Not doubling against dealer 2,3,4 at high count.

[bj bob]

..... I get a boner just thinking about doubling down with a max bet against a low card.

Moo, that line indeed proves the intellectual insight of this forum's elite members. :devil: You will have saved me a ton of money on the Blue Pill by simply having my wife dealing me dd combos with her 4 exposed. Thanks again! :grin:

 

[johndoe]

I'll buy 'em!

If any of you want to sell me any 10,11 vs 4-6 hands to decrease your variance, I will quite happily buy them, or at least pay for the double.

 

[UK-21]

Well, according to the wizardofodds, the EV of doubling 11 v. 4 is 56.6%. That's a assuming a neutral count, which it's not if you have a max bet out.

So, let's say 20 units doubled is 40 units*.56 = 22.4 units

22.4 units is a nice days work. I'm doubling down whenever I have 11 v. 4 with a max bet out.

WRONG . . . . .

If you read my original post in full, and a subsequent one, you'll see I am only proposing these plays at TC+3 and above - 8%ish of the time.

So 22.4 units x 8% = c1.80. Spread that over 8 hours of play and you can see the hourly cost is less than 0.25 units. It's not as simple as this of course, but it does put into perspective your assertion that it has a huge cost attached to it. You could of course end up with a toss hand and lose and so you'd drop an additional 20 units, or 10 hours worth of profits?? Is that worth the risk for the 8%ish edge that this play carries?

I'm still waiting for some help with the figures I produced. Is it something you can help with?

 

[StandardDeviant]

I'm still waiting for some help with the figures I produced...

Newbie, check out this article (http://www.beyondcounting.com/pdfs/scavengerbjfo.pdf (Archive copy)) and, in particular, the charts on pp. 13-14. They get at a certain aspect of the topic. It's not the full answer, but adds to the discussion.

 

[Blue Efficacy]

I don't like the idea of being the favorite to win twice my max bet. Just too darn much variance.

 

[UK-21]

I don't like the idea of being the favorite to win twice my max bet. Just too darn much variance.

Favourite, but by how much? According to Moo, 11v4 is just 6% edge on that hand. Allowing for the fact that in a pos count there will be more tens coming this will increase to what, perhaps 8%, 10%? Playing hands with a 9 or 10 it will be even less. Yet with a 10% edge we double our already high, possbily max, bets.

 

[UK-21]

If any of you want to sell me any 10,11 vs 4-6 hands to decrease your variance, I will quite happily buy them, or at least pay for the double.

Again, please read my original post. I'd play the 9/10/11 v 5 or 6. But against a 4 it becomes far more riskier - and more so against a 2 or 3. You can certainly buy these off me - the only proviso I would make is that if I had a hard nine, you took the double and drew a 2 for a standing 11 and it lost you'd refund my original bet.

What do you reckon he EV on that is?

 

[UK-21]

Newbie, check out this article (http://www.beyondcounting.com/pdfs/scavengerbjfo.pdf (Archive copy)) and, in particular, the charts on pp. 13-14. They get at a certain aspect of the topic. It's not the full answer, but adds to the discussion.

Thank you indeed. I haven't read this before - it's very thought provoking, and as you mention, touches on the basis of my thread. The probability charts are useful too. A couple of considerations fall out of it:

If you were the scavenger, would you jump in and buy a hand if the unit value of the "civilian's" was much higher than your own - thereby adversely impacting on your own RoR for your bankroll (if you play with one)?

By selling your hand as a "civilian", there still seems to be a win-win situation on the return longer term, and on the hands I'm proposing to deviate this looks to be split roughly evenly (page 11)?

With the hands I'm proposing deviating from the normal betting ramp, there is only one where there'd be the possibility of taking a second card - 9 v 2, and then pulling a 2 to give an 11. In a pos count this would then become a very pos expectation hand but if you'd doubled down on it you'd be stuffed. Every other hand would be a single draw card only, and so on these there would be no instance of a pos expectation for the scavenger and a neg for the "civilian". I think Mr G's article revolves more around taking advantage of bad play (not doubling soft hands for example) than picking up the slack on the risk adverse play that is under discussion here.

The main, and biggest, difference between myself and Mr G (and any other scavengers out there) is that I don't play the game for a living and so I'm not looking to squeeze every hundreth of a per cent out of the game to maximise my EV.

Thanks again. Very useful.

 

[moo321]

Favourite, but by how much? According to Moo, 11v4 is just 6% edge on that hand. Allowing for the fact that in a pos count there will be more tens coming this will increase to what, perhaps 8%, 10%? Playing hands with a 9 or 10 it will be even less. Yet with a 10% edge we double our already high, possbily max, bets.

No, it's not 6%, it's 56%! You got that number by multiplying it by the frequency with which it occurs. The fact remains that by refusing to double a max bet against a four, you are giving up a day's wages.

 

[moo321]

The main, and biggest, difference between myself and Mr G (and any other scavengers out there) is that I don't play the game for a living and so I'm not looking to squeeze every hundreth of a per cent out of the game to maximise my EV.

There are much better ways to cut variance.

 

[Blue Efficacy]

Not doubling against the 2-4 would certainly make cover.

After all, for a cover play to look really stupid, it helps if it actually is really stupid!:laugh:

Seriously, especially in a shoe game you can spend a decent amount of time looking for just that opportunity some days.

And you're going to let that slip by for the sake of reduced variance?

You could also reduce variance by spreading 1-4 in a 6d game. Sure you'll barely break even, but you don't have to worry about getting burned when you have the big bets out, either.

 

[UK-21]

No, it's not 6%, it's 56%! You got that number by multiplying it by the frequency with which it occurs. The fact remains that by refusing to double a max bet against a four, you are giving up a day's wages.

I didn't. I got that number by inferring the win rate was 56%, not the EV was 56%. My error.

Sorry, but I'm still not convinced that the loss in profitability is as bad as you state. If not making this play cost 22+ units per day, over 8 hours of play, with a 2 unit per hour expectation, this would mean the loss would be greater than the longer term gain - ie the game would become overall a negative expectation one. This is not the case. How can it be if you only hit rather than double this hand 8% of the time, and there is a positive expectation of a win with both?

 

[UK-21]

Following on from Moo's post where he quoted a win expectation on one of the hands under discussion taken from the WoO's site (appendix 1), I have visited his site and pulled the other win expectation figures for the other hands. With these I have crunched a fourth set of calcs that calculate the difference between the doubling win expectation and hitting win expectation for each hand, based on it's frequency at TC+3+ per 100,000 hands, my betting ramp (1-8) and the unit bet value. Contrary to Moo's post, if you hit rather than doubled 11 v 4 you would not be giving up all of the longer term win expectation of 56% of your bet as there'd still be a win expectation if you just hit.

Please look at the attached sheet and the difference between the doubling/hitting win expectations and what they work out to in ££s. The figure at the bottom is the overall loss for these (average of) 275 hands per every 100,000. Divide this by 100,000 to give the average cost per hand played in general, and by 100,000/(70*7) to give an average daily cost (based on 7 playing hours in a day, 70 hands per hour). The overall daily cost works out at 56p (1/6 of a unit). The figures are, of course, based on OTT BS and in practice the win rates will increase as the TC goes up. But they won't change so much that the output figures to fall out of the exercise are significantly affected (they won't increase by, say, 6 times).

If you divide the sample of 100,000 by the 275 instances of these hands at TC+3+, it works out at an instance every 364 hands. Divide that by 70 hands per hour, you get an average of the frequency that these plays turn up - once every 5 hours or so. Perhaps that puts it all in perspective. There's just no way that hitting rather than doubling on these plays at TC+3+ can have a huge impact on the overall EV of a game as, on average, you just don't make those plays often enough for it to do so.

I would still welcome anyone's calculations that can fine tune my own.

 

[psyduck]

Newb,

Have you asked yourself why players have advantage when TC is high? Is it because you win more hands? Or is it because you have more successful doubles and splits?

I still say what I said early in this thread: if you hesitate to double when you have your max bet out, it means your max is too big for you.

 

[StandardDeviant]

If it looks like a duck, walks like a duck, quacks like a duck...

...if you hesitate to double when you have your max bet out, it means your max is too big for you.

Not necessarily...doubling for less lowers EV and also standard deviation. If one wants to lower volatility of bankroll growth, making smaller bets is one way to do it.

Not all players want to drive to max EV, some want to grow EV subject to a constraint on volatility.

 

[ycming]

The purpose of RA indexes is to increase EV with no increase in risk. But, if you simply start making strategy changes to reduce doubling and splitting, this will reduce EV. And if you reduce EV, overall risk will increase. If you are concerned about adding to your initial bet, the solution is to reduce the initial bet (i.e. reduce your unit size), not mess up your strategy.

I think this post is the most informative to your question mate!

"And if you reduce EV, overall risk will increase."

Ming

 

[psyduck]

"If it looks like a duck, walks like a duck, quacks like a duck... "

....it is still not psyduck!

You do not seem to get my point. Using a smaller max bet and not hesitating to double already lowered variance.

Just my opinion. By no means am I pretending to be an expert like you.

 

[StandardDeviant]

I think this post is the most informative to your question mate!

"And if you reduce EV, overall risk will increase."

Ming

Except that if you reduce EV by lowering your bet, overall risk, as measured by standard deviation, generally decreases, not increases.

 

[StandardDeviant]

"If it looks like a duck, walks like a duck, quacks like a duck... "

....it is still not psyduck!

You do not seem to get my point. Using a smaller max bet and not hesitating to double already lowered variance.

Just my opinion. By no means am I pretending to be an expert like you.

Psyduck, I think we're actually both right. We're just saying that same thing different ways! At the end of the day, the more money on the felt, the higher the risk.

 

[psyduck]

At the end of the day, the more money on the felt, the higher the risk.

The higher the possibility of profit, too (at high enough TC, of course). It is all about risk and reward. If I want zero risk, I stay home.