Results 1 to 12 of 12

Thread: Cacarulo: OPP vs KO vs Hi-Lo

Hybrid View

Previous Post Previous Post   Next Post Next Post
  1. #1
    Cacarulo
    Guest

    Cacarulo: OPP vs KO vs Hi-Lo

    Hi to all,

    The purpose of the following sims is to show accurately how much can be gained
    by using OPP/SC?. I've also simulated KO and Hi-Lo under the same conditions
    so you can compare OPP/SC? against them. I did the study using two different
    strategies:

    a) Generic BS (BS)
    b) Counter BS (CBS)

    Note: NO INDICES were used in any of the sims.

    At the end you'll find some extra sims in which I added more players (only BS)
    so you can see how they impact on each of the different systems.

    Game analyzed: 6D,S17,DOA,DAS,SPA1,SPL3,NS,5/6 pen,5000 million rounds per sim

    ##################################################  #############################


    1) KO + BS (heads up)

     play-all | rounds played = 100.00% 
    spread ev/h sd/h ror% n0 di score ekb avb unit kelly
    ------------------------------------------------------------------------------
    1 - 1 -0.428 11.542 100.00 0 0.00 0.00 10000.00 1.000 1.000 0.000
    1 - 4 0.364 22.719 13.53 389866 1.60 2.56 1418.55 1.607 7.049 1.000
    1 - 8 1.274 36.008 13.53 79822 3.54 12.53 1017.32 2.111 9.830 1.000
    1 - 12 2.079 48.173 13.53 53669 4.32 18.63 1116.01 2.539 8.961 1.000
    1 - 16 2.836 59.869 13.53 44553 4.74 22.45 1263.68 2.943 7.913 1.000
    1 - 20 3.547 70.905 13.53 39955 5.00 25.03 1417.27 3.320 7.056 1.000


    2) Hi-Lo + BS (heads up)

     play-all | rounds played = 100.00% 
    spread ev/h sd/h ror% n0 di score ekb avb unit kelly
    ------------------------------------------------------------------------------
    1 - 1 -0.428 11.542 100.00 0 0.00 0.00 10000.00 1.000 1.000 0.000
    1 - 4 0.467 25.994 13.53 309605 1.80 3.23 1446.36 1.839 6.914 1.000
    1 - 8 1.404 38.868 13.53 76679 3.61 13.04 1076.28 2.362 9.291 1.000
    1 - 12 2.251 51.662 13.53 52661 4.36 18.99 1185.53 2.852 8.435 1.000
    1 - 16 2.970 62.294 13.53 43994 4.77 22.73 1306.59 3.250 7.654 1.000
    1 - 20 3.731 74.177 13.53 39525 5.03 25.30 1474.69 3.697 6.781 1.000


    3) OPP + BS (heads up)

     play-all | rounds played = 100.00% 
    spread ev/h sd/h ror% n0 di score ekb avb unit kelly
    ------------------------------------------------------------------------------
    1 - 1 -0.428 11.542 100.00 0 0.00 0.00 10000.00 1.000 1.000 0.000
    1 - 4 0.287 25.270 13.53 776469 1.13 1.29 2226.76 1.806 4.491 1.000
    1 - 8 1.116 41.159 13.53 136006 2.71 7.35 1517.90 2.473 6.588 1.000
    1 - 12 1.857 55.726 13.53 90027 3.33 11.11 1672.01 3.057 5.981 1.000
    1 - 16 2.562 69.848 13.53 74312 3.67 13.46 1904.05 3.616 5.252 1.000
    1 - 20 3.195 82.349 13.53 66415 3.88 15.06 2122.22 4.103 4.712 1.000


    ##################################################  #############################


    1) KO + CBS (heads up)

     play-all | rounds played = 100.00% 
    spread ev/h sd/h ror% n0 di score ekb avb unit kelly
    ------------------------------------------------------------------------------
    1 - 1 -0.471 11.704 100.00 0 0.00 0.00 10000.00 1.000 1.000 0.000
    1 - 4 0.425 22.975 13.53 292503 1.85 3.42 1242.56 1.604 8.048 1.000
    1 - 8 1.452 36.359 13.53 62678 3.99 15.95 910.26 2.106 10.986 1.000
    1 - 12 2.364 48.700 13.53 42447 4.85 23.56 1003.35 2.535 9.967 1.000
    1 - 16 3.218 60.512 13.53 35350 5.32 28.29 1137.71 2.938 8.790 1.000
    1 - 20 4.025 71.720 13.53 31759 5.61 31.49 1278.12 3.317 7.824 1.000


    2) Hi-Lo + CBS (heads up)

     play-all | rounds played = 100.00% 
    spread ev/h sd/h ror% n0 di score ekb avb unit kelly
    ------------------------------------------------------------------------------
    1 - 1 -0.471 11.704 100.00 0 0.00 0.00 10000.00 1.000 1.000 0.000
    1 - 4 0.539 26.334 13.53 238386 2.05 4.19 1285.76 1.839 7.777 1.000
    1 - 8 1.592 39.289 13.53 60920 4.05 16.42 969.73 2.358 10.312 1.000
    1 - 12 2.549 52.304 13.53 42091 4.87 23.76 1073.07 2.853 9.319 1.000
    1 - 16 3.353 62.947 13.53 35242 5.33 28.38 1181.68 3.245 8.463 1.000
    1 - 20 4.212 75.003 13.53 31705 5.62 31.54 1335.49 3.695 7.488 1.000


    3) OPP + CBS (heads up)

     play-all | rounds played = 100.00% 
    spread ev/h sd/h ror% n0 di score ekb avb unit kelly
    ------------------------------------------------------------------------------
    1 - 1 -0.471 11.704 100.00 0 0.00 0.00 10000.00 1.000 1.000 0.000
    1 - 4 0.341 26.100 13.53 587135 1.31 1.70 1999.90 1.824 5.000 1.000
    1 - 8 1.279 42.180 13.53 108721 3.03 9.20 1390.79 2.500 7.190 1.000
    1 - 12 2.116 57.022 13.53 72627 3.71 13.77 1536.70 3.090 6.507 1.000
    1 - 16 2.884 70.763 13.53 60192 4.08 16.61 1736.11 3.626 5.760 1.000
    1 - 20 3.645 84.652 13.53 53940 4.31 18.54 1966.04 4.166 5.086 1.000


    ##################################################  #############################


    1) KO + BS (4 Players playing the same strategy - Seat #1)

     play-all | rounds played = 100.00% 
    spread ev/h sd/h ror% n0 di score ekb avb unit kelly
    ------------------------------------------------------------------------------
    1 - 1 -0.425 11.541 100.00 0 0.00 0.00 10000.00 1.000 1.000 0.000
    1 - 4 0.357 22.518 13.53 397984 1.59 2.51 1420.55 1.595 7.040 1.000
    1 - 8 1.261 35.862 13.53 80896 3.52 12.36 1020.00 2.099 9.804 1.000
    1 - 12 2.065 48.160 13.53 54369 4.29 18.39 1122.95 2.531 8.905 1.000
    1 - 16 2.804 59.553 13.53 45122 4.71 22.16 1265.01 2.921 7.905 1.000
    1 - 20 3.517 70.730 13.53 40448 4.97 24.72 1422.50 3.304 7.030 1.000


    2) Hi-Lo + BS (4 Players playing the same strategy - Seat #1)

     play-all | rounds played = 100.00% 
    spread ev/h sd/h ror% n0 di score ekb avb unit kelly
    ------------------------------------------------------------------------------
    1 - 1 -0.425 11.541 100.00 0 0.00 0.00 10000.00 1.000 1.000 0.000
    1 - 4 0.462 25.859 13.53 313279 1.79 3.19 1447.36 1.828 6.909 1.000
    1 - 8 1.399 38.908 13.53 77314 3.60 12.93 1081.84 2.357 9.244 1.000
    1 - 12 2.236 51.536 13.53 53120 4.34 18.83 1187.78 2.840 8.419 1.000
    1 - 16 2.955 62.263 13.53 44396 4.75 22.52 1311.91 3.240 7.622 1.000
    1 - 20 3.708 74.051 13.53 39889 5.01 25.07 1478.97 3.682 6.761 1.000


    3) OPP + BS (4 Players playing the same strategy - Seat #1)

     play-all | rounds played = 100.00% 
    spread ev/h sd/h ror% n0 di score ekb avb unit kelly
    ------------------------------------------------------------------------------
    1 - 1 -0.425 11.541 100.00 0 0.00 0.00 10000.00 1.000 1.000 0.000
    1 - 4 0.233 25.240 13.53 1170552 0.92 0.85 2730.78 1.775 3.662 1.000
    1 - 8 1.000 41.172 13.53 169435 2.43 5.90 1694.76 2.462 5.901 1.000
    1 - 12 1.684 55.645 13.53 109238 3.03 9.15 1839.12 3.033 5.437 1.000
    1 - 16 2.315 69.144 13.53 89196 3.35 11.21 2065.02 3.554 4.843 1.000
    1 - 20 2.941 82.792 13.53 79232 3.55 12.62 2330.45 4.079 4.291 1.000


    ##################################################  #############################


    From the above charts we can draw some interesting conclusions:

    A) For a 1-16 spread OPP/SC captures from the other systems the following percents:

    1) From KO    +  BS = 13.46/22.45*100 = 59.96% 
    2) From Hi-Lo + BS = 13.46/22.73*100 = 59.22%
    3) From KO + CBS = 16.61/28.29*100 = 58.71%
    4) From Hi-Lo + CBS = 16.61/28.38*100 = 58.53%


    B) Note that if we were to compare OPP/SC against KO or Hi-Lo with their corresponding
    C22 indices we would have a terrible performance under the same spread. To wit:

    1) KO    + C22 = $40.69 
    2) Hi-Lo + C22 = $40.43


    OPP + CBS would capture 16.61/40.69*100 = 40.82% from KO
    and 16.61/40.43*100 = 41.08% from Hi-Lo.

    C) Using more players makes OPP/SC's SCORE to deteriorate more than the other systems do.

    Hope this helps.

    Sincerely,
    Cacarulo

  2. #2
    MGP
    Guest

    MGP: Re: OPP vs KO vs Hi-Lo

    Thank you very much the sims Cacarulo.

    It's interesting that you're sims are actually giving better EV's than the OPP designers own sims.

    One question I have is for anyone as chicken as, and with as small as a bankroll as me, and if you have the time and don't mind - what is the EV for a 1 - 2 spread OPP (Heads up or 4 players, BS) and at what count would you begin doubling the bet?

    I tried using the MathProf numbers as well as the OPP designer's numbers and both gave a negative EV for that spread where the bet was doubled at the first +EV count. The sims they ran were for DD (MathProf and OPP) and 6D (OPP). Linear approximation of your numbers though suggest it is close to 0 and probably slightly negative (-0.1), but since it's not exactly linear it's not clear. Hi-Lo looks like it might actually be positive even with that low spread.

    Thanks in advance,
    MGP

  3. #3
    Cacarulo
    Guest

    Cacarulo: Re: OPP vs KO vs Hi-Lo

    > Thank you very much the sims Cacarulo.

    > It's interesting that you're sims are actually giving
    > better EV's than the OPP designers own sims.

    Probably because I'm using optimal betting ramps.

    > One question I have is for anyone as chicken as, and
    > with as small as a bankroll as me, and if you have the
    > time and don't mind - what is the EV for a 1 - 2
    > spread OPP (Heads up or 4 players, BS) and at what
    > count would you begin doubling the bet?

    With 1-2 it's still negative. You need at least 1-3 but the gain is really poor. You should be better off wonging instead of play-all. The same goes for Hi-Lo and KO.
    However, if you insist on playing all with a 1-3 spread the betting scheme goes as follows:

    Start at an IRC = 0
    If RC <= +4 bet 1 unit
    If RC > +4 bet 3 units

    > I tried using the MathProf numbers as well as the OPP
    > designer's numbers and both gave a negative EV for
    > that spread where the bet was doubled at the first +EV
    > count. The sims they ran were for DD (MathProf and
    > OPP) and 6D (OPP). Linear approximation of your
    > numbers though suggest it is close to 0 and probably
    > slightly negative (-0.1), but since it's not exactly
    > linear it's not clear. Hi-Lo looks like it might
    > actually be positive even with that low spread.

    > Thanks in advance,

    You're welcome.

    Sincerely,
    Cac

  4. #4
    T. Hopper
    Guest

    T. Hopper: Re: OPP vs KO vs Hi-Lo

    > Thank you very much the sims Cacarulo.

    > It's interesting that you're sims are actually giving
    > better EV's than the OPP designers own sims.

    > One question I have is for anyone as chicken as, and
    > with as small as a bankroll as me, and if you have the
    > time and don't mind - what is the EV for a 1 - 2
    > spread OPP (Heads up or 4 players, BS) and at what
    > count would you begin doubling the bet?

    Instead of trying to spread 1-2, check out the "Safest" spreads in T-H Basic Blackjack. They are optimized for the lowest possible risk of ruin assuming you play all hands.

  5. #5
    Norm Wattenberger
    Guest

    Norm Wattenberger: Interesting study

    Many thanks for the study. Some comments:

    1. Obviously your results for OPP will be superior to the OPP as originally stated since you maximized Win Rate instead of EV. After all, winning is the aim

    2. The fact that OPP deteriorates with more players at the table matches our beliefs and warns us that heads-up sims are misleading. Particularly since small players generally don?t play heads-up.

    3. Even though you added an optimized betting ramp and a Counter?s Basic Strategy to OPP; it still had 40.82% of the Win Rate of KO. With LSR the results are bound to have been worse yet. This is in agreement with MathProf?s sim.

    4. OPP with an optimized ramp and a CBS added won 40% less than crippled KO with no indexes and no Insurance. (Also known as KO Rookie.) This is worse than I estimated.

    I continue to believe that KO Rookie, which has no indexes, no Insurance and no TC conversion; but has about double (1.98 times) the Win Rate of optimized OPP is a better choice. And the player can add indexes in the future without changing the count and have a true professional level count.

  6. #6
    Cacarulo
    Guest

    Cacarulo: Re: Interesting study

    > Many thanks for the study. Some comments:

    > 1. Obviously your results for OPP will be superior to
    > the OPP as originally stated since you maximized Win
    > Rate instead of EV. After all, winning is the aim

    > 2. The fact that OPP deteriorates with more players at
    > the table matches our beliefs and warns us that
    > heads-up sims are misleading. Particularly since small
    > players generally don?t play heads-up.

    > 3. Even though you added an optimized betting ramp and
    > a Counter?s Basic Strategy to OPP; it still had 40.82%
    > of the Win Rate of KO. With LSR the results are bound
    > to have been worse yet. This is in agreement with
    > MathProf?s sim.

    > 4. OPP with an optimized ramp and a CBS added won 40%
    > less than crippled KO with no indexes and no
    > Insurance. (Also known as KO Rookie.) This is worse
    > than I estimated.

    I thought KO Rookie used Insurance.

    > I continue to believe that KO Rookie, which has no
    > indexes, no Insurance and no TC conversion; but has
    > about double (1.98 times) the Win Rate of optimized
    > OPP is a better choice. And the player can add indexes
    > in the future without changing the count and have a
    > true professional level count.

    And I agree.

    Thanks for the comments.

    Sincerely,
    Cac

  7. #7
    Norm Wattenberger
    Guest

    Norm Wattenberger: KO Insurance

    > I thought KO Rookie used Insurance.

    I believe KO Rookie is betting only. The book has a KO Preferred: Insurance Only as a stepping stone from Rookie to Preferred.

  8. #8
    Parker
    Guest

    Parker: KO Rookie

    > I believe KO Rookie is betting only. The book has a KO
    > Preferred: Insurance Only as a stepping stone from
    > Rookie to Preferred.

    I've always had a special fondness for KO Rookie, perhaps because it was the first system I learned.

    It has no insurance; indeed, it uses no indices at all. It doesn't really have a bet ramp, it simply uses a "big bet" and a much smaller waiting bet.

    I think a great system for the casual player would be a sort of "KO Semi-Rookie," which would be the Rookie system with insurance and indices for 16 vs 10 and 15 vs 10, along with a proper betting ramp.

    My KO book arrived in the mail the day I was leaving on a business trip to Las Vegas. I learned the Rookie count on the 1-hour flight from San Diego and used it for the first time in a casino that night.

    I don't necessarily recommend this. :-)

  9. #9
    T. Hopper
    Guest

    T. Hopper: Re: Interesting study

    > 3. Even though you added an optimized betting ramp and
    > a Counter?s Basic Strategy to OPP; it still had 40.82%
    > of the Win Rate of KO.

    And K-O CBS had 69.53% the win rate of K-O C22. Why would you compare one count with no indices to a different count with nearly two dozen?

  10. #10
    Norm Wattenberger
    Guest

    Norm Wattenberger: Re: Interesting study

    > And K-O CBS had 69.53% the win rate of K-O C22. Why
    > would you compare one count with no indices to a
    > different count with nearly two dozen?

    Hi T-HOP. Why not? How else would you measure the effectiveness of CBS in comparison to indexes

    I find quite humorous the comments that it is not correct to compare two counts unless you cripple one to look like the other. How is this useful? It's like saying that a Yugo is almost as good as a Formula 1 Ferrari. As long as you seal off eight of the cylinders, put on 13" wheels, a Yugo automatic transmission, leaf springs, and add 2,000 pounds of lead. How does this compare a Yugo and a Ferrari? Saying that OPP has 75% of the power of HiLo is seriously misleading. And, it's not even true if you did strip HiLo bare.

  11. #11
    T. Hopper
    Guest

    T. Hopper: Re: Interesting study

    > Hi T-HOP. Why not? How else would you measure the
    > effectiveness of CBS in comparison to indexes

    I just think it would be more useful to compare OPP to K-O Rookie, and then K-O Rookie to K-O Preferred or C22 or whatever. By adding one more number you put everything in context.

  12. #12
    Norm Wattenberger
    Guest

    Norm Wattenberger: Re: Interesting study

    You should be able to do this with the sims I just posted. BTW, keep in mind that if you ask for something on BJF, I am barred there.

    > I just think it would be more useful to compare OPP to
    > K-O Rookie, and then K-O Rookie to K-O Preferred or
    > C22 or whatever. By adding one more number you put
    > everything in context.

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
  •  

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.