As someone was running a two of in a Reno deck yesterday in the SOU tourney, I was interested in the probability of the "Reno" - card being drawn, but none of the two duplicate cards. Which means you could play the "Reno" - card but the effect would not be active.
I have attached a plot of this probability. You can see that by about 10 cards drawn this probability reaches its max, which is about 15.5%. So roughly every 6-7. game you would have drawn your "Reno" - card within the first 10 cards, but the effect would not be active. (By this time you should already have enough mana to play the card aswell) Also from the 4th card on the probability is already above 10%, which is releveant if you want to play "Zephrys the Great" early on.
I have done a simulation of 10 000 games in R by random sampling from 1 to 30, which leads to the same results as in my plot. So my math should be correct.
Therefore I would conclude that it is not worth running any card twice in a Reno deck.
Trump was running two of the 4-mana 3/3 Hyena that spawns 2x 2/2 minions if you have a secret, in a Highlander Hunter deck the other day - it's done quite often as long as you are careful what duplicates you run and are willing to take the chance that both are not buried in the bottom of the deck.
This was tested ad infinitum last time "no duplicate" cards were in standard. It was a terrible idea then and it is a terrible idea now. It doesn't matter how good Hyena Alpha is, having a slightly weaker card in place of the second copy is going to cost you way less games than having your two most powerful bombs be inactive at a key moment in ~15% (1 in 6!!) of your games.
Only two cards in your whole deck rely on the singleton requirement for the deck - so you have a fairly equal chance of both of your bombs sitting on the bottom of your deck and you never see them either. That doesn't mean that there is no risk involved - there always will be; but I think it is probably less of a problem than expected. But then, it will be more down to actual testing with the deck and seeing how it feels when this sort of risk is included - in many times, perhaps the outcome would not have been much different. We will have to see when the decks have had some playtesting and different iterations.
Back in the day I used to run double Mountain Giant and double Twilight Drake in Renolock. Really wasn't an issue aside from a small handful of games. Most of the time it worked out just fine.
I dont get it why the probability is lower before turn 10. Shouldnt the prob dicrease as the number of turn goes (which mean you have less and less cards to draw ?).
I dont get it why the probability is lower before turn 10. Shouldnt the prob dicrease as the number of turn goes (which mean you have less and less cards to draw ?).
Thanks anyway I also like this kind of info
At first the prob increases because your chance of drawing the Reno card increases with more cards drawn. But at one point it gets more and more unlikely that you have drawn your Reno card, but not any of the two duplicates.
Also depends on how key your "bomb" cards are to your win condition. If they're essential, then it makes no sense to run duplicates, but if Zephrys is in there just as an end game powerplay, then running some duplicates may be worth it.
3. I see it is enough nerdy thread to share my Renounce Darkness higlander cards synergy calculation(calculation done for wild Rise of the shadows card pool):
This was tested ad infinitum last time "no duplicate" cards were in standard. It was a terrible idea then and it is a terrible idea now. It doesn't matter how good Hyena Alpha is, having a slightly weaker card in place of the second copy is going to cost you way less games than having your two most powerful bombs be inactive at a key moment in ~15% (1 in 6!!) of your games.
I never run duplicates and no good players do either. 95% of the Reno lists contain no duplicates, those that do are plain bad and you just need to cut the duplicate for a card that is 95% as good in order not to lose 15% of games when Reno whiffs.
This was tested ad infinitum last time "no duplicate" cards were in standard. It was a terrible idea then and it is a terrible idea now. It doesn't matter how good Hyena Alpha is, having a slightly weaker card in place of the second copy is going to cost you way less games than having your two most powerful bombs be inactive at a key moment in ~15% (1 in 6!!) of your games.
Surely it depends to some degree on how quickly you are drawing through your deck.
But in the case of Hunter, that's not very quickly, so it probably is bad. On the other hand, there's no such thing as "slightly weaker" than Hyena Alpha, one of the strongest minions in an extremely strong set.
Rollback Post to RevisionRollBack
"Why, you never expected justice from a company, did you? They have neither a soul to lose nor a body to kick." -- Lady Saba Holland
This was tested ad infinitum last time "no duplicate" cards were in standard. It was a terrible idea then and it is a terrible idea now. It doesn't matter how good Hyena Alpha is, having a slightly weaker card in place of the second copy is going to cost you way less games than having your two most powerful bombs be inactive at a key moment in ~15% (1 in 6!!) of your games.
I never run duplicates and no good players do either. 95% of the Reno lists contain no duplicates, those that do are plain bad and you just need to cut the duplicate for a card that is 95% as good in order not to lose 15% of games when Reno whiffs.
Trump does, and I'd happily argue that he can be considered a "good player". He's certainly in the "Pro" category.
As someone was running a two of in a Reno deck yesterday in the SOU tourney, I was interested in the probability of the "Reno" - card being drawn, but none of the two duplicate cards. Which means you could play the "Reno" - card but the effect would not be active.
I have attached a plot of this probability. You can see that by about 10 cards drawn this probability reaches its max, which is about 15.5%. So roughly every 6-7. game you would have drawn your "Reno" - card within the first 10 cards, but the effect would not be active. (By this time you should already have enough mana to play the card aswell)
Also from the 4th card on the probability is already above 10%, which is releveant if you want to play "Zephrys the Great" early on.
I have done a simulation of 10 000 games in R by random sampling from 1 to 30, which leads to the same results as in my plot. So my math should be correct.
Therefore I would conclude that it is not worth running any card twice in a Reno deck.
I have played some duplicates in my Reno decks over the years and that worked just fine.
Double Dead Man's Hand in Reno warrior for example.
Trump was running two of the 4-mana 3/3 Hyena that spawns 2x 2/2 minions if you have a secret, in a Highlander Hunter deck the other day - it's done quite often as long as you are careful what duplicates you run and are willing to take the chance that both are not buried in the bottom of the deck.
This is so cool! I'm a real nerd for such statistics! Already thought about doing something like that.
Thanks a lot! For my part, this really helps with my decision!
This was tested ad infinitum last time "no duplicate" cards were in standard. It was a terrible idea then and it is a terrible idea now. It doesn't matter how good Hyena Alpha is, having a slightly weaker card in place of the second copy is going to cost you way less games than having your two most powerful bombs be inactive at a key moment in ~15% (1 in 6!!) of your games.
Only two cards in your whole deck rely on the singleton requirement for the deck - so you have a fairly equal chance of both of your bombs sitting on the bottom of your deck and you never see them either. That doesn't mean that there is no risk involved - there always will be; but I think it is probably less of a problem than expected.
But then, it will be more down to actual testing with the deck and seeing how it feels when this sort of risk is included - in many times, perhaps the outcome would not have been much different. We will have to see when the decks have had some playtesting and different iterations.
Can u share a source code? I like to mess around in R. I like the graph as well ;)
Back in the day I used to run double Mountain Giant and double Twilight Drake in Renolock. Really wasn't an issue aside from a small handful of games. Most of the time it worked out just fine.
Missing lethal since June 2015.
Sure:
## plot ##
x=c()
for (n in seq(1:28)){
x=c(x,choose(27,n-1)/choose(30,n))
}
plot(seq(1:28),x,xlab='number of cards drawn',ylab='prob',main='Prob. of Reno drawn but effect not active')
lines(seq(1:28),x,col='red')
##########
## simulation ##
d=15 ### Number of cards drawn
y=c()
x=c()
for (n in seq(1:10000)){
x=sample(30,d,replace=FALSE)
if ((1 %in% x) == TRUE){
if ((29 %in% x) == FALSE){
if ((30 %in% x) == FALSE){
y=c(y,1)
} else{
y=c(y,0)
}
} else{
y=c(y,0)
}
} else{
y=c(y,0)
}
}
z=0
for (n in y){
if (n==1){
z=z+1
}
}
try=c(1-z/length(y),z/length(y)) ## prob of simulation
prob=c(1-choose(27,d-1)/choose(30,d),choose(27,d-1)/choose(30,d)) ## computed prob
try
prob
Threads being made on Hearthpwn based on actual testing, simulations and calculations?
The world is over as we know it, ladies and gentlemen.
Dad, husband, gamer, fueled by coffee.
Currently playing Dragon Galakrond Priest, Dragon Galakrond Warrior and Highlander Dragon Hunter.
You're all forgetting that Bomb Warrior is probably going to kill all Highlander decks.
True, but I’m still going to play highlander decks!
I dont get it why the probability is lower before turn 10. Shouldnt the prob dicrease as the number of turn goes (which mean you have less and less cards to draw ?).
Thanks anyway I also like this kind of info
At first the prob increases because your chance of drawing the Reno card increases with more cards drawn. But at one point it gets more and more unlikely that you have drawn your Reno card, but not any of the two duplicates.
Also depends on how key your "bomb" cards are to your win condition. If they're essential, then it makes no sense to run duplicates, but if Zephrys is in there just as an end game powerplay, then running some duplicates may be worth it.
"I knew it."
1. I run 2 ice blocks with 1Mad Scientist and 1Arcanologist without any other secrests and it works for me.
2. I run maximum duplicates in decks with Fel Reaver and Keening Banshee with some highlander cards
3. I see it is enough nerdy thread to share my Renounce Darkness higlander cards synergy calculation(calculation done for wild Rise of the shadows card pool):
I never run duplicates and no good players do either. 95% of the Reno lists contain no duplicates, those that do are plain bad and you just need to cut the duplicate for a card that is 95% as good in order not to lose 15% of games when Reno whiffs.
Surely it depends to some degree on how quickly you are drawing through your deck.
But in the case of Hunter, that's not very quickly, so it probably is bad. On the other hand, there's no such thing as "slightly weaker" than Hyena Alpha, one of the strongest minions in an extremely strong set.
"Why, you never expected justice from a company, did you? They have neither a soul to lose nor a body to kick." -- Lady Saba Holland
Trump does, and I'd happily argue that he can be considered a "good player". He's certainly in the "Pro" category.
https://www.youtube.com/watch?v=hHn6BnPI3dg&t=838s