See the top rated post in this thread. Click here

Page 1 of 2 12 LastLast
Results 1 to 13 of 26

Thread: Blackjack House Edge Tests

  1. 10-29-2016, 12:56 PM #1
    CoopOwnz
    CoopOwnz is offline
    Member
    Join Date
    Oct 2016
    Posts
    42

    Did you find this post helpful? Yes | No

    Blackjack House Edge Tests

    Hello, I'm running some tests to calculate house edges with various rule variations.
    I just built my own simulation test program and I'm getting results that seem odd.
    Example test:
    I'm running 100,000 shoes of 6 decks with 75% penetration. This usually comes out to just under 4.5 million hands played.
    Splitting up to 4 hands. Double after split allowed. Late surrender allowed. Dealer Stays on S17. Can double on any 2 cards. Draws 1 card only after splitting aces. Blackjack pays 3:2.
    I flat bet $5 each hand and after running this sim about 10 times I'm getting approximately 0.675% house edge.
    This seems a little high from a lot of the reading I've been doing.
    The player is following the guidelines I found in this link:
    http://wizardofodds.com/games/blackj...ategy/4-decks/
    using the Dealer stands on soft 17 chart.

    Any suggestions on why this might be coming out high? Also, almost every website or book I've looked at shows different house edge percents and a lot show different basic strategy charts.
    Thanks.
    Reply With Quote Reply With Quote
  2. 10-29-2016, 12:57 PM #2
    ZenMaster_Flash
    ZenMaster_Flash is offline
    Banned or Suspended
    Join Date
    Dec 2011
    Location
    Eastern U S A
    Posts
    6,830

    Did you find this post helpful? Yes | No


    Why re-invent the wheel ?
    Reply With Quote Reply With Quote
  3. 10-29-2016, 01:05 PM #3
    CoopOwnz
    CoopOwnz is offline
    Member
    Join Date
    Oct 2016
    Posts
    42

    Did you find this post helpful? Yes | No
    I wanted a way to practice programming and learn more about blackjack along the way.
    Reply With Quote Reply With Quote
  4. 10-29-2016, 01:12 PM #4
    Exoter175
    Exoter175 is offline
    Banned or Suspended
    Join Date
    May 2014
    Posts
    1,289

    Did you find this post helpful? Yes | No
    Quote Originally Posted by CoopOwnz View Post
    Hello, I'm running some tests to calculate house edges with various rule variations.
    I just built my own simulation test program and I'm getting results that seem odd.
    Example test:
    I'm running 100,000 shoes of 6 decks with 75% penetration. This usually comes out to just under 4.5 million hands played.
    Splitting up to 4 hands. Double after split allowed. Late surrender allowed. Dealer Stays on S17. Can double on any 2 cards. Draws 1 card only after splitting aces. Blackjack pays 3:2.
    I flat bet $5 each hand and after running this sim about 10 times I'm getting approximately 0.675% house edge.
    This seems a little high from a lot of the reading I've been doing.
    The player is following the guidelines I found in this link:
    http://wizardofodds.com/games/blackj...ategy/4-decks/
    using the Dealer stands on soft 17 chart.

    Any suggestions on why this might be coming out high? Also, almost every website or book I've looked at shows different house edge percents and a lot show different basic strategy charts.
    Thanks.
    It would appear as if you're running roughly 2x the House Edge as you're supposed to be running, your results would be close to that of an H17 game where you could only double on any 9-11. Two things I'd do before anything else. First, run the sim again to see if you get similar results. If the results are similar, (within about 10%) I'd check your coding for a problem.
    Reply With Quote Reply With Quote
  5. 10-29-2016, 01:37 PM #5
    ZenMaster_Flash
    ZenMaster_Flash is offline
    Banned or Suspended
    Join Date
    Dec 2011
    Location
    Eastern U S A
    Posts
    6,830

    Did you find this post helpful? Yes | No
    Quote Originally Posted by moses View Post
    "because casino's can take away your car?"
    w. t. f. ?
    Reply With Quote Reply With Quote
  6. 10-29-2016, 01:39 PM #6
    CoopOwnz
    CoopOwnz is offline
    Member
    Join Date
    Oct 2016
    Posts
    42

    Did you find this post helpful? Yes | No
    I just ran another test of about 9 million hands. Here are the results:
    Total Hands: 8,994,036
    Time to complete: 00:00:11.9001209
    Total Bet: $48,660,343
    Total Won/Lost: $-339,415
    Approx. House Edge: 0.6975187 %
    Win percentage: 42.63364 %
    Lose percentage: 44.7137 %
    Standoff percentage: 8.307038 %
    Surrender percentage: 4.345624 %

    I think the other percentages seem to be pretty accurate.
    I calculate house edge using:
    ((Total Won/Lost) / (Total Bet)) * 100
    Total Bet is adding up all the bets including splits and doubles. And for surrenders it is only adding half the bet. Does that seem correct?
    Reply With Quote Reply With Quote
  7. 10-29-2016, 02:25 PM #7
    Exoter175
    Exoter175 is offline
    Banned or Suspended
    Join Date
    May 2014
    Posts
    1,289

    Did you find this post helpful? Yes | No
    Quote Originally Posted by CoopOwnz View Post
    I just ran another test of about 9 million hands. Here are the results:
    Total Hands: 8,994,036
    Time to complete: 00:00:11.9001209
    Total Bet: $48,660,343
    Total Won/Lost: $-339,415
    Approx. House Edge: 0.6975187 %
    Win percentage: 42.63364 %
    Lose percentage: 44.7137 %
    Standoff percentage: 8.307038 %
    Surrender percentage: 4.345624 %

    I think the other percentages seem to be pretty accurate.
    I calculate house edge using:
    ((Total Won/Lost) / (Total Bet)) * 100
    Total Bet is adding up all the bets including splits and doubles. And for surrenders it is only adding half the bet. Does that seem correct?
    I'm assuming you rounded one up, and if so your percentages are coming in at 100%, so its at least tracking correctly and recording correctly. The fact that the second result returned a higher HE% bothers me some. Can you post a snapshot of the coding for the various rule sets? Something has to be fuddled up in oversight, as the numbers aren't far off enough to suggest a real "fuck up" in the coding.
    Reply With Quote Reply With Quote
  8. 10-29-2016, 02:58 PM #8
    CoopOwnz
    CoopOwnz is offline
    Member
    Join Date
    Oct 2016
    Posts
    42

    Did you find this post helpful? Yes | No
    It's a lot of code so before I dig into that I want to make sure I'm doing the house edge calculation properly.
    So lets say I am betting $5 each hand.
    Hand 1: I'm dealt blackjack. (Bet Total = 5) (Total Won/Lost = 7.5)
    Hand 2: I split once & win both (Bet Total = 15) (Total Won/Lost = 17.5)
    Hand 3: I double down & lose (Bet Total = 25) (Total Won/Lost = 7.5)
    Hand 4: Dealer gets BJ (Bet Total = 30) (Total Won/Lost = 2.5)
    Hand 5: I surrender (Bet Total = 32.5) (Total Won/Lose = 0)

    That is essentially how my program is handling each situation.
    At the end of all the hands just doing ((Total Won/Lost) / (Total Bet)) * 100 for house edge. What do you think?
    Thanks for your thoughts so far.
    Reply With Quote Reply With Quote
  9. 10-29-2016, 05:51 PM #9
    CoopOwnz
    CoopOwnz is offline
    Member
    Join Date
    Oct 2016
    Posts
    42

    Did you find this post helpful? Yes | No
    So I found an error in my code. I had a - instead of a + in one spot...
    My new tests have been with over 22 million hands per run since I was still getting a lot of variation in the smaller runs.
    Same rules apply as I had above and getting about 0.1% house edge each time.
    Reply With Quote Reply With Quote
  10. 10-29-2016, 05:57 PM #10
    Norm
    Norm is offline
    Random number herder Norm's Avatar
    Join Date
    Dec 2011
    Location
    The mote in God's eye
    Posts
    12,473
    Blog Entries
    59

    Did you find this post helpful? Yes | No
    Not enough runs.
    "I don't think outside the box; I think of what I can do with the box." - Henri Matisse
    Reply With Quote Reply With Quote
  11. 10-29-2016, 07:57 PM #11
    DSchles
    DSchles is offline
    Senior Member
    Join Date
    Apr 2012
    Posts
    5,140

    Did you find this post helpful? Yes | No
    "Hand 5: I surrender (Bet Total = 32.5) (Total Won/Lose = 0)"

    This is wrong. Your bet is $5. You lose $2.50 when you surrender, but that can't change the size of the original bet. You bet $5, and that's how it should be counted. Also, by convention, house advantage is quoted as what we refer to as IBA, that is, as a percentage of the original, or initial, bet. You are quoting TBA, total bet advantage, so, even if done properly, your numbers will never match those normally quoted. Total bets are about 13.5% higher than original bets, so edges quoted using total bets will be the reciprocal or about 1/1.135, or 88% of the values obtained by using IBA.

    Finally, values you'll see for BS in the literature or online are almost always for off-the-top scenarios and not dealing down to a cut card, which introduces a cut-card effect that adds to the house edge. It is a very small value for the 6-deck game you're using (about 0.01%), but it would be much more significant in a SD game.

    Don
    Last edited by DSchles; 10-29-2016 at 08:14 PM.
    Reply With Quote Reply With Quote
  12. 10-30-2016, 05:55 AM #12
    Stealth
    Stealth is offline
    Senior Member Stealth's Avatar
    Join Date
    Mar 2012
    Location
    Las Vegas
    Posts
    2,008

    Did you find this post helpful? Yes | No
    Quote Originally Posted by CoopOwnz View Post
    My new tests have been with over 22 million hands per run since I was still getting a lot of variation in the smaller runs.
    You might think about the need for 400-500 million rounds to get accurate information. Refer to CVDATA documentation for suggestions.
    Luck is nothing more than probability taken personally!
    Reply With Quote Reply With Quote
  13. 10-30-2016, 02:47 PM #13
    ZenMaster_Flash
    ZenMaster_Flash is offline
    Banned or Suspended
    Join Date
    Dec 2011
    Location
    Eastern U S A
    Posts
    6,830

    Did you find this post helpful? Yes | No

    Actually, I am thinking that there is a coding error as such an

    egregious difference in H.E. cannot be that far off due only to

    insufficient sample size. Besides, if that was the case, wouldn't

    we be seeing an outsized S.E    . ?
    Reply With Quote Reply With Quote
Page 1 of 2 12 LastLast
« Previous Thread | Next Thread »

Similar Threads

  1. Why does casino assign high limits to blackjack games with lower house edge?
    By seriousplayer in forum General Blackjack Forum
    Replies: 32
    Last Post: 12-04-2015, 09:32 AM
  2. Does Blackjack Switch have a lower House Edge than regular Blackjack?
    By Inv in forum General Blackjack Forum
    Replies: 5
    Last Post: 12-24-2014, 12:00 PM
  3. Blackjack switch house edge question
    By MrPrime in forum General Blackjack Forum
    Replies: 10
    Last Post: 06-04-2013, 03:12 PM
  4. illsur5: Spanish 21 house edge less than blackjack?
    By illsur5 in forum Blackjack Main
    Replies: 4
    Last Post: 01-15-2003, 06:40 PM

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  

Forum Rules

Other QFIT Sites

Internal Links

About Blackjack: The Forum

BJTF is an advantage player site based on the principles of comity. That is, civil and considerate behavior for the mutual benefit of all involved. The goal of advantage play is the legal extraction of funds from gaming establishments by gaining a mathematic advantage and developing the skills required to use that advantage. To maximize our success, it is important to understand that we are all on the same side. Personal conflicts simply get in the way of our goals.