LoquaciousMoFW
LoquaciousMoFW
Joined: Aug 24, 2014
  • Threads: 0
  • Posts: 120
Thanks for this post from:
RealizeGaming
March 19th, 2020 at 11:37:11 AM permalink
I agree with the increased wager amount; I'd actually play this in a casino for $30 a hand, and I usually max out at $5 a deal. Of course that is probably because the RTP is currently high :) Got up to $24,000 in that session.

The dealt royal was nice, but the resulting 9 card royal interested me more.
RealizeGaming
RealizeGaming
Joined: Aug 1, 2013
  • Threads: 55
  • Posts: 564
March 19th, 2020 at 11:50:50 AM permalink
Quote: LoquaciousMoFW

I agree with the increased wager amount; I'd actually play this in a casino for $30 a hand, and I usually max out at $5 a deal. Of course that is probably because the RTP is currently high :) Got up to $24,000 in that session.

The dealt royal was nice, but the resulting 9 card royal interested me more.



I'd be interested in how your others sessions go without the royal.

The nice thing about this game being a video poker game, a company could easily build in other bonuses such as 6, 7, 8 card royals or even multiple four of a kind bonuses, etc.
gordonm888
gordonm888
Joined: Feb 18, 2015
  • Threads: 37
  • Posts: 2396
March 19th, 2020 at 1:01:04 PM permalink
I too am finding this game too be addictive.

Trade + 2 places mode: Very fun, but certainly +EV for the player with current wager size
3 Places mode: Not fun but certainly is -EV for the player

I imagine that once your mathematician figures out the EV for the player in both modes, you will adjust the frequency of the Trade + 2 places mode: downward. But if that trade mode becomes too infrequent I would not play this game - the trade decision is what is making this game fun for me. So you might want to consider increasing the initial wager size or adjusting paytable downward

I have seen that I play the game faster as I become more accustomed to it.
So many better men, a few of them friends, were dead. And a thousand thousand slimy things lived on, and so did I.
RealizeGaming
RealizeGaming
Joined: Aug 1, 2013
  • Threads: 55
  • Posts: 564
March 19th, 2020 at 3:19:53 PM permalink
Quote: gordonm888

I too am finding this game too be addictive.

Trade + 2 places mode: Very fun, but certainly +EV for the player with current wager size
3 Places mode: Not fun but certainly is -EV for the player

I imagine that once your mathematician figures out the EV for the player in both modes, you will adjust the frequency of the Trade + 2 places mode: downward. But if that trade mode becomes too infrequent I would not play this game - the trade decision is what is making this game fun for me. So you might want to consider increasing the initial wager size or adjusting paytable downward

I have seen that I play the game faster as I become more accustomed to it.



Glad you are finding the game to be addictive!

Thanks for your thoughts on the number of trades and places. I'm sure as we get closer to the math we can find something that works. I'm sure we will look at reducing the appearance of the trade card to help balance the game, but I still want it to happen often. In this version of the game, we could go to a trade plus one place or without a trade card it could be 2 or 3 placed. Definitely something to look at. If people enjoy the current trades and place amounts, I'd first look at raising the bet as I really don't think people will mind since it is very rare that you go winless.

Have you had a chance to try the 3x3 version of the game in the other thread? Just curious to hear your thoughts on it. I actually like both, but you can expect that since they are my own games.
gordonm888
gordonm888
Joined: Feb 18, 2015
  • Threads: 37
  • Posts: 2396
Thanks for this post from:
RealizeGaming
March 20th, 2020 at 10:25:58 PM permalink
You have reduced the frequency of the "Trade+ 2 Places" in the demo game! AArgh! Its still fun , but its also still +EV. The issue is the frequency with which player can make full houses and 4oaks.

I am a gaming mathematician, and I have tried to dream up ways to analyze this game. There are 17 or 18 cards involved in the game, so looping through all the possibilities is not possible. Also, given 15 cards on the board, a Trade decision involves 75 different options and a place decision involves 15 different options.

So the total number of combinations to be evaluated to create optimum player decisions in a brute force looping code is:

Place 3
c(52,5) *c(47,5)* C(42,5) * 37 * 15 *36 * 15 * 35 * 15 = 7.6 x 10^25

Trade + Place 2
c(52,5) *c(47,5)* C(42,5) *75 * 37 * 15 *36 * 15 = 5.3 x 10^26

So, I believe that is too many to do by a brute force looping code.

One might try generating the 15 card board via monte carlo with an RNG, and then looping through all the possible decisions to generate optimum player strategy for each randomly-selected 15 card board. That would still be a massive analysis.

OR, you might allow players to play a demo game that you post online and accumulate information on the outcomes of the Trade mode and the Place 3 mode.

Can you share with us the approach for analyzing this game?
So many better men, a few of them friends, were dead. And a thousand thousand slimy things lived on, and so did I.
GaryJKoehler
GaryJKoehler
Joined: Oct 22, 2015
  • Threads: 3
  • Posts: 139
March 21st, 2020 at 5:34:16 AM permalink
Quote: gordonm888


One might try generating the 15 card board via monte carlo with an RNG, and then looping through all the possible decisions to generate optimum player strategy for each randomly-selected 15 card board. That would still be a massive analysis.


This seems like the only realistic approach to getting an estimate of the EV.
RealizeGaming
RealizeGaming
Joined: Aug 1, 2013
  • Threads: 55
  • Posts: 564
Thanks for this post from:
charliepatrick
March 21st, 2020 at 8:41:13 AM permalink
Trade N' Place - developmental version 3x3 version
-use this link to check the logic of the auto placer
http://realizegamingllc.com/dev/tradeNPlace3x3/

Trade N' Place - original 3x3 version
-use this link for trade + 1 place or just 1 place
https://realizegamingllc.com/demo/tradeNPlace3x3/

Trade N' Place - original 5x3 version
-use this link for trade + 2 place or just 3 places
https://www.realizegamingllc.com/demo/tradeNPlace/
RealizeGaming
RealizeGaming
Joined: Aug 1, 2013
  • Threads: 55
  • Posts: 564
March 21st, 2020 at 8:44:26 AM permalink
Quote: gordonm888

You have reduced the frequency of the "Trade+ 2 Places" in the demo game! AArgh! Its still fun , but its also still +EV. The issue is the frequency with which player can make full houses and 4oaks.

I am a gaming mathematician, and I have tried to dream up ways to analyze this game. There are 17 or 18 cards involved in the game, so looping through all the possibilities is not possible. Also, given 15 cards on the board, a Trade decision involves 75 different options and a place decision involves 15 different options.

So the total number of combinations to be evaluated to create optimum player decisions in a brute force looping code is:

Place 3
c(52,5) *c(47,5)* C(42,5) * 37 * 15 *36 * 15 * 35 * 15 = 7.6 x 10^25

Trade + Place 2
c(52,5) *c(47,5)* C(42,5) *75 * 37 * 15 *36 * 15 = 5.3 x 10^26

So, I believe that is too many to do by a brute force looping code.

One might try generating the 15 card board via monte carlo with an RNG, and then looping through all the possible decisions to generate optimum player strategy for each randomly-selected 15 card board. That would still be a massive analysis.

OR, you might allow players to play a demo game that you post online and accumulate information on the outcomes of the Trade mode and the Place 3 mode.

Can you share with us the approach for analyzing this game?



gordonm888,

I appreciate all your incite on the math for Trade N' Place. I will have my math and demo guys read your post over to see if they can provide more information. In my opinion, both your ideas are very valid.
charliepatrick
charliepatrick
Joined: Jun 17, 2011
  • Threads: 30
  • Posts: 1987
Thanks for this post from:
RealizeGaming
March 21st, 2020 at 12:43:22 PM permalink
Quote: RealizeGaming

.....Trade N' Place - original 5x3 version
-use this link for trade + 2 place or just 3 places
https://www.realizegamingllc.com/demo/tradeNPlace/

Nice game especially when this happens!
VladAlex1
VladAlex1
Joined: Dec 4, 2015
  • Threads: 12
  • Posts: 118
March 21st, 2020 at 12:54:11 PM permalink
By Alex Dunaevsky.


We haven’t started on the math analysis for this game since we’re focusing on the 3x3 version.

Our approach is to write a script that determines the best moves given specific cards on the table. We’ll test this script in the game by asking for player’s review and perhaps with an option to select a specific dealt set of cards to analyze tricky situations. Once we’re happy with the performance of the auto-play mode, we’ll take the script and run it outside the game in a terminal with no user interface. It should be able to play a billion rounds of the game in about 24 hours. We’ll only run a billion rounds once we see that it’s giving us a good RTP in shorter 5 to 20 min. simulations.

In the trade phase, the script might follow some steps like this:

1) Evaluate the 3 hands before the trade. Rank them form best to worst. Save that result for later to see if it’s better or worse after a potential trade move. Perhaps “No trade” is the best move. When ranking the hands, it might be necessary to apply the classic poker strategy and see which cards make sense to hold. Although there is no “hold” option in this game, that strategy can be used to give better rankings to the 3 hands.

2) Record the hand results of the 75 possible trade moves. Record their rankings using the same classic poker strategy.

3) Compare the possible trade moves with the original 3 hands and find the best move.

When making a swap move, the best move seems to depend on if it’s the last swap, or there are 1 or 2 more swaps to go.

For the last swap of the round, it’s a basic check of all 15 moves to see which one pays out the most.

For the first or second last swap, it would require further ranking of the hands before and after the swap to find the best 3 hands.

It’s a lengthy process to get the final script working, and we haven’t even started on the 5x3 version. It’s amazing how fast the computer can play the game, and even if it takes 3 days to simulate a billion rounds, it’s not a problem to us.
I’d rather have to be a lucky player than good one.

  • Jump to: