Increased win rate in shallow pen w/ SC'ing 5's

bjcount

Well-Known Member
#1
Ok sim fans help me figure this one out.

Below are the results of using a balanced count, LVL2, RPC & ZEN, both reckon the 5 with a tag of 2.

Setting up the sims using a side count of 5's for betting purposes with a value of +1 gave me a steadily increasing win rate as the pen got worse.

Could it be that the extra 5's in the deck which help make our large number of stiff hands we're dealt is offsetting the effects of poor pen? Or is it just that we are putting more money out, increasing our variance, and in turn substantially increasing our risk?

Notice at 5/6d pen we get a decrease in the win rate by using the 5sc.

Some thoughts please.

BJC


Action and win rate is per hour
(DD game is 1-6spread same rules)
(zen bet spread is a 1-12 opt not the noted in the table)
 

Attachments

#2
How are you implementing the 5's sidecount? If you have it set up as an unbalanced count, betting in running count mode, that could explain it, as such counts tend to have you overbetting deep in the shoe.

Note the dramatic increase in action with the sidecount. Guaranteed this has more to do with the way you are using the sidecount than the effect of the 5's themselves.
 

bjcount

Well-Known Member
#3
Automatic Monkey said:
How are you implementing the 5's sidecount? If you have it set up as an unbalanced count, betting in running count mode, that could explain it, as such counts tend to have you overbetting deep in the shoe.

Note the dramatic increase in action with the sidecount. Guaranteed this has more to do with the way you are using the sidecount than the effect of the 5's themselves.
The only gains, through all my test sims, were incurred when you left the counts balanced (normal tags) and setting the side count as a betting strategy, not a playing strategy w/ MP charts. If I understand the betting strategy correct it would work similarly to HO2's Ace side count, adjusting the RC based on the number of cards left, rich or depleted (side counted) prior to placing your bet, and then use the unadjusted RC for playing decisions.

(understanding here that HO2 is not ace reckoned, just using it as an example.)

BJC

edit: I haven't tried to develop a MP chart for this scenario yet.
 
Last edited:

jack.jackson

Well-Known Member
#4
Floating advantage

My only explanation would be since, your adding an extra +1, and since your that late in the deck, your TC becomes very sensitive, and must be accurate. Since the fact, that adding the extra +1, for betting, could have you betting slightly early, making your results slightly inferior to the TCs you normally bet at. The Floating advantage, could be partially to blame as well.
 

bjcount

Well-Known Member
#5
jack said:
My only explanation would be since, your adding an extra +1, and since your that late in the deck, your TC becomes very sensitive, and must be accurate. Since the fact, that adding the extra +1, for betting, could have you betting slightly early, making your results slightly inferior to the TCs you normally bet at. The Floating advantage, could be partially to blame as well.
I agree it may have you betting too much too early based on your advantage, thats why the action/ hr rate jumps so significantly. Now, why do the results drop like a rock if I value the 5 as a +2 in the side count or even -1?

BJC
 
Last edited:

Nynefingers

Well-Known Member
#6
Can you post another chart including score? At a glance, it looks to me like the winrate has gone up, but not enough to compensate you for the increased variance. Of course, that doesn't mean there is no way to make this work, but it looks like the way you've got it simmed right now, you'd be better off simply playing with no side count and increasing your unit size.
 
#7
bjcount said:
The only gains, through all my test sims, were incurred when you left the counts balanced (normal tags) and setting the side count as a betting strategy, not a playing strategy w/ MP charts. If I understand the betting strategy correct it would work similarly to HO2's Ace side count, adjusting the RC based on the number of cards left, rich or depleted (side counted) prior to placing your bet, and then use the unadjusted RC for playing decisions.

(understanding here that HO2 is not ace reckoned, just using it as an example.)

BJC

edit: I haven't tried to develop a MP chart for this scenario yet.
OK I think I see what's happening. You are forcing the software to use an unbalanced count in true counted mode. When it sees these 5's come out the count is rising such that it is getting a positive count, even though an RPC player seeing the same thing would see a neutral count.

You can do that just fine, but what you have to do is set an IRC of 4* number of decks, and adjust your betting spread so that a neutral count is -4, just like you were using true counted KO or BRH-1. That's why you're getting so much action with the sidecount.
 

jack.jackson

Well-Known Member
#8
bjcount said:
I agree it may have you betting too much too early based on your advantage, thats why the action/ hr rate jumps so significantly. Now, why do the results drop like a rock if I value the 5 as a +2 in the side count or even -1?

BJC

First, can you ascertain, to AM's theory, whether or not, its simming it as unbalanced?

Bjcount, you could try 5.5/6 pen, and see if the margin in your WR, becomes even greater. That should tell us something.

AM, by the results of his sims, I dont see that as the case.
 
Last edited:

bjcount

Well-Known Member
#9
Nynefingers said:
Can you post another chart including score?
In response to some of the questions:

Nynefingers: It's my understanding that when you use a betting side count the accuracy of the SCORE value diminishes. For an example of this you can try to run a sim using HO2 in a DD game. First run it without the ace side count, then run it with an ace side count for betting. You will see the SCORE goes does significantly when you add the Ace side count even though you should use the ace side count when using HO2.

Automatc monkey: OK I think I see what's happening. You are forcing the software to use an unbalanced count in true counted mode. When it sees these 5's come out the count is rising such that it is getting a positive count, even though an RPC player seeing the same thing would see a neutral count.

You can do that just fine, but what you have to do is set an IRC of 4* number of decks, and adjust your betting spread so that a neutral count is -4, just like you were using true counted KO or BRH-1. That's why you're getting so much action with the sidecount.
Now I tried to set the IRC to -24 (6d x -4) but the win rate results drop again substantially, but so does the action. ~$1,800/hr w/~ $18.50 win rate/hr.

JJ: First, can you ascertain, to AM's theory, whether or not, its simming it as unbalanced?

Bjcount, you could try 5.5/6 pen, and see if the margin in your WR, becomes even greater. That should tell us something.

AM, by the results of his sims, I dont see that as the case.
I ran the sims as if it is a balanced count, using the original RPC tags. I just added a side count of 5 to the betting strategy placing a value on the 5's at +1. Why +1 you may ask, because it was the only value that increased the win rate.

If you look at the OP spreadsheet, as the pen hit 5/6d the win rate was lower with the 5 side count than without it. After running one at 5.5/6d the results dropped even lower, (-12%).

So the results still confirm that the shallow pen gave a higher win rate/ hour, with large increases in Action & SD.

BJC
 
Top