Random Thoughts Thread - Murder House Game
- Thread starter GTO
- Start date
Exactly, along with the indignant womenfolk. You may "PLAY TO WIN THE GAME!" but we also recognized that you play games to have fun.
I do! And I really enjoyed the banter and the limited amount of general crabbiness involved in this game.
Completely agree. I would also like to say thanks to everyone who participated and made it fun. Props to all the people that continued visiting the thread even after losing. This just started from a small idea (RTT house) and it just took off from there. Additionally, I'd like to thank all the ghosts for continuing to play and reply to all my PMs.
Can't say I totally understand it, but can definitely look into it. We found another one which works like the NBA Draft lottery. It lets you assign a total number of balls (ghosts) to two players and then randomly pick based on probability (somebody with 1 ghost vs. somebody with 5 ghosts would have a 1 in 5 chance of winning the battle).
http://www.draftpicklottery.com/index.php
Maybe we just keep it to RTT and if anyone is interested from outside, we can let them in. We only had two who were rarely in RTT but played this time around (Farnsworth, CoKane). Not likely many outside people want to visit the thread when they see the words Random Thoughts Thread on the title, anyway. Oh, and someone needs to convince Carvers to play next time around. That needs to happen.
During lunch I played around with using a random number generator to create a weighted system of doing battle depending on the total number of ghosts available and the number of ghosts for each player.
Say 10 total players so 10 ghosts available
Player 1 has 5 ghosts + 1 for them self = 6
Player 2 has 2 ghosts + 1 for them self = 3
Player 1 would get a random number between 6 and 16
Player 2 would get a random number between 3 and 13
Whomever gets the highest number would win. As it is now a tie is a do over.
Player 1 has better odds of getting a higher number because he has a higher range of numbers to select from.
I just did a 50 run sample and Player 1 would have won 40 times
Subsequent 50 sample runs gave Player 1 victories anywhere from 31 - 41 out of 50
Switching the ghosts to player 1 with 7 and player 2 with 2
Player 1 won 48 times over a 50 run sample
Ran that test a second time and player 1 won 40 times.
Could probably figure out different ways to additional weights. Like ghosts won in battle weigh more than ghosts that were attached to you.
Could also add trivia ghosts won to a players range.
So if we took the original example with 5 trivia ghosts
Say 10 total players so 10 ghosts available
Player 1 has 5 ghosts + 1 for them self + 5 trivia ghosts = 11
Player 2 has 2 ghosts + 1 for them self = 3
Player 1 would get a random number between 11 and 26 (total range increased from 10 to 15 due to 5 trivia ghosts)
Player 2 would get a random number between 3 and 13 (standard range of 10 because no trivia ghosts)
Or
Player 1 has 5 ghosts + 1 for them self = 6
Player 2 has 2 ghosts + 1 for them self + 5 Trivia ghosts = 8
Player 1 would get a random number between 6 and 16 (standard range of 10 because no trivia ghosts)
Player 2 would get a random number between 8 and 23 (total range increased from 10 to 15 due to 5 trivia ghosts)
Player 2 won 40 times
Subsequent 50 run samples produced 39, 41, 34 wins for Player 2.
So depending on the number of ghosts a player has, their odds of winning a battle would increase but it isn't just a simple win strictly due to number of ghosts
Ok, I went beyond just lunch.
Say 10 total players so 10 ghosts available
Player 1 has 5 ghosts + 1 for them self = 6
Player 2 has 2 ghosts + 1 for them self = 3
Player 1 would get a random number between 6 and 16
Player 2 would get a random number between 3 and 13
Whomever gets the highest number would win. As it is now a tie is a do over.
Player 1 has better odds of getting a higher number because he has a higher range of numbers to select from.
I just did a 50 run sample and Player 1 would have won 40 times
Subsequent 50 sample runs gave Player 1 victories anywhere from 31 - 41 out of 50
Switching the ghosts to player 1 with 7 and player 2 with 2
Player 1 won 48 times over a 50 run sample
Ran that test a second time and player 1 won 40 times.
Could probably figure out different ways to additional weights. Like ghosts won in battle weigh more than ghosts that were attached to you.
Could also add trivia ghosts won to a players range.
So if we took the original example with 5 trivia ghosts
Say 10 total players so 10 ghosts available
Player 1 has 5 ghosts + 1 for them self + 5 trivia ghosts = 11
Player 2 has 2 ghosts + 1 for them self = 3
Player 1 would get a random number between 11 and 26 (total range increased from 10 to 15 due to 5 trivia ghosts)
Player 2 would get a random number between 3 and 13 (standard range of 10 because no trivia ghosts)
Or
Player 1 has 5 ghosts + 1 for them self = 6
Player 2 has 2 ghosts + 1 for them self + 5 Trivia ghosts = 8
Player 1 would get a random number between 6 and 16 (standard range of 10 because no trivia ghosts)
Player 2 would get a random number between 8 and 23 (total range increased from 10 to 15 due to 5 trivia ghosts)
Player 2 won 40 times
Subsequent 50 run samples produced 39, 41, 34 wins for Player 2.
So depending on the number of ghosts a player has, their odds of winning a battle would increase but it isn't just a simple win strictly due to number of ghosts
Ok, I went beyond just lunch.
Last edited:
Half way through my brain hurt from the math. Hopefully GTO is better at math than me. 8 for sure is.
If you have 6 ghosts to someone's 3, and you only win 31/50 times, that doesn't seem weighted enough to me. You have double the ghosts but only will win a little over half the time?
I like the idea of the 6 sided dice solution mentioned earlier. If someone has 1 ghost they are rolling one virtual die. If they have two ghosts they roll 2 virtual dice. So in that case it is a probability spread of 1-6 vs. 2-12. It is weighted since there is only one combination that will yield a roll of 2 with two dice and one that will yield a 12, 2 combinations that will yield a 3 or 11 but multiple combinations that will yield results of 4-10. It takes pretty crazy rolls for a player with 5 dice (all 1's) to get beat by a player with 1 (a 6) but it can happen.
Probably would, I never looked at that
In regards to your other question, Instead of just going a straight number of ghosts vs number of ghosts, your odds would be based on your number of ghosts in relation to the total ghosts available. Then add the trivia ghosts to that when they came available.
It was just something that popped into my head so I started playing with it.
Being as close as I am to the Timberwolves and seeing their history of having a lot of chances in draft lotteries and seeing them only get the first pick once in their history and repeatedly ending up below their pick if they picked by record rather than lottery, I'm not sure I'm a big fan of the NBA model.
A little dated since they did finally get a #1 pick out of the lottery last year but still...
http://bleacherreport.com/articles/...luckiest-teams-in-nba-lottery-history/page/10
Warning: BR
I don't like using your chances to win based on the overall number of ghosts available because is 1 v 1 competition. How many ghosts Velo has shouldn't effect my odds when I'm playing Bawbie.
The NBA lottery is designed to be far enough away from "fair" to dissuade NBA teams from tanking for the draft pick. I think we want to throw in an element of surprise and uncertainty in there without making it a crap shoot.
I'm not an actuary so I can't say for certain how all that odds of thing happen works but I don't think it really works that way. It is your number vs Bawbies in regards to the total available. I could add in a extra bit to take into account the difference in ghosts just like with trivia ghosts earned to weight the odds a bit more in favor.
Or I could go mtown
