Newbie with little experience!!

[TXAP420]

Hey guys,

The name is Hugh from Dallas and I am new to the forum. I got into card counting about 3 months ago. I have already wet my feet a little at my local casino which unfortunately is located in Oklahoma( home of the $0.50 ante. I was off to a good start getting up to $500 from $300 bet ended up losing it all back due to negative swings and antes. I was back counting while my girlfriend was sitting down making the bets. Any tips on how this can work in our favor in the future ?? Should I continue playing at these casinos ( there's a few) with the ante in play but with favorable rules like DAS RSA upto4x double on any %80 pen s17 or should I not play at all? And also a question that I have not been able to find the answer to.. When a new shoe is in play, the running count is at +1 with almost 6 decks remaing, is the true count still +6 therefore indicating for me to bet 5 units right off the bat? I may be wrong left and right but I have no one to ask and everything I have learned so far is from books and research on the internet with a fair share of YouTube videos. I am here to learn so any tips or advice would be greatly appreciated and if I am wayyyy off, please let me have it! I am very eager to learn and am like a little sponge ready to soak everything up! Thanks in advance!

[DSchles]

"And also a question that I have not been able to find the answer to. When a new shoe is in play, the running count is at +1 with almost 6 decks remaining, is the true count still +6 therefore indicating for me to bet 5 units right off the bat?"

This is a very scary misunderstanding on your part of the relationship of RC to TC, and you shouldn't be risking real money until you understand it. The true count (TC) is the running count (RC) divied by the number of decks remaining. So, with 6 decks remaining, you would need a RC of +6 to have a TC of +1, and not vice versa!!

Don

[mrw464]

^ As Don indicates, your confusion of RC to TC is pretty dangerous. I don't have math to prove this, but I think even with the ante, with 80% pen and those generous rules you should do just fine, unless there is heat.

[blackjackplyr]

I might be able to help with the math, maybe not.. but my understand is this.
average wager * average advantage = EV.

So lets say you play $10 min and spread to 2x60 (1:12 spread). Average wager is some where around $18 (purely guessing), with an average advantage of 1% advantage
18*.01 =.18

If my understanding / math is correct your EV is about 18 cents per hand, and it costs you 50 cents just to play, losing 32 cents per hand.
If it's wrong people show me the flaws!

Thanks!

[mrw464]

that could be true, in my mind i was considering some wonging and a 1:20 spread

[SiMi]

Howdy, Hugh-TXAP420!

I'm assuming you are using HiLo as your count, right? If that's the case, it seems that you might be a little confused about the Running Count and the True Count. The Running Count (RC) is called a 'running' count because you keep it going (or 'running' like a car that's idling) throughout the entire shoe or deck. You start at 0 with a new shoe and just add 1, subtract 1 or do nothing (for HiLo: 2-6=+1; 7-9=0; 10-Ace=-1) as the cards go by.

In a game with more than 1 deck with HiLo, you have to “adjust” the RC to take into account the fact that there are more 'copies' of each card than there are in a Single Deck (SD) game. Unlike in a SD game, in a 6 Deck shoe game (6D), there are 24 of every card to begin with instead of only 4 of each card. When you see the first card and it's an Ace, for example, your RC will go down by 1 (from 0 to -1) but that is only 1/6^th as significant in a 6D game with 6 times as many Aces as in a SD game.

To get a “true” picture of the count (and your edge), you must divide the RC by the number of remaining decks unseen. So, the True Count (TC) is the RC divided by the number of unseen decks or Decks Remaining (DR). TC =RC/DR. When that 1^st Ace came out, there were 6 decks remaining in the shoe so the TRUE COUNT was, at that moment, -1 divided by 6 or -1/6. (You could round this to zero or possibly -1, depending on the rounding technique you use.)

Many people estimate the unseen decks by estimating the number of decks in the discard tray and then subtracting that number from the starting number of decks. When you see about 1 deck in the discard tray,there are about 5 decks remaining in the shoe. So, you will divide the RC by 5 at that point to get the True Count. With a RC of +1 and 1 deck in the discard tray, you would have a TC of +1/5 which you would still round to zero.

In a typical 6D game, you usually won't get to a +2 or greater TC until several rounds have been played because you are dividing the RC by the large number of Decks Remaining. When the RC gets to +6 or so and/or the DR gets small, things can start to get interesting. If the RC = +6 and there are 3 decks remaining in the shoe, this makes the TC = +2 (+6/3 = +2TC) and you may have an edge at that point depending on the rules, penetration, etc. If so, your bet will normally go up then.

Your bet should be based on your advantage (bet more when you have more edge and vice versa) and your play can vary depending on the TC (these play-variations at given TC's are called 'indexes'). I use computer simulation programs to calculate where my advantage occurs in a given game and how much it is. The best one I am aware of is "CVCX" by Norm Wattenberger and you should buy it and the game simulator, "CVBJ", ASAP.

How much to bet, how much to spread, money management and cover, for example, are all very important and, if you haven't already done so, you should research them. Since you have a potential teammate, you can do a lot of interesting things with that help. It can get very advanced. Before you bet significant money, you should practice a lot at home, with computers and in smaller places with low minimums.

As for the ante issue, I ran a simulation on my program, BJ678, using the rules and penetration you said and guessed at a bet ramp of $10 -$120 (12:1 spread) Wonging out at -2. Your win rate is 1.2% and 22 cents per hand. The ante of 50 cents will more than consume your win. You will lose about 28 cents per hand on average after paying the 50 cent ante. If you play 100 hands per hour, you will lose $28 per hour over the Long Haul. Unless I'm missing something, you should find a place to play that does not charge an ante.

I hope this helps.
Bestof luck!!
SiMi

[Skull]

As for the ante issue, I ran a simulation on my program, BJ678, using the rules and penetration you said and guessed at a bet ramp of $10 -$120 (12:1 spread) Wonging out at -2. Your win rate is 1.2% and 22 cents per hand. The ante of 50 cents will more than consume your win. You will lose about 28 cents per hand on average. If you play 100 hands per hour, you will lose $28 per hour over the Long Haul. Unless I'm missing something, you should find a place to play that does not charge an ante.

Hey SiMi!

I am just trying to get the hang of my finally-recently-bought CVCX, so I did a similar simulation to check my results. I used Hi-Lo with Illustrious 18, wonging at TC <= -2, optimal 1-12 (10$-120$) betting ramp, 3 players, S17 DAS RSA with 80% pen.

I got a rather different 1.423% advantage or 15 cents per round.

Did I do anything wrong or it's just that there are so many details that it is hard to get all of them the same?

Best!

tmpsim.jpg

[metronome]

TXAP, welcome. I wouldn't make a habit of playing ante games. I do, on occasion back count at Choctaw while the wife plays VP. I end up standing around a lot!
And now they've installed continous shufflers on about 50% of the tables. Except for entertainment value, Oklahoma gaming is mostly a waste of time.
And the last time there, I actually saw one 6D game with $.50 ante and NO MID SHOE ENTRY.

[SiMi]

Hey Skull!

Thanks for doing that and posting it!!! I doubt you did anything wrong. I've suspected for a long time that the sim program I use (Blackjack 678) would produce different numbers. Your bet ramp was different than mine so I re-ran the sim and used all the parameters you did (except I had 7 players at the table because I've never figured out how to change that!). I burned 1 card (unseen), took insurance at +3. There's probably some other minor differences that I'm missing but I don't think the numbers should be that far off... Rules and pen. look the same (S17, DAS RSA, NO SURRENDER and 80% pen). Notice I used NO Surrender.

It seems the big differences are:

The AVERAGE BET. CVCX shows $22.18 while BJ678 shows $19 (it doesn't show cents). That's a big difference.
The win rate. CVCX shows 1.42% while BJ678 shows only 1.09% when I re-ran it just now with 20M rounds. That's also big.

My sim just now showed a win rate of 20 cents per hand which I would think would be the average bet of $19 times the average win of 1.09%. That math works.
On your sim, though, you show an average bet of $22.18 and a win rate of 1.423%. So that should be about 31 cents per hand. Yet, the sim window shows 15 cents per hand... so I'm confused there. I'm probably messing something up there...

I wish I could explain the differences but I can't because I don't know the details underlying the 2 programs. Because I'm not comfortable without knowing exactly HOW something like this works, I'm in the process of writing my own BJ programs for counting skills, including drills, a game sim and a stats sim in Java and will report when I'm done. I've been working on it almost a year now and it's coming along very nicely. I have the drills done and the game sim can play 6 deck shoe games with various rules and up to 5 players (1 human and 4 computer + Dealer) according to various strategies (BS or counting) so far. I will have to remove the human player to make it into a pure simulator.

If you see something I messed up, please let me know. In the end, for this particular question, the 50 cent ante seems to be a killer no matter how you look at it!

Thanks again!
SiMi

[Norm]

you show an average bet of $22.18 and a win rate of 1.423%. So that should be about 31 cents per hand. Yet, the sim window shows 15 cents per

Because 48.9% of hands were played.

I re-ran it just now with 20M rounds.

20M rounds is way too few. CVCX won't even allow you to run that few.

I will have to remove the human player to make it into a pure simulator.

A simulator is very different from a game. Just making a game autoplay would be slow beyond belief.

[Skull]

Hey Skull!

Thanks for doing that and posting it!!! I doubt you did anything wrong. I've suspected for a long time that the sim program I use (Blackjack 678) would produce different numbers. Your bet ramp was different than mine so I re-ran the sim and used all the parameters you did (except I had 7 players at the table because I've never figured out how to change that!). I burned 1 card (unseen), took insurance at +3. There's probably some other minor differences that I'm missing but I don't think the numbers should be that far off... Rules and pen. look the same (S17, DAS RSA, NO SURRENDER and 80% pen). Notice I used NO Surrender.

Yup, everything matches.

It seems the big differences are:

The AVERAGE BET. CVCX shows $22.18 while BJ678 shows $19 (it doesn't show cents). That's a big difference.
The win rate. CVCX shows 1.42% while BJ678 shows only 1.09% when I re-ran it just now with 20M rounds. That's also big.

My sim just now showed a win rate of 20 cents per hand which I would think would be the average bet of $19 times the average win of 1.09%. That math works.
On your sim, though, you show an average bet of $22.18 and a win rate of 1.423%. So that should be about 31 cents per hand. Yet, the sim window shows 15 cents per hand... so I'm confused there. I'm probably messing something up there...

As Norm explained, you probably forgot about the backcounting. Taking into consideration the 48.9 hands played, the math is right.

I wish I could explain the differences but I can't because I don't know the details underlying the 2 programs. Because I'm not comfortable without knowing exactly HOW something like this works, I'm in the process of writing my own BJ programs for counting skills, including drills, a game sim and a stats sim in Java and will report when I'm done. I've been working on it almost a year now and it's coming along very nicely. I have the drills done and the game sim can play 6 deck shoe games with various rules and up to 5 players (1 human and 4 computer + Dealer) according to various strategies (BS or counting) so far. I will have to remove the human player to make it into a pure simulator.

Nice! Do report when done!

Because I'm not comfortable without knowing exactly HOW something like this works, I'm in the process of writing my own BJ programs

I am like that as well and I even wrote in one of my first few posts here that I wanted to program my own stuff to really understand what's going on. Well, while I still do write some lines of code to really grasp the statistics behind BJ, when I downloaded the demos of CVBJ, CVCX and CVData, I quickly understood that I would NEVER be able to program anything even just a very tiny little bit similar to that, so I just resigned to my mortal and humble self and just bought everything

If you see something I messed up, please let me know. In the end, for this particular questions, the 50 cent ante seems to be a killer no matter how you look at it!

With that our two programs agree

Thank again!
SiMi

Thank you SiMi!

Great cards!

[SiMi]

Hi, Norm!

Thanks! I was hoping you'd help out. That explains some things.

I have found that running 20M hands or 2 Billion hands (the max) doesn't really change it much on BJ678. I just re-ran the same sim with 500M rounds and it showed 1.10% win rate compared to the 1.09% win rate for 20M hands. I've seen that over and over with BJ678 and I've learned it's just not worth the extra time for these little sims where you want to change things a lot and time is short. Hal Marcus says in the documentation that 120M rounds "gives very good results" and 500M gives "superb" results.

I will time my BJ game on "autoplay" for X rounds and see how long it takes. Even if it takes days, it's worth it if I can verify certain things to my own satisfaction. I hate not knowing how it's done... And, it's fun!

Anyway, thanks again!!
SiMi

[Norm]

Simi, that means that the RNG is no good for more than a relatively small number of hands.

For BJA Chapter Ten, I ran 20 billion per sim.