Love
Deity
This seems bad.
General Turn Discussion
- Thread starter Krill
- Start date
shadyforce
Chieftain
I question those odds. It looks like the percentages were calculated from the relative strengths shown but completely ignores the fact that the aggressors are all full health and the defenders are all weakened. Those 11.16 vs 11 type ~50% battles should actually be well in favour of the uninjured attacker.
Mojoqmeyvam
Warlord
- Joined
- Jul 25, 2009
- Messages
- 129
11.0 vs. 11.1 would give the 11.0 unit like 38% odds if both had 100hp. But if that 11.16 Knight was a CII knight (likely), then it would be at 93HP, and would die in 5 hits from a CI Knight at 100hp (11 Str). However, our CI knight would also die in 5 hits, so the listed odds of 49.3% for that 11 v. 11.16 fight look reasonable (we win 49.64% of each round of combat, so 49.3% chance of being first to 5 hits is true or sufficiently close to true). IIRC, odds are more likely to screw up with handling First Strikes properly than they are with messing up on HP totals.
According to those stats, we won all 4 low-odds attacks (10-30%), got a 30% withdrawal, won all 4 coin-flips (44-49%), and the 3 attacks that we were favored in, at 70-85%. Going by the odds given, that's 1-in-92000, ignoring the nearly-zero odds of winning the suicided cats. Those odds do require assuming that the RNG isn't streaky. Memphus needs a better defense than "the RNG was just streaky", though.
According to those stats, we won all 4 low-odds attacks (10-30%), got a 30% withdrawal, won all 4 coin-flips (44-49%), and the 3 attacks that we were favored in, at 70-85%. Going by the odds given, that's 1-in-92000, ignoring the nearly-zero odds of winning the suicided cats. Those odds do require assuming that the RNG isn't streaky. Memphus needs a better defense than "the RNG was just streaky", though.
Yeah I am traveling right now... I didn't do a turn update (havn't done many of them recently)
Those are the results. We should of lost 3-6 units in a 50-50 world. That is what I said to slaze in game.
As for the cav knights, we never killed cav knights, we killed cav musketmen.
Kaz mace stacks we won battles at low odds, but had troops in reserve had we not won.
The Galley was 33% odds, not 10%. Still not great.
Overall the RNG has caused us to lose alot less troops then we should of....but had we lost those troops there was more left.
I am not sure what else we are suppose to do? not attack? the question really is, is the RNG dependant or independant? I believe it is independant and therefore drawing conclusions from one battle to the next while it is easy to do, and get frustrated with, isnt the right way to do it.
This battle, and each attack within the battle is seperate and not depedant on the result before.
Going back to what M-H said, who attacks at low odds routinely? not too many people. And who truely understands the odds of the battle for how much dmg needs to be done each turn for each round, based on the injured units? the % arn't a true refelction of the actual odds. This is why using promos is also a skill so that units dont show up to early in the battle.
Example Knight uses STR shock and the pike goes away to the WE. Uses STR X2 and the WE goes away to another knight, odds on that knight are lower than the WE, which is lower than the pike. Breaking down all these paths and options isn't an easy task and takes alot of time. But understanding the mechanics fully via reading articles etc and you get better at this. I wrote the strategy up the last couple of times, when i had time.
So what else did you want sulla? a total breakdown of what unit attacked when and why? what the options were at any given time? I can re-create the matrix if you want...but it will take me a week until my traveling is done.
By the same slice though, Cav got a roll from the RNG (Random events) that gave thier muskets pinch right before a war with us. All thier rifles now have pinch + 2 other promos, since they paid to upgrade that way. We arguably got flanking for our knights.
Slaze also didn't know how we cut the road. I told him. (chariot which we then deleted), an unit that didn't attack that could of had we had worse rolls.
edit: at mojo. 1 in 92000 yes if the results are depedant. but they arn't. So after each battle the odds are reset.
Those are the results. We should of lost 3-6 units in a 50-50 world. That is what I said to slaze in game.
As for the cav knights, we never killed cav knights, we killed cav musketmen.
Kaz mace stacks we won battles at low odds, but had troops in reserve had we not won.
The Galley was 33% odds, not 10%. Still not great.
Overall the RNG has caused us to lose alot less troops then we should of....but had we lost those troops there was more left.
I am not sure what else we are suppose to do? not attack? the question really is, is the RNG dependant or independant? I believe it is independant and therefore drawing conclusions from one battle to the next while it is easy to do, and get frustrated with, isnt the right way to do it.
This battle, and each attack within the battle is seperate and not depedant on the result before.
Going back to what M-H said, who attacks at low odds routinely? not too many people. And who truely understands the odds of the battle for how much dmg needs to be done each turn for each round, based on the injured units? the % arn't a true refelction of the actual odds. This is why using promos is also a skill so that units dont show up to early in the battle.
Example Knight uses STR shock and the pike goes away to the WE. Uses STR X2 and the WE goes away to another knight, odds on that knight are lower than the WE, which is lower than the pike. Breaking down all these paths and options isn't an easy task and takes alot of time. But understanding the mechanics fully via reading articles etc and you get better at this. I wrote the strategy up the last couple of times, when i had time.
So what else did you want sulla? a total breakdown of what unit attacked when and why? what the options were at any given time? I can re-create the matrix if you want...but it will take me a week until my traveling is done.
By the same slice though, Cav got a roll from the RNG (Random events) that gave thier muskets pinch right before a war with us. All thier rifles now have pinch + 2 other promos, since they paid to upgrade that way. We arguably got flanking for our knights.
Slaze also didn't know how we cut the road. I told him. (chariot which we then deleted), an unit that didn't attack that could of had we had worse rolls.
edit: at mojo. 1 in 92000 yes if the results are depedant. but they arn't. So after each battle the odds are reset.
The quoted odds are indeed what's in the combat logs.
They seem unlikely, but to accuse someone publicly of being a cheat is very serious and requires evidence to back it up. Posting odds isn't proof, no matter what odds they are.
BTW: the figure of 1 in 100,000 is really misleading and it annoys when people post odds like this and claim it means anything. If you have a series of 14 events, then the probability of any distinct set of outcomes is extremely unlikely. But that doesn't mean anything. Plus you'd have to question that, if you were able to hack into Daveshack's server and fix the result, would you be silly enough to fix the irrelevant battles?
They seem unlikely, but to accuse someone publicly of being a cheat is very serious and requires evidence to back it up. Posting odds isn't proof, no matter what odds they are.
BTW: the figure of 1 in 100,000 is really misleading and it annoys when people post odds like this and claim it means anything. If you have a series of 14 events, then the probability of any distinct set of outcomes is extremely unlikely. But that doesn't mean anything. Plus you'd have to question that, if you were able to hack into Daveshack's server and fix the result, would you be silly enough to fix the irrelevant battles?
Mojoqmeyvam
Warlord
- Joined
- Jul 25, 2009
- Messages
- 129
It's a p=1.1-1.2*10^-5 that we'd do this good or better if the RNG is truly independent (adding winning the 5.0% odds for the HA that withdrew, and small odds of the catas withdrawing, which would make winning later much better). That's...statistically significant. Unless the RNG is streaky. In which case, well, creating a p-value is meaningless.
shadyforce
Chieftain
Ok, I guess I thought 11.xy might be half strength units with loads of modifiers but if they are near full health then I guess those odds are true.
Sullla
Patrician Roman Dictator
Look, I'd be willing to accept the argument that the RNG was streaky (even though the Civ4 one isn't) if this had been a one-time deal. But... it's not. It's not at all. The odds have always broken in SANCTA's favor, in every single combat throughout the game. Here's the complete list of every combat initiated by SANCTA against other human teams so far in the game, giving units, strengths, odds, and result:
SANCTA vs. Kaz (525BC)
axe v chariot, 4.5 v 4, 70.6%, win
SANCTA vs. Kaz (500BC)
axe v chariot, 4.29 v 3.82, 57.1%, win
SANCTA vs. Kaz (400BC)
axe v chariot, 4.52 v 4.40, 52%, win
SANCTA vs. Kaz (75BC)
trireme v trireme, 2 v 2.2, 32.2%, win
SANCTA vs. Kaz (50BC)
axe v mace, 6 v 8.8, 10.7%, win
chariot v mace, 4.4 v 8.8, 0.9%, lose (72/100 on mace)
archer v mace, 3 v 8, 0.2%, lose (52/100 on mace)
axe v horse archer, 5.5 v 6, 32.6%, win
axe v damaged mace, 5 v 6.3, 49.1%, win
longbow v catapult, 7.2 v 5, 78.3%, win
axe v chariot, 5.5 v 4.4, 75.1%, win
chariot v catapult, 4 v ?, 67.1%, win
SANCTA vs. Kaz (1AD)
chariot v war elephant, 4 v 8.8, 0.1%, withdraw (20% rate)
longbow v war elephant, 6 v 8.8, 13.5%, lose (36/100 on elephant)
chariot v maceman, 4 v 8.8, 0.1%, lose (100/100 on mace)
longbow v maceman, 6.6 v 8.8, 26.7%, win
spear v horse archer, 4 v 3.6, 73%, win
longbow v damaged war elephant, 6.6 v 3.2, 99%, win
axe v maceman, 5.5 v 8.8, 20.5%, lose
axe v damaged maceman, 6 v ?, 99.8%, win
SANCTA vs. Cavs (820AD)
knight v musket, 11 v 13.5, 25%, win
horse archer v musket, 6 v 13.5, 0.0%, withdraw (50% rate)
knight v musket, 11 v 13.5, 25%, win
SANCTA vs. Cavs (940AD)
cat v rifle, 5 v 24.5, 0.0%, lose (80/100 rifle)
cat v rifle, 5 v 19.6, 0.1%, lose
rifle v rifle, 15.4 v 22.4, 10.9%, win
knight v pike, 10 v 13.37, 21.7%, win
knight v pike, 11 v 13.37, 26.6%, win
rifle v knight, 14 v 10.23, 79.5%, win
knight v war elephant, 12 v 12.51, 47.5%, win
knight v pike, 11 v 12.05, 44.6%, win
spear v knight, 4.4 v 5.5, 24.7%, win
horse archer v war elephant, 6 v 11.42, 5.0%, withdrawl (30% rate)
knight v knight, 11 v 11.16, 49.3%, win
knight v knight, 12 v 10.32, 71.2%, win
knight v knight, 10 v 10.23, 48.8% win
axe v war elephant, 5 v 4.03, 85.2%, win
Here's my summary table:
Under 21% Odds ("Suicidal"): 12 combats, 2 wins, 7 losses, 3 withdraws
21-40% Odds ("Bad"): 8 combats, 8 victories
41-60% Odds ("Coinflip"): 7 combats, 7 victories
61-80% Odds ("Good"): 7 combats, 7 victories
81-99+% Odds ("Near-Certain"): 3 combats, 3 victories
Total: 37 combats, 27 victories, 7 losses, 3 withdraws
The "lowest odds" battle SANCTA has ever lost has been a 20.5% battle against Kaz. Every other battle with odds higher than that - EVERY OTHER BATTLE - your team has won. All of them. 8 battles under 40% odds, all victories. 7 coinflip battles, all victories. 7 battles in the "Good" odds range, all victories. In anything that's not a "sacrifice" combat, the team is 25-0 with zero losses. SANCTA has emphatically not been fighting combats in the pushover range (only two 99% battles), yet this team wins them all regardless. I mean... come on.
If we throw out the under 1% and over 99% battles, SANCTA's record is 25 wins, 2 losses, 1 withdraw. That's essentially a statistical impossibility. Doubly so when most of those combats were in the 20-80% range.
Now I feel bad about making this accusation. As I said, I've defended this team in the other forums on multiple occasions about the combat odds. And yes, I know that the in-game results don't perfectly match up with the real numbers - but they aren't THAT far off either. I don't want to do this, but in the interests of fairness, I have to speak up. It's simply not possible for these to be random results. I'm sorry Memphus, but I don't believe you. No one can possibly be this lucky, so consistently, over such a long period of time.
I'd like to see what the lurkers on SANCTA think about this. Take yourself out of the team context for a moment. If you saw these combat results, what would your reaction be? And how would you feel if you were on one of the competing teams, and you always - always! - lost every single battle over 20% odds?
SANCTA vs. Kaz (525BC)
axe v chariot, 4.5 v 4, 70.6%, win
SANCTA vs. Kaz (500BC)
axe v chariot, 4.29 v 3.82, 57.1%, win
SANCTA vs. Kaz (400BC)
axe v chariot, 4.52 v 4.40, 52%, win
SANCTA vs. Kaz (75BC)
trireme v trireme, 2 v 2.2, 32.2%, win
SANCTA vs. Kaz (50BC)
axe v mace, 6 v 8.8, 10.7%, win
chariot v mace, 4.4 v 8.8, 0.9%, lose (72/100 on mace)
archer v mace, 3 v 8, 0.2%, lose (52/100 on mace)
axe v horse archer, 5.5 v 6, 32.6%, win
axe v damaged mace, 5 v 6.3, 49.1%, win
longbow v catapult, 7.2 v 5, 78.3%, win
axe v chariot, 5.5 v 4.4, 75.1%, win
chariot v catapult, 4 v ?, 67.1%, win
SANCTA vs. Kaz (1AD)
chariot v war elephant, 4 v 8.8, 0.1%, withdraw (20% rate)
longbow v war elephant, 6 v 8.8, 13.5%, lose (36/100 on elephant)
chariot v maceman, 4 v 8.8, 0.1%, lose (100/100 on mace)
longbow v maceman, 6.6 v 8.8, 26.7%, win
spear v horse archer, 4 v 3.6, 73%, win
longbow v damaged war elephant, 6.6 v 3.2, 99%, win
axe v maceman, 5.5 v 8.8, 20.5%, lose
axe v damaged maceman, 6 v ?, 99.8%, win
SANCTA vs. Cavs (820AD)
knight v musket, 11 v 13.5, 25%, win
horse archer v musket, 6 v 13.5, 0.0%, withdraw (50% rate)
knight v musket, 11 v 13.5, 25%, win
SANCTA vs. Cavs (940AD)
cat v rifle, 5 v 24.5, 0.0%, lose (80/100 rifle)
cat v rifle, 5 v 19.6, 0.1%, lose
rifle v rifle, 15.4 v 22.4, 10.9%, win
knight v pike, 10 v 13.37, 21.7%, win
knight v pike, 11 v 13.37, 26.6%, win
rifle v knight, 14 v 10.23, 79.5%, win
knight v war elephant, 12 v 12.51, 47.5%, win
knight v pike, 11 v 12.05, 44.6%, win
spear v knight, 4.4 v 5.5, 24.7%, win
horse archer v war elephant, 6 v 11.42, 5.0%, withdrawl (30% rate)
knight v knight, 11 v 11.16, 49.3%, win
knight v knight, 12 v 10.32, 71.2%, win
knight v knight, 10 v 10.23, 48.8% win
axe v war elephant, 5 v 4.03, 85.2%, win
Here's my summary table:
Under 21% Odds ("Suicidal"): 12 combats, 2 wins, 7 losses, 3 withdraws
21-40% Odds ("Bad"): 8 combats, 8 victories
41-60% Odds ("Coinflip"): 7 combats, 7 victories
61-80% Odds ("Good"): 7 combats, 7 victories
81-99+% Odds ("Near-Certain"): 3 combats, 3 victories
Total: 37 combats, 27 victories, 7 losses, 3 withdraws
The "lowest odds" battle SANCTA has ever lost has been a 20.5% battle against Kaz. Every other battle with odds higher than that - EVERY OTHER BATTLE - your team has won. All of them. 8 battles under 40% odds, all victories. 7 coinflip battles, all victories. 7 battles in the "Good" odds range, all victories. In anything that's not a "sacrifice" combat, the team is 25-0 with zero losses. SANCTA has emphatically not been fighting combats in the pushover range (only two 99% battles), yet this team wins them all regardless. I mean... come on.
If we throw out the under 1% and over 99% battles, SANCTA's record is 25 wins, 2 losses, 1 withdraw. That's essentially a statistical impossibility. Doubly so when most of those combats were in the 20-80% range.
Now I feel bad about making this accusation. As I said, I've defended this team in the other forums on multiple occasions about the combat odds. And yes, I know that the in-game results don't perfectly match up with the real numbers - but they aren't THAT far off either. I don't want to do this, but in the interests of fairness, I have to speak up. It's simply not possible for these to be random results. I'm sorry Memphus, but I don't believe you. No one can possibly be this lucky, so consistently, over such a long period of time.
I'd like to see what the lurkers on SANCTA think about this. Take yourself out of the team context for a moment. If you saw these combat results, what would your reaction be? And how would you feel if you were on one of the competing teams, and you always - always! - lost every single battle over 20% odds?
I think it's a lot more complicated than that. First of all, the events are not independent. Each event directly affects the odds of the following events. Secondly, what would be more meaningful, though impossible to calculate, would be "what are the chances of the results being good enough to be accused of cheating". For example, if we had lost one of the 30% battles and won the rest then memphus would likely still be accused of cheating. And yet it's very hard to work these odds out because losing that battle will affect the odds of all of the following battles. One would have to calculate every series of events that would still get Memphus accused of cheating, which is not possible because it's not a black and white issue. But assuming it were possible, you'd have to calculate their odds, remembering that the odds listed above would not apply anymore because the events are dependent on each other, and then add them up. You'd get a result significantly larger than the 1 in 100 million or whatever.
Again, let me say that I'm not saying Memphus is or is not a cheat, and no one can because no one has any evidence except for some odds.
lurker's comment: Any suggestion how this supposed cheating has happened? I'm not aware of any ways of altering the combat odds in a pitboss environment, though I'm not up to date on what's happening in the multiplayer world.
OK, honestly, I would feel suspicious. I would be looking for ways that cheating would be possible. But until I found anything, I would like to hope I'd not reach a conclusion one way or the other.
So what is the plan forward then Sulla?
Here is what the outcomes are now with this:
1. We attack again and get lucky. Everyone on other teams get more upset
2. We attack again and lose a reasonable number of battles. (as soooo pointed out though what is reasonable? But don't tell me be because if you did and it happened then that would prove cheating... Or would it?
3. We attack again and get crushed. Them everyone will assume it is due to the scrutiny going on
4. Someone else plays and all the above situations arise again but it is now directed solely at me and my ability to "cheat" especially if that player listens to advice on which unit to attack with.
So now we are in a lose lose.
Btw soooo for your dependance, it is also then affected by other players on thier turn whose results we don't know. Just like our barb stats should be in there too.
Here is what the outcomes are now with this:
1. We attack again and get lucky. Everyone on other teams get more upset
2. We attack again and lose a reasonable number of battles. (as soooo pointed out though what is reasonable? But don't tell me be because if you did and it happened then that would prove cheating... Or would it?
3. We attack again and get crushed. Them everyone will assume it is due to the scrutiny going on
4. Someone else plays and all the above situations arise again but it is now directed solely at me and my ability to "cheat" especially if that player listens to advice on which unit to attack with.
So now we are in a lose lose.
Btw soooo for your dependance, it is also then affected by other players on thier turn whose results we don't know. Just like our barb stats should be in there too.
Mojoqmeyvam
Warlord
- Joined
- Jul 25, 2009
- Messages
- 129
Except that after the 2 suicide rounds, there was no collateral damage done. So no unit fought twice, with the one exception of the unit the HA slammed into, and withdrew from, and we got 90% odds on that unit when it came back into the stack. And I wasn't trying to draw a line of "here's where accusations would start", that is impossible. What I did say was "we should do this good or better in about 1 in 90000 times", and, frankly, yes, that doesn't look good. If you can find where I'm off by a factor of 10 or 100, then go ahead. If we'd lost a fight that we didn't, we'd have done worse, so the effects of not winning are irrelevant to the "this good or better" question.
For the "1%-21%" set:
SANCTA vs. Kaz (50BC)
axe v mace, 6 v 8.8, 10.7%, win
SANCTA vs. Kaz (1AD)
longbow v war elephant, 6 v 8.8, 13.5%, lose (36/100 on elephant)
axe v maceman, 5.5 v 8.8, 20.5%, lose
SANCTA vs. Kaz (940AD)
rifle v rifle, 15.4 v 22.4, 10.9%, win
horse archer v war elephant, 6 v 11.42, 5.0%, withdrawl (30% rate)
2 wins, 2 losses, 1 WD. Counting the WD as a loss (and none of the other units could have withdrawn anyways), and only looking at those 5 combats, we get 2 wins in those 5 combats almost 10% of the time. Sure, usually one only wins 1 or 0, but 10% odds is simply lucky.
Here's the damning bit: 8/8 for the 21-40% range: 0.066%, if they're all 40% odds.
7/7 for 40-60%: 2.80%, if they're all 60% odds.
7/7 for 60-80%: 21.0%, if they're all 80% odds.
Combined odds of getting 22/22 out of that set: 4 in a million. Sure, talk about "combat's interdependent": sorry, that doesn't fly. I'm looking at each individual combat, should we win or not? If we won fight A that we lost, and won fight B, sure, that would change the odds of future fights, but that doesn't change the odds of winning fight A. And we keep on winning fight A, even when the odds say otherwise.
Sure, I know what you were doing and I don't doubt your maths for the question you asked. However, "what are the chances of doing this good or better" is certainly easier to calculate, but not as meaningful a question as "what are the chances of doing so well that we get accused of cheating". You have to account for cases where we lose one, two or maybe more battles, and in those cases the non-independence of the events certainly is a factor. The "as good or better" question is not so meaningful when the (many) instances that the results are worse but still good enough to get accused of cheating do not differ in relevance (I'm talking in-game relevance here
) that much when compared to the actual result.Mojoqmeyvam
Warlord
- Joined
- Jul 25, 2009
- Messages
- 129
1 battle wouldn't have done anything to accusations. My off-the-cuff guess is that would put the odds down to the 1-in-3000 range (11 battles, average odds are like 30-40% or so, there's a factor of 2 or 3 error there, I'm also ignoring how hard it would be to win battles against damaged units, although 70-90% odds are fairly likely). 2 battles, however, and we'd be looking at "unlikely enough": 1-100 or so, maybe as much as 1-500. But it's not like this is a perfectly random system, freaky things can happen. And when my null hypothesis is 'nothing funny is going on', either with the RNG being temporarily streaky, or god knows what, and the game says that I'd see my current results in about 1 in 90,000 instances, I'm rejecting the null.
Maybe the RNG was really streaky for 50BC and 940AD, which would have used strings of seeds, and we've won the other battles through a fair amount of luck. 1AD looks perfectly normal: we got a 20% withdraw, and won one of three 15-35% battles, and won the 73% battle.
The other individual combats, which we swept (counting the 50% withdraw as a win), were 1-155 to happen that way, although that's with the HA doing 0 damage, if we'd gotten any hits in, then those odds would jump to as high as 1-50 or 1-100. Lucky, but not implausibly so. p happens.
Disproving this last hypothesis, that the RNG was streaky in 50BC and 940AD, and we've gotten very favorable, but not absurdly so, rolls in the other years, is much harder.
Maybe the RNG was really streaky for 50BC and 940AD, which would have used strings of seeds, and we've won the other battles through a fair amount of luck. 1AD looks perfectly normal: we got a 20% withdraw, and won one of three 15-35% battles, and won the 73% battle.
The other individual combats, which we swept (counting the 50% withdraw as a win), were 1-155 to happen that way, although that's with the HA doing 0 damage, if we'd gotten any hits in, then those odds would jump to as high as 1-50 or 1-100. Lucky, but not implausibly so. p happens.
Disproving this last hypothesis, that the RNG was streaky in 50BC and 940AD, and we've gotten very favorable, but not absurdly so, rolls in the other years, is much harder.
Again, disclaimer about not wanting to pre-judge what happened.
But if you took this attitude, you'd arrest every lottery winner for fraud. There are a lot of games of civilization reported on the internet.
Anyway, equating the probability of getting the observed results or better with the probability of the innocence of Memphus is a prosecutor's fallacy. See Meadow's Law, which became quite famously wrong in this country because it resulted in the wrongful conviction of a mother accused of murdering her two daughters. It's only correct to equate those two factors if, without the "evidence" of the combat log odds, the probability of memphus being innocent is only 50%. And seeing as there is no other evidence, it cannot be anywhere close to 50%.
But if you took this attitude, you'd arrest every lottery winner for fraud. There are a lot of games of civilization reported on the internet.
Anyway, equating the probability of getting the observed results or better with the probability of the innocence of Memphus is a prosecutor's fallacy. See Meadow's Law, which became quite famously wrong in this country because it resulted in the wrongful conviction of a mother accused of murdering her two daughters. It's only correct to equate those two factors if, without the "evidence" of the combat log odds, the probability of memphus being innocent is only 50%. And seeing as there is no other evidence, it cannot be anywhere close to 50%.
Let's illustrate this with an example of what people here are doing:
A lottery winner is in the dock. Sullla relays the following odds, given to him by team Cav and confirmed to be true by sooooo:
1st ball: Chance of getting this ball is 6/49
2nd ball: Chance is 5/48
3rd ball: Chance is 4/47
4th ball: Chance is 3/46
5th ball: Chance is 2/45
6th ball: Chance is 1/44
Mojoqmeyvam then calculates the p-value, the chance of achieving these results or better. He decides that probability is 1 in 14 million. He states: "1 in 14 million. Sure, talk about "balls are interdependent": sorry, that doesn't fly. I'm looking at each individual ball, should we win or not?"
Sullla observes that getting the first ball was pretty unlikely. The second ball also very lucky - he's getting suspicious. And then he gets the third ball! And ... oh my god ... he's got the fourth ball too! And so on until the 5th and 6th ball all match the lottery result's ticket. Sullla says "No one can possibly be this lucky, so consistently, over such a long period of time."
The lottery winner is then wrongly convicted of fraud. The jury have not taken into account how many people play civ, sorry, I mean the lottery. The argument is "But we're looking a this specific lottery ticket!"
Now if other evidence were to emerge later, such as the accused also being discovered to have manufactured the lottery balls, then that would be different.
A lottery winner is in the dock. Sullla relays the following odds, given to him by team Cav and confirmed to be true by sooooo:
1st ball: Chance of getting this ball is 6/49
2nd ball: Chance is 5/48
3rd ball: Chance is 4/47
4th ball: Chance is 3/46
5th ball: Chance is 2/45
6th ball: Chance is 1/44
Mojoqmeyvam then calculates the p-value, the chance of achieving these results or better. He decides that probability is 1 in 14 million. He states: "1 in 14 million. Sure, talk about "balls are interdependent": sorry, that doesn't fly. I'm looking at each individual ball, should we win or not?"
Sullla observes that getting the first ball was pretty unlikely. The second ball also very lucky - he's getting suspicious. And then he gets the third ball! And ... oh my god ... he's got the fourth ball too! And so on until the 5th and 6th ball all match the lottery result's ticket. Sullla says "No one can possibly be this lucky, so consistently, over such a long period of time."
The lottery winner is then wrongly convicted of fraud. The jury have not taken into account how many people play civ, sorry, I mean the lottery. The argument is "But we're looking a this specific lottery ticket!"
Now if other evidence were to emerge later, such as the accused also being discovered to have manufactured the lottery balls, then that would be different.
This would be my question. If anyone can shed light on this then it would help a lot.
Sullla--do you have any idea how one could do this?
Similar threads
