Comparative Strategy Analysis

Setting and influencing the dice roll is just part of the picture. To beat the dice you have to know how to bet the dice. Whether you call it a "system," a "strategy," or just a way to play - this is the place to discuss it.

Moderators: 220Inside, DarthNater

Mad Professor
Posts: 1830
Joined: Fri Jul 29, 2011 12:15 pm

Re: Comparative Strategy Analysis

Post by Mad Professor » Fri Sep 07, 2012 11:00 am

Hi Ahigh,

There is a Short Bus full of guys who would LOVE nothing more than to have the MP $204-Across Regression proven ineffective against a reasonably-sized SRR-6.8 roll-trial.

This should be good. I'm sure they are licking their chops (or other body parts) in eager anticipation of you showing how badly the MP-$204 performs at the SRR-6.8 level of de-randomization.

But what if it works? :o

What if the MP $204-Across Regression w/Passive-Aggressive Pressing shows a profitably-sustainable bankroll-doubling and re-doubling effect? :shock:

What then?

Since you overlooked the DO NOT USE ON RANDOM-ROLLERS warning (much to your chagrin); would it be unreasonable if I asked if you have coded the actual post-regression (and quite critical) press-schedule for the $44-Inside into your bet-script for this method, or has that also been left out by oversight?


MP

Ahigh
Posts: 266
Joined: Tue Aug 28, 2012 11:49 pm

Re: Comparative Strategy Analysis

Post by Ahigh » Fri Sep 07, 2012 12:23 pm

I'll put it through the ringer. And I want everyone to know that I'm not playing favorites with anybody.

My credibility is my most valuable asset, and I'm not picking sides. But I generally know what to tell you about your strategy is that the exposure has a spike in it. I think a better way to profit from a good shooter is to turn everything off on the comeout, but no pressure and just grind out the profit. The volatility from gambling is _not_ a good thing if you want to profit. It is as likely to hurt you as it is to help you unless you have an unlimited bankroll.

Mad Professor
Posts: 1830
Joined: Fri Jul 29, 2011 12:15 pm

Re: Comparative Strategy Analysis

Post by Mad Professor » Fri Sep 07, 2012 12:42 pm

Ahigh wrote:The volatility from gambling is _not_ a good thing if you want to profit. It is as likely to hurt you as it is to help you unless you have an unlimited bankroll.


Maybe that's an assumption you should put that to the test too. :D

When you run the MP $204-Across Regression w/Passive-Aggressive Pressing against the SRR-6.8 roll-sim; would your programining skills allow you to also test it for bankroll-volatility?

That is, can you test the MP-$204's DEEPEST bankroll-drawdowns on its way to doubling and re-doubling that very same bankroll?

That way, you would be able to show, without equivocation, whether "pos-ex (SRR-6.8) volatility HURTS as much as it HELPS"...while also showing what kind of overall bankroll-requirements the MP-$204 really has.

That would effectively end all of the seat-of-the-pants subjective speculation, and replace it with a modicum of objectivity, would it not? ;)

MP

Ahigh
Posts: 266
Joined: Tue Aug 28, 2012 11:49 pm

Re: Comparative Strategy Analysis

Post by Ahigh » Fri Sep 07, 2012 3:17 pm

I believe I could do that. It sounds like you understand my suggestion. I mean it is _boring_ to profit from a good shooter, or it should be. If you profit from gambling, that is fun as hell. And the flip side when you lose from gambling, that sucks big time.

Just like the house, nobody likes volatility that goes against you. It just sucks big time when you're trying to engineer a win, it just gets in the way.

And I don't think ANYBODY need instructions on how to gamble and win if they are lucky. Taking risks is pretty straight forward, and doing that leads to profits for those who are lucky.

Ahigh
Posts: 266
Joined: Tue Aug 28, 2012 11:49 pm

Re: Comparative Strategy Analysis

Post by Ahigh » Sat Sep 08, 2012 12:27 am

Here's a the last roll I recorded at home. This was on 2012-Sep1-10:19pm .. thrown from Stick Right 1 by yours truly.

Look at the contrast between the familiar 345x odds with 1,2,3,4,5 and 6 points compared to MP204.

53`62`32`42`42`53`61`22`66`61`54`22`43`52`54`43`54`52`61`21`51`42`43`62`41`21`41`54`21`44
22`31`32`66`31`64`55`55`65`62`33`33`54`51`54`31`66`41`32`42`42`21`21`22`21`42`11`32`31`11
51`42`65`21`42`54`41`52

I broke this up into two sections, because the first 23 rolls were LOADED with sevens.

Total rolls: 23
1) 5 10.87% - 16.67 = (-5.80)--------------------------------- 1
2) 12 26.09% - 16.67 = (+9.42)------------------------------------------------------------------------------- 2
3) 6 13.04% - 16.67 = (-3.62)---------------------------------------- 3
4) 9 19.57% - 16.67 = (+2.90)----------------------------------------------------------- 4
5) 8 17.39% - 16.67 = (+0.72)----------------------------------------------------- 5
6) 6 13.04% - 16.67 = (-3.62)---------------------------------------- 6

11: 2
21: - 3
22: ---- 4
31: 4
32: - 5
41: 5
33: 6
42: --- 6
51: - 6
61: --- 7
52: -- 7
43: --- 7
53: -- 8
44: 8
62: - 8
63: 9
54: --- 9
55: 10
64: 10
65: 11
66: -- 12

2) 0 0.00% - 2.78% = -2.78% (-0.64) 2
3) 1 4.35% - 5.56% = -1.21% (-0.28)- 3
4) 2 8.70% - 8.33% = 0.36% (+0.08)-- 4
5) 1 4.35% - 11.11% = -6.76% (-1.56)- 5
6) 4 17.39% - 13.89% = 3.50% (+0.81)---- 6
7) 8 34.78% - 16.67% = 18.12% (+4.17)-------- 7
8) 3 13.04% - 13.89% = -0.85% (-0.19)--- 8
9) 3 13.04% - 11.11% = 1.93% (+0.44)--- 9
10) 0 0.00% - 8.33% = -8.33% (-1.92)10
11) 0 0.00% - 5.56% = -5.56% (-1.28)11
12) 1 4.35% - 2.78% = 1.57% (+0.36)-12

Total sevens 8 - Seven outs 5 (62.50%) - Seven winners 3 (37.50%)
Pairs 3 13.04% - 16.67% = -3.62% (-0.83 rolls)
Hards 2 8.70% - 11.11% = -2.42% (-0.56 rolls)
HiLos 1 4.35% - 5.56% = -1.21% (-0.28 rolls)
H2 0/0 ( 0.00% - 2.78% = -0.64)
H4 2/1 ( 8.70% - 2.78% = +1.36)
H6 0/0 ( 0.00% - 2.78% = -0.64)
H8 0/0 ( 0.00% - 2.78% = -0.64)
H10 0/0 ( 0.00% - 2.78% = -0.64)
H12 1/0 ( 4.35% - 2.78% = +0.36)

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

The the next 45 rolls were 44 rolls with no red, then a seven.

Total rolls: 45
1) 20 22.22% - 16.67 = (+5.56)------------------------------------------------------------------- 1
2) 21 23.33% - 16.67 = (+6.67)---------------------------------------------------------------------- 2
3) 11 12.22% - 16.67 = (-4.44)------------------------------------- 3
4) 16 17.78% - 16.67 = (+1.11)------------------------------------------------------ 4
5) 13 14.44% - 16.67 = (-2.22)-------------------------------------------- 5
6) 9 10.00% - 16.67 = (-6.67)------------------------------ 6

11: ---- 2
21: ------ 3
22: ---- 4
31: ---- 4
32: --- 5
41: ---- 5
33: ---- 6
42: ----- 6
51: -- 6
61: 7
52: - 7
43: 7
53: 8
44: -- 8
62: -- 8
63: 9
54: ---- 9
55: ---- 10
64: - 10
65: -- 11
66: ---- 12

2) 2 4.44% - 2.78% = 1.67% (+0.75)-- 2
3) 6 13.33% - 5.56% = 7.78% (+3.50)------ 3
4) 6 13.33% - 8.33% = 5.00% (+2.25)------ 4
5) 7 15.56% - 11.11% = 4.44% (+2.00)------- 5
6) 9 20.00% - 13.89% = 6.11% (+2.75)--------- 6
7) 1 2.22% - 16.67% = -14.44% (-6.50)- 7
8) 3 6.67% - 13.89% = -7.22% (-3.25)--- 8
9) 4 8.89% - 11.11% = -2.22% (-1.00)---- 9
10) 3 6.67% - 8.33% = -1.67% (-0.75)---10
11) 2 4.44% - 5.56% = -1.11% (-0.50)--11
12) 2 4.44% - 2.78% = 1.67% (+0.75)--12

Total sevens 1 - Seven outs 1 (100.00%) - Seven winners 0 (0.00%)
Pairs 11 24.44% - 16.67% = 7.78% (+3.50 rolls)
Hards 7 15.56% - 11.11% = 4.44% (+2.00 rolls)
HiLos 4 8.89% - 5.56% = 3.33% (+1.50 rolls)
H2 2/0 ( 4.44% - 2.78% = +0.75)
H4 2/1 ( 4.44% - 2.78% = +0.75)
H6 2/0 ( 4.44% - 2.78% = +0.75)
H8 1/0 ( 2.22% - 2.78% = -0.25)
H10 2/0 ( 4.44% - 2.78% = +0.75)
H12 2/0 ( 4.44% - 2.78% = +0.75)

First 23

Image

Last 45

Image

All 68

Image

First 23

Image

Last 45

Image

All 68

Image

Ahigh
Posts: 266
Joined: Tue Aug 28, 2012 11:49 pm

Re: Comparative Strategy Analysis

Post by Ahigh » Sat Sep 08, 2012 8:03 am

Here's the modified random data with a SRR of 6.8. I went to great pains to ensure that random distribution was preserved with the exception of the desired SRR. So here's the code for anybody who wants to critique how I dialed the SRR down to the desired level starting with purely random data.

In laymans terms for those who can't read the code, I build up a list of sevens in the data, then I randomly select sevens out of the list of sevens and I replace those specific sevens in the roll data with a random result until the random result that I replace them with is something besides a seven.

At 100,000 events a SRR of 6.8 is absolutely going to make money with the MPP strategy. My basic findings are that IN THE LONG HAUL, you can count on about $1 per roll given an SRR of 6.8 on the MP204 strategy.

However, you can also LOSE $1 per roll in the short term (up to about 2000 rolls) before the long term trends take over.

Here's the code for how I modified the random data to achieve the desired SRR ratio.

I did some further testing, and the SRR needed to break even with MP204 is about 6.35 or so (for the very long haul with up and downs in the tens of thousands). I would have to run the computer for a couple more hours to get a more precise requirement at where the strategy starts to be overtaken by the edge in the long run. But even at 6.35 which is break-even in the long run, you can lose $20,000 in 5000 rolls when things don't go your way even at a SRR of 6.35 just the order of the numbers can get you. A combination of CONFIDENCE IN THE SHOOTERS SRR of 6.8 or higher *AND* committment to stick with it for 4000 rolls or more are key. If either one of these doesn't happen, You can lose thousands of dollars. If your shooter stays at SRR of 6.8 or higher and you stick with it for 8000 rolls, you can expect to get $8000 reward in the end after potentially enduring $2000 underwater in the first 1000 to 3000 rolls.

If your shooter has a SRR of 4 for the first 50 rolls, you are either burned or committed to follow through depending on your way of doing things and how much money you have to throw at the problem.

Code: Select all

    while( $srr > 0 && $sevens > 0 && ( $numrolls / $sevens ) < $srr )
    {
        my( @sevenlist );
        my( $sevens_to_delete ) = 1;
        $sevens_to_delete++ while( $sevens > $sevens_to_delete && $numrolls / ( $sevens - $sevens_to_delete ) < $srr );
        $i = 0;
        foreach( @roll )
        {
            my( $d1 ) = substr( $_, 0, 1 );
            my( $d2 ) = substr( $_, 1, 1 );
            if( int( $d1 ) + int( $d2 ) == 7 )
            {
                push( @sevenlist, $i );
            }
            $i++;
        }
        foreach( @sevenlist )
        {
#           print "Found seven at idx $_ value is $roll[$_]\n";
        }
        for( $i=0; $i<$sevens_to_delete && $#sevenlist >= 0; $i++ )
        {
            my( $slidx );
            my( $rlidx );
            my( $d1 );
            my( $d2 );

            $slidx = int( rand( $#sevenlist ) );
            $rlidx = $sevenlist[ $slidx ];

#           print "sevenlist: @sevenlist\n";

            do
            {
                $d1 = int( rand( 6 ) + 1 );
                $d2 = int( rand( 6 ) + 1 );
            }
            while( int( $d1 ) + int( $d2 ) == 7 );
#           print "Changing index $rlidx from $roll[$rlidx] to $d1$d2\n";
            $roll[$rlidx] = sprintf( "%d%d", $d1, $d2 );
            $sevenlist[$slidx] = pop @sevenlist;
        }

        $sevens -= $sevens_to_delete;

        printf( "Modified SRR on roll data is %.2f\n", ( $numrolls / $sevens ) ) if( $sevens > 0 );
    }
Here's the best case scenario for getting your buck-a-roll upward movement:

Image

Here's the you-gotta-stick-with-it-cause it can be a buck-a-roll-loss for the first couple THOUSAND rolls cases

Image

Image

Ahigh
Posts: 266
Joined: Tue Aug 28, 2012 11:49 pm

Re: Comparative Strategy Analysis

Post by Ahigh » Sat Sep 08, 2012 8:14 am

Now I'm going to make a few additional points here:

Point 1) I have busted my ass on running this strategy through my simulator
Point 2) I'm personally down about $600 from mistakes I made on the table related to this strategy
Point 3) In general, I do not recommend strategies for people to employ at the tables .. I recommend choosing the bet according to the edge
Point 4) Cost per roll is THE SINGLE MOST EFFECTIVE WAY to gauge your expected LOSS when you are at the tables
Point 5) If you expect to go to the tables and WIN, you are setting yourself up no matter WHAT YOU BELIEVE.

Be careful out there guys. I hope that my charts help you guys out, and I hope you guys appreciate what I'm doing.

If I were to charge for this software for what I get for programming on my real job, my total bill would be somewhere around five grand by now. I hope that collectively between all you guys, there's that much benefit and that it's not time just wasted. I only have so much programming manpower in me, and this is no small effort.

My ultimate goal is to put this software to work to tell me how to bet for exactly how I roll personally. So I am COMPLETELY DONE with MP204 analysis, now, guys. I like helping you guys out, but I have my own work to do with this stuff.

My final point is about those first 23 rolls up there. That is real roll data from me .. right before rolling 44 times with no seven, I had a SRR of 3! MP204 lost about a grand in 23 rolls on my amazing rolls (okay maybe I'm not amazing, but I did do this). That is an average of losing $40 per roll for 23 rolls. It might be appropriate to compare this loss level to someone betting a boxcar for $40 on every roll for 23 rolls without getting a single $1200 payday to try to wrap your head around what it has to feel like getting beat up like that. The difference is that you wouldn't expect to win $1200 in 23 rolls doing that, but you might expect to win with MP204 because the chances of winning are greater in the first 4 to 5 rolls.

But once you've lost that much, you start to rewind and wonder "WHAT WENT WRONG?"

And if the answer is that you expected to win, just be careful out there. That's about all.

MMMmmmm kay?

Love you guys!

Mad Professor
Posts: 1830
Joined: Fri Jul 29, 2011 12:15 pm

Re: Comparative Strategy Analysis

Post by Mad Professor » Sat Sep 08, 2012 8:47 am

Well first I want to thank you, Ahigh, for putting all of your effort into doing this project. I know how much work is involved, as do several others, because we have engaged in similar undertakings in the past.

I also want to thank you for proving that the MP $204-Across Regression w/Passive-Aggressive Pressing is as net-profitable in the hands of modestly-skilled dice-influencers as I said it would be.

In fact, your sim showed EXACTLY the SAME profit-production rate as the Testing, Testing...Formulized D-I Betting-Strategies experiment did for the MP-$204. In that one, ~20,000 de-randomized rolls tossed by three-dozen variously-skilled dice-influencers produced +$20,000 with the MP-$204.

Do I feel vindicated?

Sure, but I already knew that despite your extreme skepticism, that a properly run test would re-prove what I've been saying all along. So the fact that you've 'discovered' the same thing at this late juncture, is only icing on the cake....but very appreciative-icing nonetheless. :D

Yes, this betting-method will inevitably encounter some of that icky, evil volatility along the way to making all of that money in the hands of a modestly-skilled D-I; but frankly, if you aren't prepared for volatility; then a casino is the LAST place you'll want to be spending your money-making time in.

Again Ahigh, thanks for all of your efforts; it is well and truly appreciated.


MP

Ahigh
Posts: 266
Joined: Tue Aug 28, 2012 11:49 pm

Re: Comparative Strategy Analysis

Post by Ahigh » Sat Sep 08, 2012 10:02 am

I don't recall any skepticism of mine saying that a 6.8 SRR would show a loss with a lot of rolls thrown at it. And I never doubted any simulations of any kind.

My skepticism was in having a betting strategy that included bets with a 4% edge.

I haven't arrived at any new information for myself as a result of these tests. I just went and did what you asked me to do because you asked me to do it, and reiterated my concerns about losses in the short term.

Since I did what you asked me to do, let me just ask you to advise anyone who uses this strategy to be prepared for short-term losses in the multiple thousands of dollars until you get a good solid 4000 rolls from a DI with a SRR of 6.8 or better FOR THOSE 4000 rolls.

My concern is just that too many people are just applying this strategy and getting pounded with SRR's in the range of 4.0 to 6.2 in the first couple hundred rolls. If you're not concerned about it, that's fantastic. But there are a lot of people who aren't reading all the fine print and think that this is more than a $1/roll grind that assumes a good shooter plus 4000 or more rolls to yield results consistently.

2/3rd of the people applying this strategy get lucky. But I'm just talking about the other third.

Mad Professor
Posts: 1830
Joined: Fri Jul 29, 2011 12:15 pm

Re: Comparative Strategy Analysis

Post by Mad Professor » Sat Sep 08, 2012 10:48 am

Hi Ahigh,

If UNDER-FUNDED skilled-shooters use this method and choose to entirely ignore or discount any of my adequate-funding-required advice; then they do so at their own wanton folly and peril. In other words, they deserve to LOSE!

That is, if they choose to ignore the thousands upon thousands of times that I have already stated that an adequate volatility-tolerant bankroll is needed to play ANY of these higher-volatility methods; then frankly, they deserve the losses they get.

Plain enough? :geek:

The first vérité of advantage-play dice-influencing is that your positive-expectation edge will not gloriously manifest itself on each and every toss of each and every hand of each and every session. To think other wise would be foolish to the point of being rickdickulous!


MP

Ahigh
Posts: 266
Joined: Tue Aug 28, 2012 11:49 pm

Re: Comparative Strategy Analysis

Post by Ahigh » Sat Sep 08, 2012 9:43 pm

Rickdickulous? Really?

Alright.

Are you guys in love? For real... It's sort of gross hearing this stuff.

Why do you gotta bring up all the drama? Get a room you two. Always talking about each other and shit. It's just gross.

Ahigh
Posts: 266
Joined: Tue Aug 28, 2012 11:49 pm

Re: Comparative Strategy Analysis

Post by Ahigh » Sat Sep 08, 2012 9:48 pm

I'm really turned of by you and SuperRick at each others throats all the time. Can't you two just figure out how to not constantly talk about one another?

House of Orange
Posts: 1328
Joined: Wed Mar 07, 2012 11:09 am

Re: Comparative Strategy Analysis

Post by House of Orange » Sun Sep 09, 2012 7:38 am

Don't be so touchy. It's Rickdickulous!

Ahigh
Posts: 266
Joined: Tue Aug 28, 2012 11:49 pm

Re: Comparative Strategy Analysis

Post by Ahigh » Sun Sep 09, 2012 10:05 am

Well, I just want to focus on the work and not on personal stuff. If I had unlimited time, I would feel otherwise. But the lengths at which some people have gone to examine the personal lives of other people has become exhausting for me. Comic relief is great if we can keep moving on. But I want to FOCUS!

memo
Posts: 870
Joined: Sat Jul 23, 2011 4:47 pm

Re: Comparative Strategy Analysis

Post by memo » Sun Sep 09, 2012 10:25 am

For what it is worth Ahigh...

I can't say I really understand all the things you did or went through in your analysis, but I do appreciate the time and thoroughness that went into it...

The result jives quite nicely with my own personal experience with the MP204, and pretty much what I have surmised and posted in the past.

Thanks for reaffirming that.

Memo

Mad Professor
Posts: 1830
Joined: Fri Jul 29, 2011 12:15 pm

Re: Comparative Strategy Analysis

Post by Mad Professor » Sun Sep 09, 2012 11:14 am

There's no denying that high-volatility betting-approaches, including any that involve an Initial Steep Regression (ISR); are going to be subject to a lot of hand-to-hand and even session-to-session volatility-spikiness.

I don't think anyone has been silly enough to suggest otherwise.

Where you have high-volatility, you also have to have the STOMACH for high bankroll-variance, because even when you are playing with a valid in-casino advantage; your bet-funding has to be large enough (and the bettor themselves have to be variance-tolerant enough) to handle those inevitable drawdown-valleys on his way to achieving ever-higher profit-peaks.

Ahigh's experiment showed bankroll-drawdowns as deep as -$4000 (if I am reading his charts correctly). That equates to around twenty full $204-Across losses that this approach experienced...but let's keep our eye on the BIG PICTURE and acknowledge that those drawdowns occurred to the MP-$204 on it's way to making something like +$100,000.

Are there many players who are willing to endure drawdowns of -$4000 (or potentially even -$6000) in order to sustainably make net-profits exceeding +$100,000 or more?

No, probably not...but that doesn't mean there aren't any. ;)

It's certainly true that the vast majority of players aren't banked for this (neither financially nor psychologically); but that doesn't automatically exclude every advantage-player out there. :D

Here's the main takeaway though.

~Everything that I have always said about the MP $204-Across Regression has once again been proven correct. That applies to its bankroll-doubling and re-doubling UPSIDE, as well as it's volatility-subjective DOWNSIDE.

Is it for everyone?

Of course not.

But can it be used for what it was designed for...to maximally exploit multi-shooter, variable-skill group-DI situations far more often and far more compellingly-profitable if you are properly bankrolled (both financially and psychologically) to handle all of the shooter-to-shooter and session-to-session variant-flux?

Yer damn right it can be. :D


Edited to add more color.
8-)

MP
Last edited by Mad Professor on Sun Sep 09, 2012 11:34 am, edited 2 times in total.

Ahigh
Posts: 266
Joined: Tue Aug 28, 2012 11:49 pm

Re: Comparative Strategy Analysis

Post by Ahigh » Sun Sep 09, 2012 11:30 am

Yeah. I've only been working on this software for a couple of weeks now, and this was a good fit on the timing since I was implementing strategies anyway, but I just have to keep moving along. If I were to tweak it, I'd say if you're wearing the "edge is important" hat, bet the four and ten for $25 each instead of $44 inside and omit the 5 and 9 on the initial 204 across adding that to the even numbers, and if you insist on betting the 5 and 9 on the $204 leg saying the edge isn't as important as the consistency of dodging sevens, in that case I would suggest instead of missing opportunities on the aces, ace-deuce, yo, and 12, hop up those bets to engineer an equal pay when anything but seven gets rolled. It has to suck sitting there with 204 across and watching 6 horns come followed by four reds.

I know the strategy is sort of in the middle ground between these two alternate paths in terms of modifying the betting strategy, but if it's more about dodging the red and confidence of that really quick in and out, when you have misses on the horn, you are prolonging your exposure which can be dangerous.

As far as winning $100,000 .. it's still an average of a buck per roll, and it's absolutely a grind. But do _not_ forget that this house is built on the foundation of an SRR of 6.8. And even on that SRR, you can have those drawdowns. With an SRR or 3, 4, or 5, which absolutely happens, it can become downright scary.

I still feel compelled to urge caution to any readers contemplating using this strategy based on any charts provided with an SRR of 6.8.

Final words: THE HOUSE GETS LUCKY TOO! MIND YOUR EXPOSURE!

User avatar
Maddog
Posts: 456
Joined: Thu Jul 14, 2011 1:10 pm

Re: Comparative Strategy Analysis

Post by Maddog » Mon Sep 10, 2012 2:23 pm

Ahigh wrote:...if you're wearing the "edge is important" hat, bet the four and ten for $25 each instead of $44 inside and omit the 5 and 9 on the initial 204 across adding that to the even numbers, and if you insist on betting the 5 and 9 on the $204 leg saying the edge isn't as important as the consistency of dodging sevens caution to any readers contemplating using this strategy based on any charts provided with an SRR of 6.8.
Ahigh wrote:I don't recall any skepticism of mine saying that a 6.8 SRR would show a loss with a lot of rolls thrown at it. And I never doubted any simulations of any kind.

My skepticism was in having a betting strategy that included bets with a 4% edge.
One suggestion that I would like you to consider as you code your model, is the notion of adjusted probability. Wincraps and BoneTracker both consider this possibility.

Consider this question: "What IS the House Advantage if a number rolls with a probability that has been altered from the well known standard distribution model."

I think we all know (or should know) the basic formula for a Place 5 bet:
[(4/10)×7 + (6/10)×(-5)]/5 = -4.00%

But what if the probability of rolling a 5 before a seven is not 4/10? Is the HA still -4.00%? Can it be more? Can it be less?

Also keep an open mind that it is not only about Seven avoidance, which most measure as SRR. When your are avoiding the seven, you need to ask, what is being rolled INSTEAD of the probable seven? I see above that you had coded into your model that you replace the seven with one of the other 10 possibilities, which is reasonable approach and the same model that Chris used in his Get the Edge at Craps book. But what if the avoided sevens were NOT evenly distributed to the other 10 numbers. That instead they were distributed with most going to a few numbers, some going to remaining numbers? Theoretically you can improve HA vs a given bet even if you maintain an SRR of 6, by simply rolling more of another number then the probabilities indicate.

Proving that divergence from random probability is a whole other topic, but for the sake of advancing the usage arguments, and the reason we are all in this endeavor, let's assume that divergence is possible and talk about how it can be leveraged.

This following quote:
Ahigh wrote:...My concern is just that too many people are just applying this strategy and getting pounded with SRR's in the range of 4.0 to 6.2 in the first couple hundred rolls. ... But there are a lot of people who aren't reading all the fine print and think that this is more than a $1/roll grind that assumes a good shooter plus 4000 or more rolls to yield results consistently. ...
I really detest the "...a lot of people..." line. What people? What is a lot of people? who are these people? We have a total of maybe 20-30 people posting in a community of 500 or so. I understand your thinking, but I feel you give our members FAR to little credit for their maturity and ability to understand the risks involved with gambling. I for one, do not believe anyone here is so naive or childish that they need to be guarded and spoon fed. I think we all understand the "Rules" that have been extolled over and over again about proper planning and practice.

I would question; is there anyone on this board that does not know that if you are going to bet $204 across, that you must then multiply that by 12 for your session buy-in (~$2500). And does anyone here not know that your session buy in should never be more then a fraction of your overall bankroll? I would never play the MP204, because I never buy in for more then 500-600 per session. And I believe that many like me read it with interest and try to re-size it for our bankroll and comfort level.

Your final quote is classic!
Ahigh wrote:Final words: THE HOUSE GETS LUCKY TOO! MIND YOUR EXPOSURE!
But I would modify it somewhat:
Maddog wrote:THE HOUSE STARTS WITH THE EDGE! MIND YOUR EXPOSURE!
With that we can quite worrying if people cannot take care of them selves and focus instead on the much more important aspect of your good work, which is what allows people to take care of themselves. That is outlining the boundaries of the volatility that come with any given betting system.

Ahigh
Posts: 266
Joined: Tue Aug 28, 2012 11:49 pm

Re: Comparative Strategy Analysis

Post by Ahigh » Mon Sep 10, 2012 3:09 pm

You have got to compartmentalize.

If you build a house on top of non-random distributions with higher edge bets, my question is why not figure out how to get your distributions higher on the bets with lower edges?

Since I've added the transform-the-set functionality to my code, I can easily see my histogram charts on the getting transformed as I would expect.

If you're betting on outcomes that are 23, 41, 63, and 54, those are four hopping combinations that could be transformed to bets with lower edges.

So I'm saying you need a bet with lower edges, and then fit your throw to be within the newer betting strategy.

If all you're DI's are having problems rolling too many 5's and 9's, I'm saying that's a problem with the DI, not with a betting strategy that bets on the 4 and 10 instead of the 5 and 9.

You have to compartmentalize. Because, and I will say it again, you are using the foundation of roll bias to build a house that includes a betting system with a higher edge than what is necessary, and as a result, you're giving more of your player's hard earned cash to the casino.

If you're telling me that none of the shooters can help but to be heavy on the five and nine and they will never be able to figure that part out, then I will agree with you then.

But I think there are other tools to deal with those problems.

And strategies alone are built on the lowest possible edges for the highest chance of success.

End of story in my opinion until someone teaches me something I don't already know to change my mind.

Ahigh
Posts: 266
Joined: Tue Aug 28, 2012 11:49 pm

Re: Comparative Strategy Analysis

Post by Ahigh » Mon Sep 10, 2012 3:12 pm

And to, again, make the point that if you don't care about the edge, I can tell you right now that I could hop the 12 ON EVERY ROLL and make money with that strategy.

But I'm not posting up my "bet the 12 every roll" strategy to everyone.

The MP204 is not this extreme, but I use the extreme case to make my point.

High edge bets are not the building blocks for good strategies.

And DI's that roll bets that you can only get with high edge bets are DI's who could use some help.

Post Reply