+ Register

[Nickmsi]

Thanks Sqzbox. . .

I agree you can speed up playing by using the last hand that completed a cycle to start a new cycle.

Could you also speed it up more  by using the last 2 hands after a cycle is completed to start a new cycle so you could be betting even earlier?

Cheers

Nick

[sqzbox]

Good point. I don't see why you shouldn't be able to back up further. Theoretically anyway. If you have completed an AP that is, for example, PPP, then why can't you start a new series with PP? What about PBPBP? Could you go right back to BPBP? According to theory it would seem that you can. But I wonder if that is pushing the limits and the result will simply be a fallback to standard expectation? Actually I'm not at all sure that there is any advantage in playing this way. Because of the situation where either of P or B would complete an AP I suspect that the result over time is going to come back to the basic probabilities of the game. Maybe the advantage is in the selectivity, or perhaps it somehow reduces the extremes.

[whopper1967]

Hi again folks, I understand to reset to hand number one after a win or an ambiguous no bet situation... But after a loss do you just continue on until hand number eight trying to complete? Or is it possibly a one shot and done deal per eight or nine hands and restart at loss?  Thanks

[Nickmsi]

Good timing with your question, Whopper.

The VDW as explained in this thread is the basic (original) method.

I have a tweaked VDW that will be used in my bot for Roulette No Zero tables.  Future plans call for the bot to play on line baccarat.

I just finished today testing my VDW to see the results if we used a Stop Loss of either -1, -2, -3, -4.

I only have the preliminary results based on 20 Sessions of 3,000 spins each for the 4 different Stop Losses.

Stop Loss = -1      Stop Loss = -2      Stop Loss = -3      Stop Loss = -4
Won 15/20 Sessions   Won 17/20 Sessions   Won 14/20 Sessions   Won 15/20 Sessions
Profit = 427 Units   Profit = 428 Units   Profit = 280 Units   Profit = 328 Units

So you can see using a Stop Loss does make a difference for my VDW for Roulette, I don't know if you will get similar results with the basic VDW method.

Next thing I have to do is test Stop Losses (-1 & -2) for 100 Sessions of 3,000 spins each to see if the results hold up.

Cheers

Nick   

[Tomla]

your moving along nick--very nice

[whopper1967]

I am definitely no mathematician but I was just reading the Wikipedia page on this theorem and a few more questions came up. As I read it in laymens terms, it seems to me that the theorem would only be a guarantee when it comes down to the 9th hand and the previous hands fell exactly as in the example on the Wikipedia page. In that case, I can see that there is no way that an AP will not be completed on the 9th hand but even though we know for sure that in that case that it is a 100% guarantee to complete, unfortunately we still do not know for sure which AP will complete. So does this still not put us back at a 50/50 chance? It seems as though the ambiguous no bet situations that we cannot bet are the only times that an AP are actually guaranteed to complete such as the occurrence of 3-6-9 banker with 1-5-9 player. Also when a tie gets thrown into the mix, I assume we would just have to restart the hand count at one because the third element of a tie. Hope this makes sense...thanks all

[sqzbox]

Firstly, ignore ties. I assume you are talking about the baccarat tie? Ties are a push - it is as if they don't exist when you are only playing B or P.

The guarantee is only that an AP will definitely occur somewhere. If it hasn't arrived by the 9th then yes, it will happen then but as you say can be either outcome as both will complete an AP.

[Nickmsi]

Yes, Whopper you are correct in your assumptions as Sqzbox points out you will have outcomes where either the P or B could complete the AP.

This is exactly the way I thought 2 years ago when I first studied VDW and I came to the same conclusion as you did so I did not pursue it as I thought it pointless.

When the VDW was revived by Priyanka on another site, I revisited it and with the fresh eyes of more experience here is what I saw:

A BET BASED ON MATH

The Van de Waerden theorem proves that you have to have a completed Arithmetic Progression(AP) within 9 spins/hands.

A NON RANDOM BET

It could care less if the last 200 spins only had 35 Reds, it could care less if you have a streak of 10 Bankers, it could care less if you has chops for the last 20 spins/hands, it could care less if you did not RTM, it could care less what the Standard Deviation is, etc.

A SIMPLE BET

Complete a 3 spin/hand AP.  That's it.

A LIMITED Bet

Within 9 spins/hands you will have a conclusion.  No waiting or no hoping.

A CONSISTENT BET

A bet that wins with regularity. Some might even call it a CWB, Consistent Winning Bet.

I thought finally, a bet that makes sense.  Now how can I make it win more than it loses.

STATISTICS was the answer.  This is a limited bet in that you can only lose -1 or -2 or -3 or -4 units during any 9 spin/hand cycle.  That is the limitation.  So, I tested to see if any of these 4 Stop Losses made a difference in the Profit and Losses.

See my reply # 123 and you will realize that Statistics do matter. Statistics is the key.

Cheers

Nick

[whopper1967]

  It looks as though on your test that a stop loss of one unit per series is giving you the most profit from the testing. I have been using it some on live dealer baccarat online. I reset the series count to one on any win or tie and only go to hand number eight. Other than the stop loss is there any other recommendation for baccarat? Thanks again for your help.

[whopper1967]

I do not have the means to test the play in simulation and I just play low stakes live dealer online baccarat. I know you have and are testing several different scenarios/options within the framework and I suspect there may be many more options that could be tested such as if you reset after a tie or continue, do you reset after a tie early in series versus late in series. I know some have said that most of the wins seem to come early. In that case maybe if there is a tie hand on the 7th hand, the simulations would show that it would be better to rest there and maybe if a tie occurred on hand 5 it would show that we should continue on. Would it possibly show that a certain AP seems to complete many more times for us where as maybe one such as the 1-5-9 would be a bad proposition. I assume it takes many, many simulated hands to safely arrive at these conclusions and maybe you have already done most of this but I thought I would just throw it out there. I mean if in a 250,000 sim a 1-3-5 completes 75% of the time and a 1-5-9 busts us 65% of the time....can we just dismiss that as random variance or does it take on a meaning....thanks for your time.

[Nickmsi]

Hi Whopper,

Yes, there are so many ways to test the VDW and I am sure I have not tested them all yet.

Let me just tell you some of my initial thoughts about the dilemma of a mutual bet.  Perhaps it will give you a little more insight.

At first I thought this VDW was a useless method because  if we win 50% of the time on non mutual bet and the mutual bet was 50/50 then it is a push, it had no merit.

But then I thought what if we bet ONE side when it is a mutual bet, like we bet Banker.  If it is 50/50 then we should win 50% of the mutuals which would be 25% of the total.   25% for mutuals plus 50%  on non mutual bets, does that not total 75% wins.

This type of thinking was based on assumptions and may not be mathematically correct, but it made me recognize that I needed some STATISTICS to support my thinking or to show another path to follow.

What we needed was some Statistics on the mutual bets, for instance, how often does a mutual bet occur? 

I know most of you don't have access to tools like a bot or simulator to give you these kinds of statistics, so let me tell you what my results were:

The Mutual Bet occurs only about 20% of the time.

The question is does this statistic help us and if so how?

Now you have 2 Statistics, 1 the Stop Loss Statistics in reply # 123 and the Mutual Bet occurrence statistic.  I hope you are beginning to see the importance of Statistics.

Cheers
Nick

[ADulay]

Stop Loss = -1      Stop Loss = -2      Stop Loss = -3      Stop Loss = -4
Won 15/20 Sessions   Won 17/20 Sessions   Won 14/20 Sessions   Won 15/20 Sessions
Profit = 427 Units   Profit = 428 Units   Profit = 280 Units   Profit = 328 Units

Nickmsi,

  This "stop loss" you're testing is for each set of 9, correct?   Lose two wagers and reset for a new set?

  AD

[Nickmsi]

Yes, Adulay .. .

The Stop Losses were based on each cycle or set of 9 spins.

They were for my VDW Roulette but I suspect similar results can be attained for baccarat.

Cheers

Nick

[ADulay]

Yes, Adulay .. .

The Stop Losses were based on each cycle or set of 9 spins.

They were for my VDW Roulette but I suspect similar results can be attained for baccarat.

Cheers

Nick

Nickmsi,

  Interesting angle.  I'll go back and look at the previous sheets and see how they change.

  AD

[ADulay]

For the 5 or so of us who are working with the "VDW" method of play, I'd like to condense down just what we're working with.

If the shoe starts out PP, then the wager is obvious under the 123AP.  Wager on P.
If it wins, you will have solved the process and can move on.

The next wagering opportunity presents itself with another 123 starting at hand #2.  How convenient!

Now, assuming you did not win that first wager at PPx, you would have PPB going into hand 4.  There is no AP that will fulfill the requirement to wager, so hand #4 is not wagered on.

After hand 4 you find your self with either PPBP or PPBB you have two choices. 

PPBP results in another "no bet" situation (neither 135 or 345 are available) leaving us only the PPBB situation to wager on with the 345AP.

See anything developing here?

With the PPBP set up you should already see that although there is no wager on hand #5, hand #6 WILL be bet and it WILL be a Player wager.  Both the 246 and 456 AP's come into play.

So, after hand #5 we have a PPBPB or a PPBPP the wager will be on P.    Pretty straight forward.

Now, let's change the result to PPBPPB (loss on hand #6).  What's the next wager?  147 is correct.  The other two iterations of "7" don't fit the box. (567 and 357).

This is also as far as you can run with the VDW on this set.  If you lose the 147AP, the 8th and 9th levels both are simultaneous bets which is no good for our purposes.  Back up and restart.

Now obviously, should you win ANY of the wagers, the set is complete and you can restart another group of 9.  Where to restart?  Simply back up and find an appropriate start based on being able to make a wager.   The example would be winning the 123AP means back up to hand #2 and start another 123AP.  Basically just stay on the side that just won if playing the 123, 234, 345, set.

The same goes for any wins on the 135, 246 and 357 sets.

The bottom line?  (For those of you who have read this far) reads like this:

2's go to 3+.   Z-runs of 3 continue on.

That's it!!

Yes, all of the writing and mathematics of the VDW Theorem show that in a wagering situation.

If you lose on a 12(3), you wait two hands to wager on the 246 or the 345 if it went the other way.   In any case, it's still a "2 goes to 3" situation.  If it loses, you're right back to the 1-2-x restart!

If the 2-4-6 wins, it drops right into a 3-5-7 and so on.

You do not wager when both sides can win, you back up and restart.

So, one more time:

#1   All 2's go to 3.  (and beyond)
#2   All ZZ runs of 3 (or 4 using the Ellis count) continue on.
#3   On any loss, move to the next AP if available otherwise back up to a suitable restarting point.

Put pencil to paper and you'll see this for yourself.   

I didn't invent this or even recommend it.  I'm just showing what the VDW system play works out to in real life.   Please continue to test and report any results that you determine would help.

AD