Head to Head Blackjack win ratio/percentage ?

[PlayHunter]

I really need your expert advice on this: - What is the win ratio/percentage in a player versus player (head to head) format for an expert blackjack player playing against a medium skilled (see below for what I call a medium skilled level) opponent ?

To give a picture/example of what exactly I want to ask, I will describe/compare with a Backgammon win ratio example:

- In Backgammon if you are a perfect player (for example a backgammon bot like GNUBG or ExtremeGammon) or near perfect player (a world champion level) and you are playing against a medium skilled opponent (this would be not a novice or beginner, but an experienced but casual player - a player which plays as s/he see/feel fit, as good as he can approximate with the eye but without really proper knowledge like reading books or studying positions about the game and its fine details about percentages etc) the world champion player will win about 74%-75% of the time.

But if you are not a a world champion level, but just an expert player (which is just one level down from being at world champion level - this would be a player with loads of experience and knowledge of lots rules of thumbs, books reading, percentages and positions as well as opening moves and reply`s well stored in his brain - but still making some blunders in the long run by not estimating the correct move every time) and you play against a medium skilled opponent, then you will win about 68%-70% of the time.

- Why do I need this similar information in regards to head to head blackjack ?

I want to find out what is the profit I am about to make from playing head to head blackjack in the long run expressed in earn per hour. - How I approach it, is to watch other players playing, seek medium or even less skilled players and then play with them as much as I can. I am looking mainly for two mistakes in their gameplay when choosing someone to play against: 1. bad or inappropriate bets on the last hand, and 2. bad or inappropriate bets in the course of the game. So, I make sure I have an edge over my opponents ! - Just how big is this edge ? - I do not know but I want to find out, and that is why this thread is here..

If I look at my last 1000 head to head blackjack games (I kept track of all of them), I have only won 57% of the time, and more than 90% of these games were all played with medium or even lower skilled opponents ! - I know that all of my last hand bets was optimal, and also the same was most (certainly not all!) of my hit/split/double decisions on the last hand. I also believe that most of my bets during the games were also optimal ones.

- The only two things which I am not so sure about is my (optimal?) play on the 9th hand which I believe is very important in the game balance (from here all my threads about next to last hand - I am trying to build rules of thumbs regarding the next to last hand for my play) and some other (less important I would say) playing decisions among other hands during a game.

Now, considering that a 10 hand game takes about ~8 minutes, and that most of my stakes are in the $10-$20 area (but most in the $10) and also adding the searching time approximately for finding a worthy opponent to play against, and also withdrawing the website fees for offering the games, I am left with a profit of ~$4 (maybe even a bit less) per ~1.5 hours which I feel is way too low to be worthy. This ($10-$20$) is the range where you can find loads of medium skilled players, so I can not increase the stakes to bust my profit. At lower stakes it might be possible to find even lower skilled players but due to the time it takes to finish a game, it would not be worthy at all, and I can not accurately do multi-table games. (two games or more at once)

- So how I can boost my profit ? (I think the answer is in improving the win percentage from 57% to something over 60% but .. HOW to do it ?)

As a comparison, a backgammon game takes about 5-6 minutes but very hard you will find medium or lower skilled players for a stake higher than $5.

 

[gronbog]

Intrigued

I'm intrigued by this head to head format for a couple of reasons.

1. First, it would seem like a great opportunity for skilled players. The more players at your table, the more difficult it is to make strategic bets early on because a good bet vs one opponent is likely to be a bad bet against another. In a head to head format, one could make strategic bets on almost every hand to keep oneself in the game.
2. Second, the head to head format makes the games go quickly, as you have noted, making a small edge very valuable via repetition.

A couple of other points

• I would like to point out that winning 57% of the time is a huge advantage. Assuming that the winner takes all, minus the rake, it represents a (0.14 x stake)-rake expected value. At the $10 level, that would be a 0.14x$10 = $1.40 EV minus the rake per game which, at 8 minutes a game (~8 games per hour) should come close to $11.20 per hour minus the rake. If you could build a list of suitable opponents, you should be able to cut down on your time searching for opponents and come close to realizing this potential.
• You seem hesitant to play in higher stakes games due to the slightly higher level of competition, but it could still be a more profitable strategy. For example, if you double the stake and lose less than 1/2 of your edge, then it is still a winning move.
• Even if you can't squeeze out a decent hourly profit, it still sounds like great practice for an almost guaranteed, if small, profit.

I am interested in which site is offering these games. If you're worried about burning out the game or attracting too many skilled players, them feel free to PM me. The reasons are twofold:

1. I myself would like to try playing at the site for the practice value.
2. Once I see the exact rules and the skill level of the opponents at each stake level, I can use my tournament simulator to estimate the optimal aggression level at each stake level which, in turn, will allow me to run further simulations to estimate the win/lose/EV numbers you have asked for.

 

[PlayHunter]

A couple of other points

• I would like to point out that winning 57% of the time is a huge advantage. Assuming that the winner takes all, minus the rake, it represents a (0.14 x stake)-rake expected value. At the $10 level, that would be a 0.14x$10 = $1.40 EV minus the rake per game which, at 8 minutes a game (~8 games per hour) should come close to $11.20 per hour minus the rake. If you could build a list of suitable opponents, you should be able to cut down on your time searching for opponents and come close to realizing this potential.
• You seem hesitant to play in higher stakes games due to the slightly higher level of competition, but it could still be a more profitable strategy. For example, if you double the stake and lose less than 1/2 of your edge, then it is still a winning move.
• Even if you can't squeeze out a decent hourly profit, it still sounds like great practice for an almost guaranteed, if small, profit.

PM sent, and anyone who is interested can also PM me about it. And I would like to address to few of your points which also can be of interest for readers:

1. The rake is 10% and is paid each game no matter if you win or lose. Each player do buy in with $11 and the pot is $20. The winner takes the pot.

1`. I used to do a list with accessible opponents, and that list was very long increasing daily, but the real problem is that these accessible players come in, play for a few 2-3 days (most of them only one day), then leave and never come back. The few ones that remain on the site to play more than that, usually become much better players in quite short time (this process does not take longer than two weeks usually) and are no longer so accessible as they was in the beginning.

2. I am not hesitant to play higher stakes at all, but the problem is that the crowd is in the $3-$20 area. Higher than that, very rarely I find anyone so worthy.

There is also a head to head format where you can play for $5 and you have to play only 5 hands. I have considered this as it usually takes about half the time to finish a $10 games where you have to play 10 hands, and the is also loads of accessible opponents, but the problem is that I found out to have a smaller win rate at this format and I believe this is happening because a player bust out of chips much more frequently in a 10 hand format than he does in a 5 hand format.

3. Yes, certainly it is great practice indeed. I started to play this game because I found it very addictive, entertaining and pleasant game to play. Later on, I have realized that I am even making a tiny profit out of it which was also very nice to me. Now, I want to learn as much as I can to improve my win ratio ..

 

[London Colin]

1. The rake is 10%

Whereas the rake at the other site which offers similar, head-to-head games is approximately 5%, plus if you are happy that you have an edge over the robot player then you can play constantly, rather that wait for opponents.

My records from both sites are a little imperfect, but my feeling is that the better rules at the 10% site give only a marginally better EV, when you factor in the bigger rake. Combine that with the ability to play much faster against the bots, and I would think the other site ought to offer a substantially better hourly win rate.

(I don't really see much point in keeping the sites' identities secret, as we have talked of them openly in the past, but I can't bring myself to be the one to break the silence.)

 

[PlayHunter]

Whereas the rake at the other site which offers similar, head-to-head games is approximately 5%, plus if you are happy that you have an edge over the robot player then you can play constantly, rather that wait for opponents.

My records from both sites are a little imperfect, but my feeling is that the better rules at the 10% site give only a marginally better EV, when you factor in the bigger rake. Combine that with the ability to play much faster against the bots, and I would think the other site ought to offer a substantially better hourly win rate.

(I don't really see much point in keeping the sites' identities secret, as we have talked of them openly in the past, but I can't bring myself to be the one to break the silence.)

Well .. to be honest, the robot player (probot) beats me ! I did played 1000 games against it and I won only 489

(I do remember our discussion, and I know you can beat it just about 53% but it seems I am a lot left to improve ..)

I don`t necessarily want to keep the website name secret, about this thing I was talking with Mr. Grongbog also, told him that me and you was telling the name of the networks where these games are running in many of our posts, and that I think the ones who do care about these games are already there .. probably.

 

[gronbog]

I see now why there's no need for secrecy. At the time I asked, it was not clear to me that the sites offering the head to head format were the same ones you had both been mentioning.