Blackjack and Card Counting Forums - BlackjackInfo.com

  #11  
Old August 4th, 2011, 08:43 AM
UK-21's Avatar
UK-21 UK-21 is offline
Executive Member
 
Join Date: Apr 2008
Location: One of the green, leafy parts of the UK
Posts: 968
Default

Amen to the advice and comments above.

If, as like me, you do not have a dedicated bankroll to play BJ with, then the "how much do I need to take with me" question is a little broader. A contributor posted here in the past that he always took 40 units with him to th felt, and played for an hour, or until he'd lost the lot or doubled his money - whichever came earlier. Personally I never play unless I've two hundred quid with me, which represents 60 or 100 units dependent on the table min, out in the shires where I live. But if you're going to start to do this in a regular methodical fashion (as opposed to being a casual player, with a replenishable bankroll so to speak - as I am), then yes, adopting a managed bankroll approach is the way forward.

A couple of other one liners that may light the way . . .

On the UK game, with ENHC, even money only etc etc, I have calculated that a 1-8 spread will give a breakeven game - in actual fact it's slightly +EV, although the frequency of hands at TC+1, +2 etc has a bearing on the calculation. Conclusion? You need to be spreading 1-12+, and have a bankroll to cover the variance, in order to make it a worthwhile enterprise. At 2 units per hour average profit, on a 2 min table you'll still be making less than the NMW longer term.

The loss in EV from only having even-money rather than full insurance is not that significant, as you'd only take insurance against a dealer ace 8% of the time - and of those times you would have a proportion would lose not only the insurance bet but the main wager as well. A lot of US contributors will denegrate this loss, although it has been discussed on these boards at length. Have a look through past posts.

Although it's possible to get away with some hefty spreading when playing, it's also possible to get busted and have your membership cancelled. As playing opportunities are few in the UK, bear this in mind (particularly as some casino chains' BJ tables are now exclusively CSM dealt - the G Casino chain comes to mind).

Good luck and good cards.
  #12  
Old August 4th, 2011, 09:38 AM
London Colin London Colin is offline
Senior Member
 
Join Date: Aug 2009
Posts: 378
Smile Penetration, Penetration, Penetration

Quote:
Originally Posted by UK-21 View Post
On the UK game, with ENHC, even money only etc etc, I have calculated that a 1-8 spread will give a breakeven game - in actual fact it's slightly +EV, although the frequency of hands at TC+1, +2 etc has a bearing on the calculation. Conclusion? You need to be spreading 1-12+, and have a bankroll to cover the variance, in order to make it a worthwhile enterprise. At 2 units per hour average profit, on a 2 min table you'll still be making less than the NMW longer term.
What penetration were you assuming for the 1-8 spread?

For comparison, Arnold Snyder quotes an edge of +0.71% for a 1-12 spread with 6D, no surrender, (and presumably with insurance), S17, US hole-card rules and a pen. of 67%.

For the same rules with a pen. of 81% he quotes +0.77%.

I suspect 1-8 wouldn't quite break even at 67% with ENHC.

Quote:
Originally Posted by UK-21 View Post
The loss in EV from only having even-money rather than full insurance is not that significant, as you'd only take insurance against a dealer ace 8% of the time - and of those times you would have a proportion would lose not only the insurance bet but the main wager as well. A lot of US contributors will denegrate this loss, although it has been discussed on these boards at length. Have a look through past posts.
The small percentage of the time you want to take insurance corresponds with your biggest bets, however.

As I said, insurance tops the list of the Illustrious 18 strategy variations. In Blackjack Attack the value ascribed to it for a HiLo player is +0.117%, more than double the next most important - 16v10.

Winning or losing the main bet is not really relevant (at least not to EV). There is an opportunity to make a +EV bet if the dealer should show an ace, but the rules do not allow it.

If full insurance is worth about +0.177%, and we can only insure (via even-money) when we get a BJ, then its value must be reduced by about 95%, based on a 5% chance of being dealt a BJ. That would be +0.009%.

Of course with higher counts the probability of being dealt a blackjack rises, so the above won't be entirely accurate. I would think the average reduction must be at least 90%, though.
  #13  
Old August 4th, 2011, 01:04 PM
UK-21's Avatar
UK-21 UK-21 is offline
Executive Member
 
Join Date: Apr 2008
Location: One of the green, leafy parts of the UK
Posts: 968
Default

Quote:
Originally Posted by London Colin View Post
What penetration were you assuming for the 1-8 spread?

For comparison, Arnold Snyder quotes an edge of +0.71% for a 1-12 spread with 6D, no surrender, (and presumably with insurance), S17, US hole-card rules and a pen. of 67%.

For the same rules with a pen. of 81% he quotes +0.77%.

I suspect 1-8 wouldn't quite break even at 67% with ENHC.
You are quite right . . . the pen will affect the outcome. I used a set of figures, from Mr Snyder's BBiBJ if my memory serves me right, which I think were based on 75%. I haven't done the sums to the finer point that you have, but I would think that any shoe with 67% pen or less would make a game -EV.

Quote:
Originally Posted by London Colin View Post
The small percentage of the time you want to take insurance corresponds with your biggest bets, however.

As I said, insurance tops the list of the Illustrious 18 strategy variations. In Blackjack Attack the value ascribed to it for a HiLo player is +0.117%, more than double the next most important - 16v10. . .
Again, you are correct. I corresponded with Mr S personally around this issue, and he considered that the loss in EV for the even money only rule was perhaps a tenth of a percent . . . not some huge figure that some US players were claiming ("I'd never play in the UK, the insurance rule makes the game unplayable . . . yada yada"). He said that he'd never calculated it, and indeed I don't think anyone ever has as part of earlier discussions.


I think that the inherent variance in the game make the finer point of the EV decimals somewhat academic?
  #14  
Old August 7th, 2011, 10:53 AM
Nimiza Nimiza is offline
Member
 
Join Date: Aug 2011
Posts: 4
Default

Quote:
Originally Posted by UK-21 View Post
I think that the inherent variance in the game make the finer point of the EV decimals somewhat academic?
Well no because the Law of Large Numbers says the more you play, the closer you will tend towards your statistical EV, no matter the variance. So yes, if you're only playing a short while, then it's unlikely to make much of a difference. But that's equivalent to saying you once bet on red and won, therefore the negative EV of roulette is academic.

Incidentally a friend of mine wrote a quick program to ensure that the strategy we are using (KO) has a positive expectation. It does, and it orders around 0.5 - 1%. If you can manage a larger spread, get some backcounting going, then obviously this increases...
BTW this was programmed for British Blackjack, i.e. ENHC rule.
  #15  
Old August 8th, 2011, 07:13 AM
UK-21's Avatar
UK-21 UK-21 is offline
Executive Member
 
Join Date: Apr 2008
Location: One of the green, leafy parts of the UK
Posts: 968
Default

Thank you for the statistics 101. Sorry, but a tenth of a per cent on the EV (one way or another) ain't going to make that much difference (if any) for the vast majority of people who play the game. And when you consider the conversation was around someone playing at table mins, the pounds and pence equivalent, even over the longer term (if someone ever plays enough hands for the variance to even out to the expected average) will be negligable.
  #16  
Old August 22nd, 2011, 08:31 AM
Nimiza Nimiza is offline
Member
 
Join Date: Aug 2011
Posts: 4
Default

Quote:
Originally Posted by UK-21 View Post
Thank you for the statistics 101. Sorry, but a tenth of a per cent on the EV (one way or another) ain't going to make that much difference (if any) for the vast majority of people who play the game. And when you consider the conversation was around someone playing at table mins, the pounds and pence equivalent, even over the longer term (if someone ever plays enough hands for the variance to even out to the expected average) will be negligable.
1. The fact it's a tiny amount is missing the point. Rules like hitting on 16v10 during normal play have a tiny difference in expectation from standing...but we still do it for the sake of every extra increment in EV. That's our game.

2. Threw some quick numbers together. Assuming a big bet frequency of around 10%, a small bet of 3 (usual table min outside of London) and a spread of 1-12. If you visited once a week and stayed for an average of 5 hours a time, managing an average rate of 50 hands per hour, over 6 months you will bet approximately 41k. So a 0.1% change in EV still represents 41. Not exactly negligible if you're playing min bet.

3. With a positive expectation of around 1%, proportionally speaking a 0.1% change is 10% of our edge.

Last edited by Nimiza; August 22nd, 2011 at 08:33 AM.
  #17  
Old August 28th, 2011, 06:36 AM
UK-21's Avatar
UK-21 UK-21 is offline
Executive Member
 
Join Date: Apr 2008
Location: One of the green, leafy parts of the UK
Posts: 968
Default

So. . . one session a week for six months at 250 hands per session - 5,150 hands in total, spreading 1-12? 41,000 total wagered which equates to around 13,650 units. Expectation @ 1% is 136 units and the 41 you discuss represents 14(ish) units?

As an earlier contributor stated in a posting, for casual players who are never going to play anywhere near enough to ensure that the EV averages out (and thereby it being possible to predict the max variance either side of it) the inherrent variance will "swamp" (his word not mine, but I think it sums things up nicely) the finer points of the maths - which is why I don't think it's worth getting too hung up about it. If you disagree that's fine.

It must cost me around 12 in diesel every time I visit the house of chance as the nearest one is around a hundred mile round trip- perhaps that (for me) puts it in perspective? How far do you take the +EV consideration to?

Good cards.

Last edited by UK-21; August 28th, 2011 at 06:36 AM. Reason: Typo
  #18  
Old August 28th, 2011, 06:03 PM
FLASH1296's Avatar
FLASH1296 FLASH1296 is offline
Executive Member
 
Join Date: Oct 2007
Location: East Coast, U S A
Posts: 3,749
Send a message via ICQ to FLASH1296 Send a message via AIM to FLASH1296
Default

The issue of e.v. can best be viewed in extremes cases.

At one extreme on this continuum is a player who wishes

to play for a few hours a couple of times a month.

He can afford to lose a couple of hundred dollars.

At the other extreme is a Card Counter relying on his skills to pay his bills.

The former will have some fun a few times a month.

The latter MUST show a long-term profit, while earning money as close to

continually as is plausible.

e.v. does NOT mean the same thing to these two players..
  #19  
Old August 29th, 2011, 04:23 AM
UK-21's Avatar
UK-21 UK-21 is offline
Executive Member
 
Join Date: Apr 2008
Location: One of the green, leafy parts of the UK
Posts: 968
Default

You sum it up very nicely there Flash.

The critial issue is the variance. People frequently talk about the "long term" and some calculate the number of hands to reach "n0" where their results, in theory, should be close to the mathematical expectation. But the truth of the matter is that the variance doesn't just go away when this point is achieved - it just represents something different based on a different basis of measurement. The roller coaster ride doesn't become a shunt across the flats.

Two examples, a new player and a many years served seasoned one both set out to play a game with an EV of +1% using the same bet spreads etc etc. The new player brings $1,000 to the felt, turns it over three times and loses the lot. Expectation is a win of $30 (+1%), actual result is a loss of $1,000 (-33.3%); his loss is 100% of his lifetime wagering at that point. Variance has taken his money.

The second player has played for many years and over this period has wagered $6m. He also loses $1,000, but this represents just 1/60 of 1% of his lifetime wagering - statistically insignificant and very different proportion of the expectation from player 1 who lost 100%.

But . . . . they've both lost $1,000 playing the same game . . . . . .

As you rightly say, it means different things to different players.
 

Tags
advice, birmingham, england, newbie

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT -6. The time now is 03:20 PM.


Forum Software vBulletin®
Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Copyright 2005-2011 Bayview Strategies LLC