Last Month there was a brief discussion of techniques for playing Shoes, from players who wanted alternatives to Wonging. At that time, I pointed out that " Play 70%, using 2 spots" was actually superior to conventional Wonging on 1 spot. I received many questions on this and related issues and I thought a more detailed explanation might be worthwhile. So in this post I will examine models for various Shoe-strategies and analyze their score.
I have made extensive use of BJRM in doing analysis, and all the data here is John's data from his SBA sims. I will focus 2 specific 6 deck games as the benchmarks: One will have S17-DAS and only 1 deck cut; the other will also permit Late Surrender, but will have a 1.5 deck cut. I will be using the HiLo System throughout.
Let me a couple of disclaimers at the outset. First, all the scores below will be "raw scores", giving results per 100 rounds. I will not attempt to adjust them for speed of play; you will have to make these adjustments based on your own conditions. Second, the methods used here do not properly consider card-eating effects. Mathematically, we are making the unrealistic assumption that when we sit out a hand, some ploppy comes in to play our spot, so that the same number of cards are dealt. Of course this doesn't happen, but this assumption makes our analysis feasible.
Wonging
Of course, the best way to play Shoes is to Back-count. We model back-counting by assuming the counter bets 0 (but consumes cards) when the TC <1. Back-counting on 1 spot gives respectable scores of $63 for 6/1 game and $59 for the 6/1.5-LS game. I computed these with BJRM using a 1-6 spread; however you really get essentially the same results if you use 1-8 or 1-10.
If we can play 2 spots when we Wong, we can achieve much higher scores: $88 and $85 respectively. This assumes that when we come in, we bet on 2 spots. In practice, this is not always feasible. Sometimes, we come in on 1 spot, and spread to 2 when the count increases. This is marginally worse, but the cost is not great: if we spread to 2 spots when the TC is 3 or above, we can score $82, $76. Here is a summary of Wonging
1 cut 1.5 cut LS Wong Play 1 spot only 63 59 Wong: Play 2 spots only 88 85 Wong: 2 spots when TC>=3 82 76Play 70% on 2 spots
Clearly Wonging into 2 spots is ideal, but there are often obstacles to implemented this strategy. First, there could be heat associated with this style of play. In addition, we may not have 2 contiguous spots available on the tables we Wong onto. Now I have been known to play 2 non-contiguous spots at a table; however this is not condoned by the pit, and I do not recommend its use. Also, sometimes there are few tables open. If you play during slow times, which you will give you nice uncrowded tables, you may find only a few tables open. The bottom line is that it is sometimes necessary to come in at the start of a Shoe, particularly if one wishes to play 2 spots.
So instead of Wonging, we do the next best thing: we sit out all hands where the TC is <= -1. This will be sitting out approximately 30% of the hands; hence I call it the "Play 70%" strategy. This 70% is only an approximation: with 1 deck cut, we actually play 70.9%; with 1.5 cut we play 72.3%
You may feel a bit squeamish about sitting out this many hands, and for that reason may wish to use a modest spread. With just a modest 1-8 spread we obtain scores of $67 and $64 respectively. If we only risk a 1-6 spread, we can still obtain 60 and 57.
These calculations assume that we play 2 spots off the top and all positive counts. In many casinos, this means that are smallest bet will be 4 times the table minimum. This means that a Green Chip player would be limited to $5 tables. An alternative is to play 1 spot of the top, and spread to 2 spots when the TC is 2 or above. This is not as good as always playing 2, but we will still obtain good scores of $64 and $60 respectively. Here is the summary of this strategy
1 cut 1.5 cut LS 1-8, 2 spots 67 64 1-6, 2 spots 60 57 1-8, 2 spots when TC>=2 64 60
Note that with the 1-8 spread, this method beats conventional 1-spot Wonging. This perhaps the key lesson of this post: if you choice is between back-counting or playing 2 spots, play 2 spots, and sit out the negatives. In practice, you may implement this by taking a break or leaving the table when the TC is <=-1.
Play 80%,90% on 2 spots
By now, you are wondering how can possibly justify all this table-hoping. No-one's bladder is that active! is ! You may feel that getting out of the 30% of the rounds is just not practical. OK, then try to sit out the hand with TC<=-2. I will call this play 80%, although this only a rough approximation. It is more like 82.2% and 84.2% with 1 and 1.5 decks cuts, respectively. This will require you to play a lot of negative hands, so you will need to spread more. With a 1-10 spread, this strategy scores 63.5 and 60 in our benchmarks games. Note that this still better than conventional Wonging.
If the table minimum does not allow you to bet 2 spots in the negative situations, you can do almost as well if spread to 2 spots at a TC of 2. The scores will be 62.5 and 61.5
If your afraid to drop out that often, at least try to stay out when the TC<=-3. I call this "Play 90%", although it is actually 90.4 and 90.4%. Unfortunately, you are playing a lot of negative rounds here, and you will need a 1-12 spread to obtain. Even then you will only score 62 and 58. Summary:
1 cut 1.5 cut LS 1-10, 2 spots 80% 63.5 60 1-10, 80% 2 spots @ TC>=2 62.5 61.5 1-12, 2 spots 90% 62 58
20-1 Partnership
Another common strategy is to bet 2 spots off the top, and drop to 1 spot when the count is negative. I will consider dropping to 1 spot when the TC<=-1. In other words, we are playing 2 spots 70%, and 1 spot 30% of the time. If our max bet is 2 spots of 10 units each, then our min bet is 1 unit. Thus, this is a 20-1 spread.
This is a popular technique for partnership play. Two players each play 1 spot and take turns sitting out. Each one only sits out 15% of the rounds, which is not unusual. It can be implemented with 1 counter and 1 BS player; the counter signals to the other partner to indicate the bet and also to indicate playing deviations. Strategies like this are often used with husband and wife teams.
Unfortunately, playing even 5% during all those negatives has a cost, and our Score is only $55.5 and $53. However, you can improve this by dropping both hands for the 10% of the rounds where the TC<=-3. These scores become $62 and $59.
1 cut 1.5 cut LS 1-2x10, play 2 70%, 1 30% 55.5 54 1-2x10, play 2 70%, 1 20% 62 60
Play All on 1 Spot
You may wonder what would happen if you could only play only 1 spot, and had to "play all". We will have to use a 12-1 spread. The pitiful results are shown below. Now even here, sitting out hands at -2 or -3 will improve your score and so I include the results of those.
1 cut 1.5 cut LS Play All 1 spot 12-1 33.5 33 Play 90%, 1 spot 12-1 42.5 40 Play 80%, 1 spot 46.5 44
Play-All should really be dead as a way of thinking about Shoe Play. I mean, you have to go to the bathroom sometime! Just by timing these to coincide with true count of -3 raises your score about 25%. You can raise it a little more if you get out when the TC is -2.
Conclusions
All of these strategies represent abstract models, which will be imperfectly implemented in the real world. However, there are two key principles to govern Shoe play:
- Avoid playing negative counts. The more negative the count, the greater the need to sit out.
- Play 2 spots as much as possible, especially when the count is positive. The more positive the count, the greater the need to play 2 spots.
And with these 2 strategies, you can get by relatively modest spreads. In the SCORE article, Don pointed out that with Wonging, you can get near optimal results with a relatively small spread. This also holds for our "play 2-spot, sit out a negative model".
In practice, try to come into a Shoe on a positive count. If it is marginal, delay until it gets strong. If it doesn't, seek another table. By coming in only on good plus counts, you will be able to play the entire first Shoe. If the count permits it, and there is an available spot, try to spread to two spots. Try to hold that spot during the Shuffle break. If the next Shoe become negative, look for creative ways to sit out. You can try to get lammers for your spots while you take a break, or you can try to find an excuse to visit another table.
As noted above, often the table minimum will prevent you from playing 2 spots in neutral counts. You will therefore have to spread to 2 spots if the count increases, and cut back if the count decreases. Now you should realize that this may draw heat, not from the pit, but from the other players. There may be a high-rolling ploppy who complains that you have "fouled up the cards". Try to avoid this. You don't want some high-roller complaining to the pitboss "That counter over there spread to 2 hands, and messed up my cards."
Try to find excuses for changing spots. If the count goes south and the dealer gets a BJ, suggest dropping a spot to "change the luck of the cards". Indicate your willingness to give up some play to help the other players. If the table is going well and a relief dealer comes, then "Ah-oh. They are going to change our luck. Your not going to burn a card, are you?" And then when the relief does the burn the card, drop a spot. Of course, if the count had headed North, you would open another box ... If another player wants to join the table and the count is negative, invite them in. tell them you'll drop a spot, so it will keep the balance of cards right. etc ....
Know you casino. Some casinos are uptight about table hoping, others are uptight are big bet variations. I recall playing a session where the dealer was telling us about the awful counters who come to that casino: "They jump in and out of games and mess up the cards for everyone." I had been quietly spreading 20 to 1 at her table without notice. On the other hand, I have been in casinos where there is a lot of table hoping. Play there with a reduced spread, and you'll fit right in.
In closing, let me emphasize the importance of playing in casinos which are not crowded. I think this is a point which is not emphasized enough in much of the BJ literature. I think experienced players have an intuitive grasp of the damage done by crowds. But texts which are read by beginners do not emphasize this enough. Now most of the literature points out that crowds conditions will slow the game down, and that alone brings down your SCORE. But in addition to this effect, crowded casinos preclude the use of the tactics discussed above.
If the only time a spot opens is at the shuffle, and then only 1 spot at a table, and if every time you sit out there are three people complaining that want your spot ... we'll you simply can't play Shoe games under these conditions. Conversely, if there are many tables with only one or two players, and you can move from table to table and play 2 spots at those tables, then you are in Shoe-Heaven.
Crowds, or the lack thereof, are very bit as important as penetration. Given a choice between a crowded casino which cuts deep and an un-crowded "shoe-heaven" with more shallow cuts, I would tend to chose the latter.
December is always my favorite month. Most gamblers are too busy with holiday activities, and casinos are nearly empty. Conditions are often favorable for implementing the strategies discussed above.