Math Guru Advice

R

Rob McGarvey

Well-Known Member
Let me, entertain you! ;> I have a serious question for the Mayor, MathProf, T-Hop, Alienated, AdM, etc. It combines risk aversive indice theory, theoretical randomness, and the fact the with online bj we get our advantage from the bonus and playing basic strategy (and maybe a few variations like standing with 3,2,4,5 v 10 up)

RA theory basically states that the more you are betting above the optimal wager, the higher the RA indice should be. See BJAttack pages 311-319. They show you 10 v 10 and the indice is +4 where the RA indice is +7. So between these two indices +4 and +7 we stand instead of double.

The optimal wager for online bj is the smallest one because we are playing a -ev game without the bonus. The required action and time constraints have an effect on that opinion, and we could be should be making larger wagers. The game shuffles after every hand, so the cards have "no memory" in this case, which can be a good thing after you beat the dealer (or vice vee) 10,A 10,10.

I have found that by avoiding certain double ups and splits the game swing stays under control during higher bets being made and I am able to nibble way way into some rather nice wins doing so. I am a lucky someB, but I do believe there is more to it that just getting good hands from an RNG.

I do know that I am probably using a cleaver to butter my bread with here (using the wrong tools for the right job), but would like a few opinions on this matter. There has to be a rational explanation for all of this.

Thanks a Billion! ;>
Robby M
 
T

The Mayor

Well-Known Member
I thought of this too

I called this "risk adverse basic strategy" -- the idea is that if we have a bonus and we have to play a certain amount of action, then we are willing to give up a little of our bonus in EV in order to avoid SD wiping out our bonus. I talked with Don a bit about this, and got shot down for it. Math Prof gave me a bit of positive feedback.

I then did a bit of experimentation with some sim's, for example, avoiding most close doubles and splits (where more money goes on the table for very little increase in EV). The net effect was that there was some theoretical gain in the fraction of interest:


(BONUS + EV)/SD,
​

but the hands that these playing deviations came on were sufficiently rare, that overall it wasn't worth the effort. I recall numbers like having a game that was -.341 becoming one that was -.347, but reducing the SD as well. Since we are playing for bonuses, it was at least worthy of note.

I spent some time with the tables of Wongs PBJ open, and I was going to write the definitive article on this, but after computing about 5 of the numbers by longhand, it dawned on me that this was a lot of work and I stopped.

I don't recall where, but I posted the few numbers I came up with. I think it was on bj21 green chip about February of this year.

Given that playing for Bonuses is not a long run phenomenon, I don't think the same sort of analysis goes into it. But, my advice, avoid close doubles (e.g. A-7 vs. 2, 11 vs. A) and close splits (22 vs. 2 or 3, etc). If it's close and you can avoid putting down more money, do so.

--Mayor
 
A

alienated

Well-Known Member
A disclaimer

Rob, I'm flattered you would make the error of including me in a list of otherwise genuine math whizzes, but my mathematics knowledge is very limited. Any low level math in my posts is strictly following the generous spoon feeding offered in Peter Griffin's book, as well as small morsels contained in the posts of the true 'math gurus', some of whom you mentioned, in which the time is taken to explain things in a way that is comprehensible to an 'educated nonmathematical reader' - assuming mathematicians will allow the notion of a nonmathematician being educated. ;-)

So for the record, I am not a mathematician, let alone a math guru, and have never claimed to be so. Still, I guess there's always dreaming...
 
R

Rob McGarvey

Well-Known Member
Re: I thought of this too

Okay. I think we are barking up the right tree, but we can't see the opossum just yet. ;> I posted it at BJMath just now to see if I can get some similar positive feedback. I often find myself ahead by a good stretch on top of the bonus and depo, and in the interest of keeping all of that extra money do not want to compromise it by doubling the risk to obtain an extra +.05 or so. This weekend I am leading my troops into battle to take on a $500 bonus for a $250 deposit. No minimum play requirements and you can play any game. The catch is the $500 can't be taken out, and if you lose into your $750 start off it is coming out of your money. Microgaming casino. It's a short term bonanza, not a long run haul, and we hope to shut this place down the same way we did the Godfather Casino. I know you don't like progressions posted so I will refrain from doing that here. Anyone with good suggestions please take a shot at this situation. Min bet is $2, max is $200.
 
R

Rob McGarvey

Well-Known Member
MathProf Responds:

Mathematics of Blackjack Discussion

Re: Internet Gambling

Posted By: Rob McGarvey <[email protected]>
Date: Saturday, 14 December 2002, at 10:00 p.m.

In Response To: Internet Gambling (MathProf)

"I am sorry but I didn't quite understand what your question was. I am not that familiar with all the aspects of Internet Gambling, so I may not be able to answer it. Steve Jacobs knows more about it, and he frequents these pages, so maybe he can help you."

Okay, I hope he sees this message and tosses his two cents in.

"My sense is that with Internet BJ there are special bonuses and promotions that are the basis for your EV. Sometimes, the Variance within the Game can help in these situations. For example, if they have Rebate on Losses than high Variance will be better."

You are right. The EV comes from a matching bonus, some as high as 200%. Our ROI can come close to this 200% but our actual EV comes in below 20% since they can require quite a bit of action before you can pull the money out.

Thanks a Million! ;>

[edited for content ]
 
T

The Mayor

Well-Known Member
Re: I thought of this too

As you said earlier, the betting strategy that is optimal is to play the min bet allowed. You can usually burn through the required betting in about 2 hours (or much less on MG if you turn off the sound). Avoid those plays that increase SD without increasing EV very much. I think it is pretty straightforward. Absolutely DO NOT play progressions. They can cost a bundle if "everything that can go wrong does go wrong." Those progressions that settle for a lot of small wins at the expense of an occasional large loss are the worst of all. Someone will get really hurt.

Bet small, avoid SD. Simple stuff.

--Mayor
 
zengrifter

zengrifter

Banned
What's 'McGarvey's Grind'?...

...and when do you recommend its use? zg

(note to moderator - RM mentioned the 'MGrind' and if its not acceptable that we discuss a method that he initiated discussion of herein, then please BUST THE ENTIRE THREAD, otherwise I really want to know the whats and whys of his Mgrind technq. Thanx.)
 
T

The Mayor

Well-Known Member
I don't see where

there is mention of "MGrind" or "McGarvey's Grind" ... let me know more specifically, and I will take the appropriate action.

Again, it is board policy to not allow posts that advocate the use of progressions.

--Mayor
 
zengrifter

zengrifter

Banned
Re: I don't see where

You are right - here is McGarvey's original post (below) wherein he DOES seek some sort of reconciliation between the statistical science and his apparent creative use of a progression technq he calls "McGarvey's Grind" - I guess YOU edited out part of his response to MathProth. CENSORSHIP! zg
---------------------------------------
From McGarvey's original post HERE -

"... You are right. The EV comes from a matching bonus, some as high as 200%. Our ROI can come close to this 200% but our actual EV comes in below 20% since they can require quite a bit of action before you can pull the money out. I have been getting great results using a combo of progressions which I call MCGARVEY'S GRIND and I am hoping that there is some solid math that will tell me why it seems to be working. I know progressions by their lonesome are bascially worthless, but coupled with the EV and tamed by the risk aversiveness that can be applied to the larger bets I really think there is a hidden edge someplace..."
 
R

Rob McGarvey

Well-Known Member
Re: I don't see where

I think the Mayor is doing a fine job and has the right to disclude any info that I may post here. We have all read about chasing bets like a typical loser does after a loss, and making parlays when winning as cover plays **within the scope of our advantage** at a table game of blackjack. It can be good cover for onLine play, but I also think that there is a possibility **within the scope of our advantage** for changes in bet size, and can include risk aversive measures since our advantage comes directly from the bonus money.

"I know progressions by their lonesome are basically worthless"
 
T

The Mayor

Well-Known Member
Re: I don't see where

Exactly! Rob knows my policy -- I just snipped out the portion advocating a progression, and left the rest unsnipped. I am going to be very strict about the no-mythology policy, but if a post is 95% non-mythology, then I think a little gentle snipping will do the trick in those cases.

--Mayor
 
zengrifter

zengrifter

Banned
Re: I don't see where

The way I read the original, Rob was seeking some validation for his 'McGrind' progression method - by deleting that element the whole context shifted to something else almost entirely. zg
 
T

The Mayor

Well-Known Member
Re: I don't see where

You said:
"I know progressions by their lonesome are basically worthless"

Take away the words "by their lonesome" and "basically" and I'll agree with you.

--Mayor
 
zengrifter

zengrifter

Banned
All the way to BJMATH...

... and NOBODY there even asked what a M'Grind was! BUT the eneffable John May did post an encouraging response tip (below) BUT he doesn't elaborate, which is vintageMay.

A thought has occurred to me - in spite of the NO MYTH RULE at CC.com, PROGRESSIONS CAN BE DISCUSSED, but such discussions are context-sensitive. For example, we can talk about the CORRECT way to scientifically use negative-progression technqs to win at BJtournies - BJtournies are alot more like BJ than i-casino bonus grifting - and we can certainly discuss progression betting as effective cover play (i like -progressions in +counts and and +progressions in -counts) - and the so-called 'parlay' is a +progression that has been the basis of many scientific BJ practioners since the early 70s - so progressions in this sense are NOT off limits even at CC.com, as far as I can tell...

... So now what in the h**l is a McGarvey Grind, for Da's sake?? zg

----------------------------------------

Mathematics of Blackjack Discussion

Re: McGarvey's Grind

Posted By: John May
Date: Sunday, 15 December 2002, at 2:38 p.m. bjmath
RE: McGarvey's Grind (ET Fan)

Progressions can help with certain types of match-play bonus, in the same manner that progression-type strategies may help with tournaments.
 
You must log in or register to reply here.
Top