Card Counting in Baccarat

It's hard to find things to write about in a time when many casinos are closed and those that are open are not trying anything new.

However, one topic I've had on my list of ideas is card counting in baccarat. It can be used to cut down the house edge, but very very rarely will it overcome it. That said, please visit my new page on Card Counting in Baccarat. It shows how to easily cut down the house edge of 0.93%.

I'm sure someone will say to just play blackjack, which I agree with. However, this at least addresses the topic in more depth than I've ever seen done before.

I welcome all comments on my article.

The question for the poll is what is your baccarat strategy?

Heh. I get it. You've been asked about this on the LiveStream so many times that you can now simply tell people to go to your website.

Mike, first off let me say thanks! Your website resources on Baccarat are totally outstanding. They really demonstrate, for those who are willing to listen, that the mathematics of Baccarat are well understood and that there’s no magic bullet to the game.

I count cards in Baccarat using Eliot Jacobson’s system (10,9 = 0, 1,2,3 =-1, 4 =-2, 5,7,8=1 and 6=2). I bet on Banker until the true count hits about +2 or so, then I'll switch to Player. If the true count is in the range of -2 to +3, I'll usually bet only half my normal wager. If the count gets anywhere from about -6 to +6, which usually happens near the end of the shoe, I’ll double my normal wager.

As you’ve pointed out before, there is NO WAY to practically use a counting system to turn Baccarat into a +EV game. The best one can hope for is to lessen the house edge a little bit. I count in order to do that, and to make the game a little more interesting.

Your system is really nice. I very much like how your Level 2 strategy uses a running count coupled with the progression on the Bead Plate to somewhat approximate the true count. Very ingenious.

Note, in the Level 2 section I was unable to see the image of the Bead Plate with the player in the sixth column (it may just have been my browser).

Quote: Gialmere

Heh. I get it. You've been asked about this on the LiveStream so many times that you can now simply tell people to go to your website.

Issue

In the Level 2 strategy section you reference an image of the Bead Plate, but no image is actually presented.

Thanks. I had some technical issues with that. Here it is.

Quote: UP84

Mike, first off let me say thanks! Your website resources on Baccarat are totally outstanding. They really demonstrate, for those who are willing to listen, that the mathematics of Baccarat are well understood and that there’s no magic bullet to the game.

Thank you! And you're welcome.

Quote:

I count cards in Baccarat using Eliot Jacobson’s system (10,9 = 0, 1,2,3 =-1, 4 =-2, 5,7,8=1 and 6=2). I bet on Banker until the true count hits about +2 or so, then I'll switch to Player. If the true count is in the range of -2 to +3, I'll usually bet only half my normal wager. If the count gets anywhere from about -6 to +6, which usually happens near the end of the shoe, I’ll double my normal wager.

Thanks for mentioning this. I didn't know Eliot wrote about this. There is a strong effect of the 4 and the 6, so maybe I'll analyze Eliot's system next.

Remember the EV of the best bet isn't the worst at zero. In the case of my count, it's at -4.

Quote: Wizard

It's hard to find things to write about in a time when many casinos are closed and those that are open are not trying anything new.

However, one topic I've had on my list of ideas is card counting in baccarat. It can be used to cut down the house edge, but very very rarely will it overcome it. That said, please visit my new page on Card Counting in Baccarat. It shows how to easily cut down the house edge of 0.93%.

I'm sure someone will say to just play blackjack, which I agree with. However, this at least addresses the topic in more depth than I've ever seen done before.

I welcome all comments on my article.

The question for the poll is what is your baccarat strategy?

My strategy for online game, use ev calculator to find ev after each hand . . . . .and yet still NOT worth to play it, unless you have other promo/rebate . . . .

Wizard,

Can you address card counting for the tie bet?

I never tried it but here is my thoughts.

In an 8 deck shoe there are 416 cards total.

128 cards are zero value cards (10, J,Q,K)

So what if you counted to determine when there is a "zero value" rich deck and bet tie.

The idea being a deck heavy in zero value cards will result in tie more often at a nice 8:1 payout

Quote: darkoz

Wizard,

Can you address card counting for the tie bet?

I have looked at that. If must have +EV, that is where you'll most often find it. However, only rarely towards the end of the shoe. The vast majority of the time, the house edge is a tall mountain to overcome.

However, maybe I'll come up with something. Baccarat is slow enough somebody could keep a side Tie count.

Newbie to Wizard forum, started today with Jacks or Better. Two things make me very curious. The first, even though most of the ANALYZE columns are self explanatory if one is familiar with game. For instance I know what the EV value represents, but I am interpreting the LOSE column as representing the number of time selection lost within the overall sample size. Second thing, why is a given HOLD selection have a "Not the Best Play" warning when preferred HOLD selection may have a higher EV value, the LOSE column may show the preferred HOLD selection has an order of magnitude more losses than what is typically advised for. Case in point has suited A, Q, J and 3. This hold selection shows the LOSE column of 39. The preferred HOLD consists of the A, Q and J even though the LOSE column shows losses of over 1,000. I may be recalling the value not exactly, but it is substantial compared to the four-to-a-flush which is my default HOLD selection other than a high pair.

I know I'm missing something crucial, would you set me straight? I haven't been to Vegas in a while and I'm itching to get back in the game. And there are a few Hard Rock casinos in Florida.

Thanks for you time
FlaTrips

Quote: FlaTrips

Newbie to Wizard forum, ....

Welcome to the forum.

Please post your question in the video poker section. I would suggest making a new thread for it.

Thank you.

I just added a level 3 to my card counting baccarat page. It widens the adjustment on the running count by rank from -1 to +1 to -6 to +6. Bottom line is a reduction in house edge of only 0.01%. Not worth the bother, but I felt I had to see how much it would help.

I saw this was already mentioned, but I have also been curious about the frequency of a shoe having enough 10 cards to have an edge with the tie bet (hitting 0-0). Unlike blackjack one good thing is it’s not as awkward to skip a lot of hands. You could even stand behind and back count the whole time since casinos aren’t really suspicious about bac being beaten. You’re just ‘looking for patterns.’ I would think the problem is the speed of the game is so slow it would take forever to see a shoe rich enough for 0-0 ties to hit frequently enough to be worth the effort.

Quote: camz1969

I saw this was already mentioned, but I have also been curious about the frequency of a shoe having enough 10 cards to have an edge with the tie bet (hitting 0-0). Unlike blackjack one good thing is it’s not as awkward to skip a lot of hands. You could even stand behind and back count the whole time since casinos aren’t really suspicious about bac being beaten. You’re just ‘looking for patterns.’ I would think the problem is the speed of the game is so slow it would take forever to see a shoe rich enough for 0-0 ties to hit frequently enough to be worth the effort.

That is addressed in my baccarat appendix 2.

Outstanding job, thank you!

Being an admirer of simplicity, I can't help but remind that 100 Bit Dice by 4ThePlayer (so-called Bitcoin Dice for regulated casinos) has 99% - for 50:50 paying 1.98x or under 49.50 paying 2.00x. That leaves only 0.05% above what's granted without thinking at all. Too little to think. :-)

May I also remind that two casinos in central London, Empire and Hippodrome, offer "Red 8" rule on Baccarat, where any Red 8 in Banker's hand pays 1:1. They still have it on at least one table. That's been mentioned back in 2012:
https://wizardofvegas.com/forum/questions-and-answers/math/7660-baccarat-variation-offered-in-the-uk/2/
Charlie calculated the overall RTP to be 99.19% through not yet authoritatively. Maybe it's a good time to add the "Red 8" rule to your appendix. That, of course, consumes all the effort in the standard game.

Still above "Red 8" is 4% commission in Isleta Resort & Casino in Albuquerque and online at 5Dimes Casino. The latter lowers the commission on Banker to 2.75% (4 pays 3.89) every Monday from 2pm to 8pm ET (6h/week) which is, essentially, a coin flip.

That's where I tried to go.
Thanks

Hi Wiz. I read through it but isn’t that whole page based on the one strategy? In the effect on edge chart you’re saying removal of 0s (10s right?) is good for the tie since it’s positive. Am I reading that right? What I was referring to is counting down shoes until you find one with almost all 10s left where 0-0 tie has a high enough probability of hitting to overcome the edge. For instance in a super rare occurrence there might be a shoe with all 10s left where you would have a 100% edge. With a few non 10 cards I know you would still have an edge but I’m not sure what exact 10/non10 ratio would give you an edge or how frequently that ratio would happen.

Hi camz1969,

you should find the links below helpful/interesting:
link , link2, link3

Thanks. Link3 hit on the all 10s left...once in every 370,370 shoes if cut card is at 14 cards and to correct what I posted earlier it is a 800% edge. That simulation was for ALL 10s though. I still wonder what 10/non10 ratio actually give you an edge. If ALL 10s is 800% edge what ratio would you need for just a 1% edge? Like if I assign +1 to ALL non 10 cards what count would give a positive expectation? This would be a different frequency (worse maybe?) than these other counts. Disclaimer: I’m not trying to go be a bac player, just curious and Wiz is looking for discussion.

Quote: camz1969

...That simulation was for ALL 10s though. I still wonder what 10/non10 ratio actually give you an edge. If ALL 10s is 800% edge what ratio would you need for just a 1% edge? Like if I assign +1 to ALL non 10 cards what count would give a positive expectation?

The problem with this type of ratio (10s/non-10s), is that at most values it's an incredibly inefficient measure of the probability of a tie. This is because some non-10 cards (A,8 & 9s) push the probability of a tie one way, while other non-10 cards push the probability of a tie the other way (the 6 and 7 pushing it very strongly the other way). For example, given the ultra rare ratio of 30 to 1, the probability of a tie is most likely enough to beat the house edge (I haven't done the math on this), but at lower and more common values, the ratio will be practically meaningless.

Right so to clarify, the count I’m talking about is basically just chasing a 0-0 tie only. I mean the simple answer is you have an edge when the 10s ratio makes you hit a 0-0 better than once every 8 bets with 8:1 payout. I’m not sure of the frequency of that happening though. The Wiz’s count is factoring in the other cards for all ties and it appears with that count a ton of 10s in the shoe is a bad thing. We’ve determined from the one article to get the full 800% edge (with 100% win rate) from all 10s left takes 320,320 shoes on average, but how frequent would you see enough 10s to hit a 0-0 even 1/8 times in the long run (0% edge)? To keep it simple assume back counting until that point so you’re not offsetting -EV.

Thought the count was A,2,3 = +1 4 = +2, 5,7,8 = -1, 6 = -2 and 0 and 9 = 0

I have a question about the main baccarat counting page on Wizard of Odds website. It states that "Looking at this from the perspective of the Player bet, a deck rich in high cards is good. That is why it is bad for the Player bet when they leave the deck. The flip side of that coin is a deck rich in low cards is good for the Banker bet. Thus, if a low card leaves the shoe, that improves the odds on the Player and if a high card leaves, it's good for the Banker."

Since we are betting on player on running counts of -4 or greater, and the cards with a negative count attached are 5 through 8, isn't this backwards? So a deck rich in high cards is good for banker, and low cards is good for player? Or am I missing something?

Thanks

Quote: Bjkdlr

I have a question about the main baccarat counting page on Wizard of Odds website. It states that "Looking at this from the perspective of the Player bet, a deck rich in high cards is good. That is why it is bad for the Player bet when they leave the deck. The flip side of that coin is a deck rich in low cards is good for the Banker bet. Thus, if a low card leaves the shoe, that improves the odds on the Player and if a high card leaves, it's good for the Banker."

Since we are betting on player on running counts of -4 or greater, and the cards with a negative count attached are 5 through 8, isn't this backwards? So a deck rich in high cards is good for banker, and low cards is good for player? Or am I missing something?

Thanks

A deck rich in low cards(Ace, 2,3,4) is good for PLAYER