YupOriginally Posted by Secretariat
Secretariat said
Even at single unit minimum bets, the positive effect on EV is significant. Think about a $25 min game with simmed EV of $150 an hour per 100 handsIf I know I can win 1 or 2 extra hands and hour with optimal card play, then why not? It adds up.
1 additional min bet win adds 16% to EV.
2 additional min bets won adds 32% to EV
No it doesn't! You can't possibly calculate it this way or think about it this way. When you make one of your dazzling departures that only a Tarzan would know, you make it sound like you automatically change a losing play into a winning one. You don't! The gain in e.v. from making your "superior" play might be infinitesimal. Your winning or losing by so doing is probably close to a coin flip. So, please don't present the math the way you did. It isn't the right way to look at it.
Don
In conclusion, the highly respected Hi-Opt II system yields a 43,7/47,7/8,6 WLP ratio but despite the titanesque work accomplished by Don over the years in evaluating counting systems and despite Norm’s fabulous CV software, we don’t know what is the upper limit in Won/Lost/Pushed hands.
I suspect it is around 44,5/46,5/9/, but I don’t know. In fact, it’s clear now that nobody knows.
In other words, we have learned a lot about blackjack math in the last 40 years but we have not learned jack shit about the ultimate human performance in blackjack while using two or three side counts. I suspect Peter Griffin would be somewhat disappointed to see that.
Even Tarzan would not know his WLP at real tables (nobody can) but he would know his playing efficiency expectation if he played 200 000 to 500 000 hands with CV Software. Apparently the super sim done by Gronbog with the Tarzan counts did not reveal Won/Lost/Push ratios.
Clearly we need to learn more about human performance and maybe someone like ICountNTrack will say 20 years from now that SCORES (with only side count or none at all) was a “cute concept” back in 2021.
The suggestion is that you win more by losing less on crappy hands. Also, not necessarily following i18 etc. QTC also has something to do with it in positive counts.
Also, a difference in 1 crap hand per hour on EV - has a big effect on EV.
What I'm trying to explain to you is that the greatest count on the face of the earth (Tarzan?) can't outperform the average ones by more than about 10%-12%. So to say that, "Look at me! Look at the great play I just made; I turned a loser into a winner," is an amateurish way of looking at things. No one cares about what you did with such-and-such hand. All that matters is the SCORE. NOTHING ELSE MATTERS! You made a brilliant play and didn't insure when the TC was +10? Guess what? Dealer had blackjack anyway. Too bad for you.
Don
You're making no sense at all. It would be child's play to add a compiler to any sim such as Gronbog's, who could tell you without the slightest problem what the WLP split was. And then what? We're trying to explain to you that no one cares! If you'd like to know the answer as a curiosity, you're entitled to your request. But please don't write about it as if you're looking for the Holy Grail.
Eric Farmer has on his site a huge amount of information re perfect play. Time and again, we are disappointed by the results. The Tarzan project taught us that his outperformance over Hi-Opt II ASC was, in fact, rather small. Eric's work shows that "perfect play," especially in a 6-deck shoe game, is not substantially greater than Tarzan, but, of course, it outperforms.
The point is, the gold standard for the measurement of that outperformance has to be SCORE. Nothing else remotely makes sense. NOTHING. You want it now to be, instead, WLP? Dead end!!
Don
CVData already provides WLP data:
-Overall
-by true count
-by running count
-by hand depth
-by hand type (surrender, splits, hard dd, soft dd, insurance)
-floating advantage
-win and lose excluding and including ties
-by first two card vs. dealer card
-cumulative by RC and TC, increasing and decreasing
-by selected deck sections
-by units bet
-by dealer upcard
-by player first card
-and with CVDATA/ST, by zones
"I don't think outside the box; I think of what I can do with the box." - Henri Matisse
And, of course you can also go here: http://www.blackjacktheforum.com/res...p?do=statspage
Choose W/L percentage by true count, and at the very top, you get the overall results. Switch from one count to another to see that the ratios are extremely stable, no matter the count. Finally, consider the following: if I wanted to win the most hands possible, I'd never surrender a single hand in a game that offered me that option. How dumb would that be?
Don
That’s all I was asking for but no one gave a straight answer and instead this thread quickly got out of hand with authorities even suggesting sarcastically to never double down and to never surrender to improve WLP.
But you now what Don! I can play perfect insurance on the side, your top I18 factor and it doesn’t do jack shit to my WLP and I don’t care. Is the I18 the Holy Grail? No but it helps and the human brain can go way beyond that. My question is always: how far can it go?
After checking your references and CV Data it seems that no one has ever achieved 44% except maybe Tarzan but we don’t know.
I never meant to downgrade SCORES, Don. After all you were able to come up with SCORES for the 4-dimensional Tarzan approach. It’s the process to get top SCORES that is fascinating… as well as to get it’s author pumped up!
Posting Permissions
-*-
Reply With Quote
Bookmarks