seven
seven
  • Threads: 25
  • Posts: 248
Joined: Oct 1, 2013
August 4th, 2019 at 7:44:42 AM permalink
Hello

I need your help. we are working on a game (android & ios) that is similar to plinko. actually a simple and easy game to play. please check screenshot (I apologize for the bad art work) https://www.screencast.com/t/Ntho5blfm9G

we will offer a daily contest. each player get 30 balls to play. the first 5 players at the end of the contest with the most accumulated points will get a Prize. as you can see there are 10 pockets. the RNG is set to 1 in 10 that means the ball will hit in average once each pocket. with 30 balls plaid each pocket should be hit 3x in average. but each point number ( as they are double/mirrored ) will be hit 6x

what are the chances that in case thousands or more players will play the daily contest that there will be 2 players with the same point amount at the end of the contest? could it happen often very often or not worth to mention?
do I need to change the formula to avoid the chance that there will be 2 players with same point amount?
I hope my bad english did not confuse you and you could understand the description and my concern

thank you all as always
seven
BleedingChipsSlowly
BleedingChipsSlowly
  • Threads: 23
  • Posts: 1033
Joined: Jul 9, 2010
August 4th, 2019 at 8:19:19 AM permalink
Please check your game design again. You say your game is like Plinko. For Plinko, the chance of having the ball land in particular slots is not the same. It is more likely the ball will be in the center than at the sides. If you simulate the ball travel to make the chance the same for all slots it will not look like the results are random.
“You don’t bring a bone saw to a negotiation.” - Robert Jordan, former U.S. ambassador to Saudi Arabia
seven
seven
  • Threads: 25
  • Posts: 248
Joined: Oct 1, 2013
August 4th, 2019 at 8:35:41 AM permalink
Quote: BleedingChipsSlowly

Please check your game design again. You say your game is like Plinko. For Plinko, the chance of having the ball land in particular slots is not the same. It is more likely the ball will be in the center than at the sides. If you simulate the ball travel to make the chance the same for all slots it will not look like the results are random.



thank you for the answer. it is similar to plinko design but has nothing to do with the plinko game itself. the outcome is according to RNG 1 in 10 as described. it is actually coded that the result is known before and the ball will fall into the pocket the RNG will tell
ThatDonGuy
ThatDonGuy
  • Threads: 118
  • Posts: 6406
Joined: Jun 22, 2011
August 4th, 2019 at 10:27:32 AM permalink
My simulation shows:
If 5000 people play, there will be a tie for first about 1/14 of the time
If 10,000 play, there will be a tie about 1/13 of the time
seven
seven
  • Threads: 25
  • Posts: 248
Joined: Oct 1, 2013
August 4th, 2019 at 10:39:56 AM permalink
Quote: ThatDonGuy

My simulation shows:
If 5000 people play, there will be a tie for first about 1/14 of the time
If 10,000 play, there will be a tie about 1/13 of the time



hello and thanks for the answer and info about your simulation. that means if 50k or 100k people will play there would be more ties and to be frank I would like to see no tie because players will not be happy to split the cash prize.
any idea how the formula should look like to avoid ties instead the 1 in 10? maybe 2 in 10?

edit:
your result is for 5000 and one time a 30 ball game? because players will play 30 times or more during the daily contest. they are allowed to play as many games they like
Last edited by: seven on Aug 4, 2019
ThatDonGuy
ThatDonGuy
  • Threads: 118
  • Posts: 6406
Joined: Jun 22, 2011
August 4th, 2019 at 11:59:30 AM permalink
Quote: seven

any idea how the formula should look like to avoid ties instead the 1 in 10? maybe 2 in 10?

edit:
your result is for 5000 and one time a 30 ball game? because players will play 30 times or more during the daily contest. they are allowed to play as many games they like


You can't avoid ties, as it will always be possible for two players to get the highest score.

One way to make it harder to have ties is, make the zeroes that are next to the 500s worth 100 points. With 10,000 plays, ties will now happen 1 time in 60.

The more different values there are, the less likely ties become. Maybe make the values something like this:
5000, 1000, 500, 25, 5, 1, 10, 100, 250, 2500
seven
seven
  • Threads: 25
  • Posts: 248
Joined: Oct 1, 2013
August 4th, 2019 at 12:20:07 PM permalink
Quote: ThatDonGuy

You can't avoid ties, as it will always be possible for two players to get the highest score.

One way to make it harder to have ties is, make the zeroes that are next to the 500s worth 100 points. With 10,000 plays, ties will now happen 1 time in 60.

The more different values there are, the less likely ties become. Maybe make the values something like this:
5000, 1000, 500, 25, 5, 1, 10, 100, 250, 2500



thank you very much for the interesting answer. how about 50 instead of 100? will this be a big difference?
edit
how about this
5000, 1000, 500, 50, 1 1, 50, 500, 1000, 5000 would this be better to avoid many ties?

the 2nd idea sounds actually nice. how about the following
5000, 1000, 500, 50, 25, 25, 100, 250, 2500
itsmejeff
itsmejeff
  • Threads: 3
  • Posts: 69
Joined: Aug 6, 2012
August 6th, 2019 at 8:40:20 AM permalink
Quote: seven

thank you very much for the interesting answer. how about 50 instead of 100? will this be a big difference?
edit
how about this
5000, 1000, 500, 50, 1 1, 50, 500, 1000, 5000 would this be better to avoid many ties?

the 2nd idea sounds actually nice. how about the following
5000, 1000, 500, 50, 25, 25, 100, 250, 2500


You will reduce risk of first place ties by increasing possible outcomes. Your first one has a problem with not using the ones or tens places. you can get values from 0 to 30*5000, but only increments of 500.

When I run the game with 100,000 players using your original numbers and 30 balls per player, I get around 160 unique results. If I take out two zeroes and replace them with a 1 and a 50, I get closer to 7000 unique results.

Here is the code in php sandbox: http://sandbox.onlinephpfunctions.com/code/a39dbd98a65a43dad61ac0c6aef27131650cf635

You can change values as needed. You only get 3 seconds of compute, so you cant go nuts with it.
Last edited by: itsmejeff on Aug 6, 2019
unJon
unJon
  • Threads: 14
  • Posts: 4655
Joined: Jul 1, 2018
August 6th, 2019 at 12:18:48 PM permalink
This would make a cool math problem from Wiz. My gut says you want to pick numbers that have the largest Least Common Multiple.
The race is not always to the swift, nor the battle to the strong; but that is the way to bet.
seven
seven
  • Threads: 25
  • Posts: 248
Joined: Oct 1, 2013
August 7th, 2019 at 10:31:20 AM permalink
Quote: itsmejeff

You will reduce risk of first place ties by increasing possible outcomes. Your first one has a problem with not using the ones or tens places. you can get values from 0 to 30*5000, but only increments of 500.

When I run the game with 100,000 players using your original numbers and 30 balls per player, I get around 160 unique results. If I take out two zeroes and replace them with a 1 and a 50, I get closer to 7000 unique results.

Here is the code in php sandbox: http://sandbox.onlinephpfunctions.com/code/a39dbd98a65a43dad61ac0c6aef27131650cf635

You can change values as needed. You only get 3 seconds of compute, so you cant go nuts with it.



hey Jeff

thanks a lot for your input. to be frank I am a noob and therefore all my questions and I appreciate all the help I get here.

to be frank I don't understand when you are saying that with 100 000 players and 30 balls per player you get 160 unique results? the rest are not unique? what does unique mean in this sample?
do you mean that 160 results are no ties?
same question for the 7000 unique results
seven
seven
  • Threads: 25
  • Posts: 248
Joined: Oct 1, 2013
August 7th, 2019 at 10:32:57 AM permalink
Quote: unJon

This would make a cool math problem from Wiz. My gut says you want to pick numbers that have the largest Least Common Multiple.



thanks

would you be so kind and give an example in numbers for the 10 pockets?
unJon
unJon
  • Threads: 14
  • Posts: 4655
Joined: Jul 1, 2018
August 7th, 2019 at 12:07:06 PM permalink
Quote: seven

thanks

would you be so kind and give an example in numbers for the 10 pockets?



Sure. Pick prime numbers.

5003
2503
1009
503
251
101
53
23
11
1
The race is not always to the swift, nor the battle to the strong; but that is the way to bet.
seven
seven
  • Threads: 25
  • Posts: 248
Joined: Oct 1, 2013
August 7th, 2019 at 12:15:37 PM permalink
Quote: unJon

Sure. Pick prime numbers.

5003
2503
1009
503
251
101
53
23
11
1



thanks again, ok now I understand what you meant.

I like the mirrored pocket numbers
would this be still ok to use

5003 2503 1009 503 251 251 503 1009 2503 5003
unJon
unJon
  • Threads: 14
  • Posts: 4655
Joined: Jul 1, 2018
August 7th, 2019 at 1:58:41 PM permalink
Quote: seven

thanks again, ok now I understand what you meant.

I like the mirrored pocket numbers
would this be still ok to use

5003 2503 1009 503 251 251 503 1009 2503 5003



I think you get more bang for the buck with 10 distinct numbers than 5. What if you pick numbers that are close together like:

5003 2503 1009 503 251 257 509 1013 2521 5009
The race is not always to the swift, nor the battle to the strong; but that is the way to bet.
seven
seven
  • Threads: 25
  • Posts: 248
Joined: Oct 1, 2013
August 10th, 2019 at 2:38:56 AM permalink
Quote: unJon

I think you get more bang for the buck with 10 distinct numbers than 5. What if you pick numbers that are close together like:

5003 2503 1009 503 251 257 509 1013 2521 5009



thanks again. I understand your point but to be frank I am afraid that players will not like or understand whats behind. I like the mirrored one best.
I will go for the start with
5000 1000 500 50 0 0 50 500 1000 5000
and lets see in real what will happen, if needed I can always change and update

thank you all
itsmejeff
itsmejeff
  • Threads: 3
  • Posts: 69
Joined: Aug 6, 2012
August 10th, 2019 at 8:15:06 AM permalink
Quote: seven

hey Jeff

thanks a lot for your input. to be frank I am a noob and therefore all my questions and I appreciate all the help I get here.

to be frank I don't understand when you are saying that with 100 000 players and 30 balls per player you get 160 unique results? the rest are not unique? what does unique mean in this sample?
do you mean that 160 results are no ties?
same question for the 7000 unique results


Sorry, distinct results. Unique results are much smaller number (around 10). There are not that many combinations possible. The range of scores is very small because of the numbers you are using. When you have the same values on each side, you reduce the possible number of outcomes.

See this thing I made.
http://www.deep13.xyz/notplinkogame.php

Changing out a two zeroes for a 1 and a 50, for example, spreads out the scores much more.
seven
seven
  • Threads: 25
  • Posts: 248
Joined: Oct 1, 2013
August 10th, 2019 at 10:48:40 AM permalink
Quote: itsmejeff

Sorry, distinct results. Unique results are much smaller number (around 10). There are not that many combinations possible. The range of scores is very small because of the numbers you are using. When you have the same values on each side, you reduce the possible number of outcomes.

See this thing I made.
http://www.deep13.xyz/notplinkogame.php

Changing out a two zeroes for a 1 and a 50, for example, spreads out the scores much more.



Hey Jeff

super helpful the link with testing option, very much appreciated
7craps
7craps
  • Threads: 18
  • Posts: 1977
Joined: Jan 23, 2010
August 10th, 2019 at 10:54:57 AM permalink
Quote: seven

any idea how the formula should look like to avoid ties

if there is a tie, make a tie breaking round or if that would not be possible

something to break the tie

as in the most highest numbers hit during a session, or player selects the total number of points will be then make the difference the deciding factor, etc.

Ties should never ruin the fun of the actual game play when it comes to giving out prizes.
winsome johnny (not Win some johnny)
seven
seven
  • Threads: 25
  • Posts: 248
Joined: Oct 1, 2013
August 11th, 2019 at 2:56:43 AM permalink
Quote: 7craps

if there is a tie, make a tie breaking round or if that would not be possible

something to break the tie

as in the most highest numbers hit during a session, or player selects the total number of points will be then make the difference the deciding factor, etc.

Ties should never ruin the fun of the actual game play when it comes to giving out prizes.



sadly I dont see a tie breaking round or something similar. this could interrupt the game flow imo
yes I agree with you that ties could ruin the fun of the game play and this was the reason that I came here and asked my question.

I did not reveal at the start that the game has a double up option after each 30 balls game play. the user can choose to use it or not. he will use it if he has a chance to get into the prize ranks. this might reduce the chance of a tie a bit more imo, not sure though.

I hope I am allowed to post my landing page (still under construction), and in case that not let the Moderator delete it please.
on the landing page you will understand more about the game and which solution I took for the start
https://frichinqo.com/test/#home

thank you all
beachbumbabs
beachbumbabs
  • Threads: 100
  • Posts: 14267
Joined: May 21, 2013
August 11th, 2019 at 8:56:11 AM permalink
Seven,

No problem with the link you posted.

Good luck with the game!
If the House lost every hand, they wouldn't deal the game.
seven
seven
  • Threads: 25
  • Posts: 248
Joined: Oct 1, 2013
August 11th, 2019 at 8:59:09 AM permalink
Quote: beachbumbabs

Seven,

No problem with the link you posted.

Good luck with the game!



thanks a lot for the message and wishes, very much appreciated

great help as always here in the forum
  • Jump to: