Losing session and streak criticism sought (not rant)

[mickeymouse]

Unfortunately, I do not have good news to report. I lost another 2.5k at the tables today playing regular blackjack (6D, D9-11, DAS, no surrender, pen 90%). I was spreading $15 to 2x$125 - conservative by my standards, and playing heads-up.

I lost 2k in one shoe, on a stretch of 15 max bets of 2x$125 (i.e. thirty hands) in which I did not net-win a single hand. The true count was +6 or over during that stretch. It fell after this stretch, suggesting excess aces and tens WERE dealt out. I did not NET-WIN a single hand (out of 15) during this stretch. There were a few pushes, and some doubles/splits which all failed. Most of the time I got stiff hands or weak pat hands against a dealer ten or an ace. Insurance bets also failed each time.

Rather than ranting, I will analyse the numbers and ask the audience for constructive criticism.

I plugged my numbers into Blackjack Risk Manager and found the following results:

- The chance of losing $2500 within a 300-hand session was approximately 1.52 percent.
- So that is pretty bad.

Over my past 25 sessions, I am now $950 in the hole. This includes a stretch of -$4800 over my last This was a mixture of Australian Pontoon (which has a slightly lower house edge) and the aforementioned blackjack.

I have not changed or altered any of my betting patterns. I simply play according to the indices and bet with the count.

Concerningly, virtually all of my losses were incurred at the max bets. Prior to tonight, I actually had two very good sessions, although most of the money was made with the moderate bets. There were virtually no max bet opportunities then. I would say that when a max bet opportunity was available, I have been running at around -25%. Most impressively all my big split/double hands are failing.

I plugged my monthly numbers, and found that in fact, the chance of me having this result or worse was 22%. So while my numbers are bad, they are not outside the realms of possibility.

The problem is I am suspecting there's something systemically wrong with the shuffling that's causing me to lose all the big hands. During this month I have yet to win a hand with two max bets expanded into multiple splits and doubles. The dealer never busts these or always draws to beat me.

My own feeling is that I should continue to plug away, simply because the math has been shown to never fail in the long run. I am still overall ahead career-wise and I do not feel bored or lacking in motivation. But I will say that it would be nice for once to not always be on the left end of the Gaussian curve.

[davethebuilder]

Unfortunately, I do not have good news to report. I lost another 2.5k at the tables today playing regular blackjack (6D, D9-11, DAS, no surrender, pen 90%). I was spreading $15 to 2x$125 - conservative by my standards, and playing heads-up.

I lost 2k in one shoe, on a stretch of 15 max bets of 2x$125 (i.e. thirty hands) in which I did not net-win a single hand. The true count was +6 or over during that stretch. It fell after this stretch, suggesting excess aces and tens WERE dealt out. I did not NET-WIN a single hand (out of 15) during this stretch. There were a few pushes, and some doubles/splits which all failed. Most of the time I got stiff hands or weak pat hands against a dealer ten or an ace. Insurance bets also failed each time.

Rather than ranting, I will analyse the numbers and ask the audience for constructive criticism.

I plugged my numbers into Blackjack Risk Manager and found the following results:

- The chance of losing $2500 within a 300-hand session was approximately 1.52 percent.
- So that is pretty bad.

Over my past 25 sessions, I am now $950 in the hole. This includes a stretch of -$4800 over my last This was a mixture of Australian Pontoon (which has a slightly lower house edge) and the aforementioned blackjack.

I have not changed or altered any of my betting patterns. I simply play according to the indices and bet with the count.

Concerningly, virtually all of my losses were incurred at the max bets. Prior to tonight, I actually had two very good sessions, although most of the money was made with the moderate bets. There were virtually no max bet opportunities then. I would say that when a max bet opportunity was available, I have been running at around -25%. Most impressively all my big split/double hands are failing.

I plugged my monthly numbers, and found that in fact, the chance of me having this result or worse was 22%. So while my numbers are bad, they are not outside the realms of possibility.

The problem is I am suspecting there's something systemically wrong with the shuffling that's causing me to lose all the big hands. During this month I have yet to win a hand with two max bets expanded into multiple splits and doubles. The dealer never busts these or always draws to beat me.

My own feeling is that I should continue to plug away, simply because the math has been shown to never fail in the long run. I am still overall ahead career-wise and I do not feel bored or lacking in motivation. But I will say that it would be nice for once to not always be on the left end of the Gaussian curve.

Mr Mouse, you sound Australian and so do the Blackjack rules. By the way, where exactly are you getting 90% pen? Even so the limitations on soft doubling, no surrender and limits on re-splitting aces mean a house edge of >0.5% which is high. Unfortunately, with the standard deviation of this game + a bit of bad luck means your bankroll may get hammered for quite a while especially if you are playing alone. Most Australian casinos use ASM's for shuffling so there should be no issue there. Playing heads up against a dealer you will benefit more by vertical rather than horizontal spreading. I would suggest only spreading to two or more hands if you have an advantage in the game as the co-variance in Australian Blackjack is high and positive so this could be a leak in your game. Also, make sure the true count actually justifies the insurance bet. You did not mention which counting system you use so I am guessing it is Hi-Lo which offers the best balance between power and ease of use for most players. While the true count does accurately reflect the advantage it is not an accurate measure of deck composition so that may also be working against you.

These days Australian Blackjack does not present an attractive AP opportunity so maybe try elsewhere or team up with one or more partners to reduce the variance as long as they are competent and honest. Another suggestion is to stick to Australian Pontoon if you can find a shoe game as the house edge is about 1/3 less. Otherwise, if your game is solid then just take a break for a while.

[Three]

Prior to tonight, I actually had two very good sessions, although most of the money was made with the moderate bets.

I am still overall ahead career-wise and I do not feel bored or lacking in motivation.

It sounds like you have a bad case of selective memory. What is your n0 and how many rounds have you played? Mixing games and/or spreads can wreak havoc on your BR due to added variance. Do you use software to help set your attack plan? What type of shuffle do they use? You seem to be suggesting cheating. IF so you can adjust play for the same but I would just not play. We all have bad runs and level 1 and ace reckoned counts are more susceptible to swings due to a wider range of actual advantage for any TC around the average you use to set your bet for that TC. The same is true for many playing decisions in a full ace reckoned count. Ace compromise counts to a much lesser extent for the playing decisions. Are you wonging out of bad counts?

I am really not sure what to tell you. I no longer have these issues. When I did I studied how I could tighten up my game to mute swings. I tried many things and abandoned some to be replaced by others. After hitting some good runs that built my BR up, which may have been just luck more than anything else, I discovered I was betting very aggressively. That is fine for a BR large enough to take the swings but not one like what I had back then. It sounds like this is an issue for you. Remember optimal bets don't just take into account EV but also variance. With the few TC bins of interest a linear counter has to deal with I am not sure that this is much of an issue unless it runs through the entire system. I recently got info for my systems about 450 bins of interest from some software update to the simulator that factored in variance for each bin. I was just using advantage and guesstimating 1/4 kelly. I did a surprisingly good job at guesstimating those 450+ TC pair bins into 1/4 kelly betting matrix but 2 bins had wider bell curves around advantage so they needed to be bet in the next lower betting bin and 29 of the 450+ were moved to the next larger betting bin due to extremely tight bell curves around the advantage estimate. 31 changes is not bad for 450+ bins considering the software originally had no ability to set an optimal ramp and I was guesstimating how much more aggressive I could bet than the linear approach due to tighter bell curves but basing everything on straight advantage estimates. I was surprised that I was that close to perfect. The point is setting an optimal ramp is tricky and requires software.

One thing about top bet results is you make them so infrequently that it is hard to get a feel for the long term for these bets. Many people have a tendency to remember the bad over the good which given the infrequency of bets made can lead to quite a misconception of how those rare top bets are actually running. The other tendency is once you get an idea that something is happening you notice it even more when it occurs and tend to not notice the opposite as much. Sort of like a self fulfilling prophecy from a perception standpoint.

You have been complaining about your bad runs for a long time but I think you are still in positive territory overall. When I was betting a BR less than $20K I went through the same sort of things despite using advanced count techniques. It was a cycle of a long good run that gradually built my BR up and then a steep decline that took away most of my winnings. At the peak every time I felt like I now had things down and would be on easy street but then a steep decline would take that confidence away. Each time more research into how to do better would follow and tweaks to my counting approach as well as all the non-counting aspects of AP would follow. Eventually I must have hit the right mix. I think the big difference was the non-counting aspects but the counting aspects did make more games playable and mute swings some while increasing EV.

Knowing what games to avoid and how to get more positive counts while eliminating most really negative counts helped the most in my opinion. I digested one chapter in Don's BJA3 slowly and then made a trip to the casino's to put what I learned to work. I had already read the book but working the math examples and concentrating on a chapter at a time kept strengthen the non-counting aspects of the game. About halfway through the chapters (not done in sequential order) the downswings became much more rare and by the last chapter I was down to 1 severe downswing a year that lasted about a week and a month or 2 of spinning wheels with not much change in my BR. I was playing over 1000 hours a year for many years. I continually try to improve my approach to the game. I guess that is just what a perfectionist does. I feel I will never know everything and try to learn more all the time. I try to find creative ways to solve issues that nobody has published solutions to. I learn from other experienced AP's and any other sources of info. Networking is a very important tool.

You don't join the ranks of the successful counters that successfully grew a small BR to a large BR just by luck. Everyone that has done it will acknowledge in retrospect, after learning enough to be called a top pro, that luck plays a big part in not losing your BR but to some extent that driven work ethic to continually better your game makes your own luck. I think almost every player that fell in this rare group of successful counters that grew their BR from winnings worked hard to make their own luck by constantly learning more about the game and adjusting what they do in the casino to seek wins.

I hope something in here will help you find the winning ways. I am thinking that you may need help with knowing when not to play. Crowding, poor cuts and rules etc are things that you should avoid. One stepping stone toward success is being able to show up at a casino and not play a hand because conditions at the time make that the wise move. After you start doing that you learn when good conditions can be found and show up at those time. Often it is when everyone else is sleeping or when they are opening more tables. If you don't know the casinos procedures or the locals gambling rhythms you will have trouble finding the conditions that fit your style of attack.

[mickeymouse]

A few more details about my tales of woe:

- I don't play crowded tables. I will generally not play any table with more than two other players, and when there is a choice of tables, I will play the table with the fewest number of players. I don't mind playing with one other, as you can Wong in and out. Playing heads up you are basically forced to play every hand as they do not shuffle when you leave the table.

- Team play is next to impossible here. In my 160 sessions playing at this venue, I have met one other person who is skilled enough to play theoretically break-even. Every other player I have met is a losing player. You are allowed to back bet other players here and no ploppy has ever had the sense to back bet me.

- About 90% of my play is in Australian Pontoon and only 10% is in regular BJ. They only offer shoe games in BJ on Friday and Saturday nights. There is one shoe game in BJ on Wednesday and Thursday night only if a CSM is faulty (as happened last night). I played last night in BJ because the two Pontoon tables were crowded and I was able to play BJ alone.

- The house edge is 0.51% in BJ under our rules cf. 0.38% in Pontoon. This is the best you will ever get within a five hour flight.

- I will only play one box at the minimum unless there is a theoretical positive EV (i.e. the count is above +1). Above that I will play two boxes, with each box at roughly $20x(TC-1) with a maximum of $125 (BJ) or $150 (Pontoon). My software shows that my ROR will be unacceptably high if I attempt to play one box heads up with double the bets. Besides, the pit will shout "paying out blacks" if they ever have to pay out a black chip. I only play with greens. "Checks play" is not in the gaming lexicon here.

- Out of 160 career sessions I have won about 57% of them. Unfortunately this is not being translated into a good total win rate - my overall profit is about 25% of expected.

- In my previous 8 sessions, I have had three of my worst 4 losses of my career (160 sessions). I am down $9000 off my all time high achieved in mid-August. The vast majority of the losses occurred at a very high count where substantial EV was present.

- Shoe dealt games are manually shuffled, not with an automatic shuffler. Before the session opens all cards are laid out in sequence by the dealer, and then shuffling occurs when the first player appears to request a game. Shuffling seems to occur much quicker than manual shuffling I have seen in the United States. The entire shoe is riffle shuffled twice, then piles of ~ 2/3 deck are stacked on top of each other before the player cuts the card. There is no burn card.

- I wonder sometimes whether the manner of shuffling actually makes the game a negative EV game, even at very high TC. My results at low and moderate TC are no worse than theoretical expectation though.

[davethebuilder]

It sounds like you are going through a downswing but at least you are still ahead overall. If the cards are manually shuffled which includes riffling, stripping and turning then I cannot see how that would affect your EV in the long run but you may be playing with an insufficient bankroll especially if bigger bets on one box dramatically increases your ROR. Also, even if the TC is very high your advantage is still only around 4-5% at best so there is a lot of room for losing and with your playing strategy the variance and co-variance increase with the TC. I think I know which casino you are playing in but I am not familiar with their tolerance levels, however, I do share your frustration with other players actions especially in Pontoon which is not easy to play well. I am surprised that a Pontoon shoe game with 90% pen is still available??

[mickeymouse]

My ROR betting two boxes is acceptable (a little under 1% overall, and 8% chance of losing 10k) but if it were combined into one box my software insists that it would increase several times. I do not wish to draw heat betting near-limit hands because I do know of individuals who have been shown the door. Personally I am surprised I've lasted this long. I haven't received any countermeasures to date, though.

The highest I ever got to was an all-time high of 13.3k including comps which was still slightly below expected EV for the number of sessions played to date then. I am currently up only 4.2k including comps. My worst losing streak lasted 49 sessions in which I went from +8.2k to a nadir of -1.7k. I stayed in net negative territory for fourteen consecutive sessions and it took me only 24 sessions from my nadir to recover completely. Unfortunately, my success was shortlived, and as I mentioned three of my four worst sessions in my career have come in the last eight sessions, all of which were attributed to the >+6 TCs. I am beginning to be convinced that the real EV is negative at high counts.

[Three]

but if it were combined into one box my software insists that it would increase several times.

Your software is correct. Besides variants such as SP21 , pontoon etc are best played at 2 hands. BJ's aren't as common as they are in BJ and you 21 is an autowinner. Given some of the disadvantageous rules like the split ace restricted rules in pontoon are pretty tough I am surprised the HE is so small. I would question that.

[mickeymouse]

I double checked the late Katarina Walker's book and confirmed that her data is predicated on no draw to split aces. Apparently the automatic win on 21 with no hole card is worth a mammoth amount. This is borne out in observed play.

[Three]

Apparently the automatic win on 21 with no hole card is worth a mammoth amount. This is borne out in observed play.

i was going to post this but wasn't sure that I would be correct so I held my tongue.

[ZenMaster_Flash]

" ... three of my four worst sessions, all of which were attributed to the >+6 TCs."

The results of your BIG bets will determine your results overall.

Your small bets -- you will be lucky to be "even" at, or you may be in the red.

However your Big Bets are not that common, and as your "selective memory" kicks in, you will distort reality.

There is almost no difference in your chances of winning your upcoming hand. On average it is but a hair under 43%.

It will pretty much always be between 40% and 48%. You will always be pushing your chips out there hoping for a 20

(that is not a push) or a BJ (that is not a push) or a solid doubling hand. Therein lies your ONLY advantage.

[mickeymouse]

Yes. But when the big hands come, I don't get the BJs. I get stiff hands and weak 17s and 18s.
And if I get a doubling hand I draw deuces.
And if I get a 20, the dealer either gets a backdoor blackjack (with a picture up) or draws to 21 (if not).

[davethebuilder]

My ROR betting two boxes is acceptable (a little under 1% overall, and 8% chance of losing 10k) but if it were combined into one box my software insists that it would increase several times. I do not wish to draw heat betting near-limit hands because I do know of individuals who have been shown the door. Personally I am surprised I've lasted this long. I haven't received any countermeasures to date, though.

The highest I ever got to was an all-time high of 13.3k including comps which was still slightly below expected EV for the number of sessions played to date then. I am currently up only 4.2k including comps. My worst losing streak lasted 49 sessions in which I went from +8.2k to a nadir of -1.7k. I stayed in net negative territory for fourteen consecutive sessions and it took me only 24 sessions from my nadir to recover completely. Unfortunately, my success was shortlived, and as I mentioned three of my four worst sessions in my career have come in the last eight sessions, all of which were attributed to the >+6 TCs. I am beginning to be convinced that the real EV is negative at high counts.

It sounds like you are also pleasant to the staff which may partly explain your longevity regardless of your session result. I am also not convinced that many surveillance personnel know how to assess Pontoon advantage play which is partly why it is dealt from CSM's in most places.

If you are only playing simultaneous hands when you have an advantage then that is the correct play as long as your bets are sized accurately for your bankroll. You did not mention whether you use BS, Ill. 18 or full indices with the count as this will definitely affect your win rate. In my mind there is no logic to why your EV should be negative at high true counts...it might just be short term bad luck.

[davethebuilder]

Your software is correct. Besides variants such as SP21 , pontoon etc are best played at 2 hands. BJ's aren't as common as they are in BJ and you 21 is an autowinner. Given some of the disadvantageous rules like the split ace restricted rules in pontoon are pretty tough I am surprised the HE is so small. I would question that.

Australian Pontoon is played with a 48 card Spanish deck and there are no 10's. This creates a 2% disadvantage for the player but it is overcompensated by a series of other player friendly rules. The big advantage is that the dealer does not take a hole card and totals of 21 are paid prior to the dealer playing their hand. In general, the house edge is about 2/3 of Australian Blackjack.