrees with him and he took Calculus in high school 8-9 years ago) Jesus christ...then he told me to look up Conditional Probability on wikipedia. Did he even freaking read the article? I would be convinced this guy was just trolling me if I didn't know him so well.
This:
Not that you're particularly looking for confirmation, but you are right about ECL. ECL is class levels + level adjustment + racial hit dice, meaning the dragon you mention is considered the equivalent of a 17th-level character before receiving any class levels.
I'm quite alarmed that you have a professional statistician in your group who doesn't understand the principle of statistical independence. In particular, he really ought to understand that statistical independence is mathematically defined in terms of conditional probability; that is, p(A|B) = p(A). That's really a fairly critical point, and it's one that should have been covered in Stats 101 and used in every other course he took toward his major.
Your DM and his statistician cohort are referring, sort of, to the Law of Large Numbers, which is also known as the Law of Averages. That principle actually says that, given a sufficiently large number of trials, the sample distribution will come to resemble the "true" distribution of the event. Your DM's mistake is a common misconception--so common that it has its own name, the Monte Carlo fallacy.
Anyway, you're still correct here. On a fair coin (or a fair die), the likelihood of getting the same result 100 times in a row is extremely low. However, if you know your coin/die is fair, the outcomes of the next event (roll, flip, etc.) is not affected by the previous outcomes, no matter how unlikely those outcomes were. That's the plain-english version of the p(A|B) bit above.
That's a good idea. It's better than losing friends over a game, anyway.
Short Version: If you have a 50/50 chance on every roll, the next one will always have that same chance. He's thinking of the odds of getting a specific sequence of results. Even then, the odds of acheiving that sequence go UP as you go, as you reduce the amount of possible mistakes left.
The DM looked at me and told me, "No, you're dumb, you're wrong."
He's wrong and he used an ad hominem as well.
I can see why you feel like ranting about this.
The DM argued that given that you flipped two heads in a row on a fair coin, your chance of getting another heads is 1/8, because "Just because you flipped/rolled already doesn't change the probability that it will happen. Its the law of probability."
Ah, the good old gambler's fallacy (previously linked). This is a text book example that's used in the article.
No matter the number of times that you previously got heads in a row, the odds of getting it the next time is 50%.
(this is actually a summarization of several arguments I've had with him on this subject, he pretty much dismisses my arguments because the BA Statistics guy agrees with him and he took Calculus in high school 8-9 years ago)
An appeal to authority as well.
:D