Card Counting! Or is it just Advantage Play now? After all you’re doing a lot more than just counting cards in order to mathematically have an edge over the casino. In the previous two articles I’ve discussed the basics of card counting, some advanced techniques/plays, and the ugly part of card counting… what to do when the casino catches on. As I’ve mentioned a couple times already, there are so many more aspects and always something to learn when it comes to counting cards. After nearly a decade, yes, I’ve got it down, but no, I don’t claim know it all. I’m learning new things each and every year. I could write so much more, perhaps a book, on the topic. Unfortunately, it would be mostly rehashing what others have said (maybe that’s how you feel now) but in the end what do you do when you’ve amassed this much knowledge about any one particular topic? You share it, and that’s what I enjoy doing. Not to build some army of card counters, or to burn the casinos down, but simply to do my part in exchanging information that helps each and every one of us grow and become wiser.
So without further ado, I present to you article 3:
In this context I’m taking this question to mean “Is counting still feasible?” and “Are there still opportunities to do so today?” The answer to both of these is yes. Almost every casino you go in has a blackjack game that can be beat. The rules are certainly deteriorating, but whether it’s a .40% house edge or a .65% house edge just means you raise your bet at TC +1 or TC +2. It does affect your hourly EV, but there are still plenty viable options. After all, $20 or $30 per hour sure beats a minimum wage job. Here are two examples with a 1-15 spread:
.40% House Edge Hourly EV: $20.24
.65% House Edge Hourly EV: $15.48
Here are two more betting quarters at the same game with a 1-10 spread:
.40% House Edge Hourly EV: $31.29
.65% House Edge Hourly EV: $20.05
The other part about being viable, in my opinion, is can you still get away with doing it in casinos or have they been too educated by now? You can absolutely still do this in almost any casino. You’ll be surprised just how ignorant most casinos are when it comes to card counting. I’ve had pit bosses, when having general conversation with them, say things like “yeah no one can count an 8 deck shoe, it’s impossible!” This is just a clear misunderstanding of what card counting is, yet that person is the one responsible for deciding to let you play or not.
Card counting is without a doubt still a viable opportunity. There are new casinos popping up in every state, many beatable games still available, and a plethora of uneducated casino personnel. Not only this, but it is my opinion that if you have a good personality, and use good cover, it should be very difficult for a dealer or pit person to figure out your counting. The only way you could get made is by playing for too long at one place and a security person watching video of the tables figuring out your spread/play. This can again be avoided with short sessions and moving tables after exposing your max bets.
Card counting, is a viable option.
Card counting is a viable option, that’s all good and well, but is it really worth it? Could you make a living doing it? Are you going to make life changing money from blackjack? Well, maybe. As I stated in my first article unless you are a professional you will probably not make any substantial life changing amount of money from blackjack. So it greatly depends on whether or not you’re a professional and what your definition of “worth it” is.
To a non professional, such as myself, I find counting cards to be worth it if it provides me with the opportunity to take vacations/trips essentially for free, and if I can make some decent extra money on the side. Given that my blackjack game/comps have fully paid for my last 7/8 Vegas trips, and that I’ve made $5,000-$7,500 per year (on average from blackjack play) I absolutely consider it worth it. It’s one of my many hobbies that enables me to travel to different places, meet different and interesting people, and have new life experiences to learn from. For me, counting cards is definitely still worth it.
So could you make a career out of counting cards? Why don’t we have a look at the math? Let’s assume the following averages for a professional card counter at 6 deck games:
House Edge: .30%
Betting Level: $100 games
Bet Spread: 4-1
Average Bet: $162.40
Hours Per Week: 30
Hands Per Hour: 100
In this scenario a professional player, from blackjack play alone, could expect to make $62.96 per hour, with a total of 1560 hours per year, yielding a total expected profit of $98,217.60 per year. Do note, this doesn’t include any comps/etc, and is pre-tax. This also allows for 30 hours of player per week but if you’re a professional you’re putting in another 10-20 hours per week in scouting, travel, etc.
Some might say you could never play that much at the black level to get away with it for a long time (many years). Then let’s also look at some math for the quarter tables (6 deck games):
House Edge: .4%
Betting Level: $25 games
Bet Spread: 12-1
Average Bet: $78.46
Hours Per Week: 30
Hands Per Hour: 100
In this scenario a professional player, from blackjack play alone could expect to make $42.46 per hour, with a total of 1560 hours per year, yielding a total expected profit of $66,237.60. Do note, this doesn’t include any comps/etc, and is pre-tax. Again, I want to note a professional would spend another 10-20 hours per week scouting and traveling.
*A note for the hourly EV’s. I set the spreads up to ramp how I would personally do it given my experience. I did not cheat and simply max bet once an advantage favorite count was met to just simply jack the hourly EV up. These are real numbers and real scenarios which are certainly viable.
More examples of this would be some professional forums members (kewlj in particular) that provide their yearly blackjack statistics (in which the average results of the past few years came about in the middle of my two example scenarios above). Given the above information, I conclude that not only is card counting a viable option, but it is most certainly still “worth it” as well.
The following are questions I’ve received on the material in the first two articles. They are paired with my responses. It is my hope that if one person asked the question, there must have been others pondering the same thought.
Q: So I’ve seen all over the place on here that you need a minimum bet spread of like 1-10 on shoe games and smaller on single/double deck games in order to “beat” the game. Would one be losing money if you were spreading 1-5 like you said?
Also in your betting ramp break down, I know you spread to two hands at +2, but just one hand the progression would go 10, 15, 20, 40, 60, 80. Why not 10, 15, 20, 30, 50, 80? Why after betting 20 do you jump to 40 right away instead of keeping it going in a 5 progression (10, 15, 20)? Is it because jumping $20 is less conspicuous than $30?
A: Phenomenal questions... I'll take them one at a time:
1) You don't have to have any 1 specific bet spread. I suggest a minimum 1-10 because for most of the deteriorating game conditions, 1-10 is at least 'slightly' worth your time. The best approach is to play with your spread for the game you want to play (keeping in mind your bankroll requirements). If you know you have a max bankroll of X, then that will drive your minimum and maximum bets. After that you just kind of fill in the blanks ramping your bets up to your max bet (which you should have out by TC +5, or TC +4 if you're slightly more aggressive).
A 1-5 spread could be 'profitable', pending the game. I know one game near me is 8D H17 no surrender so it has a house edge of like .66%. A 1-5 spread (where my bet = TC as given in example previously) on this game yields me $1.83 per hour. While "technically" I don't plan to lose money in the long run, I don't exactly consider this "profitable." The 1-5 spread is more for practicing spreading in a live casino environment... Unless you're playing black chips at a double deck table or a great 6D game, because then you could move the decimal place and make a bit more money per hour. So again, I hope you see this depends on the game you're playing as well. For the same 8D H17 game above, if you spread 1-10 (TC +1 = 10, TC +2 = 30, TC +3 = 50, TC +4 = 70, TC +5 = 100) this would yield $7.32 per hour. Again, this is less than minimum wage, but I suppose it is untaxed =p.
Now that same 1-10 spread at a better game, say S17 late surrender, would in turn yield $13.23 per hour. Do you see how just playing a better game affects your hourly EV by over $5/hour?? This is why finding good game conditions and then spreading hard on them is important to MAXIMIZE your profits and EV. Now imagine you spread 1-15 at this better game... EV = $21.20 per hour... Getting to some okay money now, huh? Especially if you wong in or out and try to avoid those negative counts.
2) Again, your spread is mostly determined by just a couple things. 1... your max bet can be determined by your bankroll and how much RoR you want to play with (which is covered in the OP). Past that the real pros will optimize their spread according to the True Count Frequencies (Frequency per 100 hands). A quick example of this is TC +3 and TC +4... While everyone in the world would think TC +4 is more profitable at face value (why not, it's 2% compared to 1.5% more favorable) TC +3 is actually more profitable because it appears more often! In a 6D game TC +3 appears ~4 hands out of 100, where TC +4 appears 2.3 hands per 100. Thus, while you do have a bigger advantage at TC +4, it's actually better to get that bet upped at TC +3 which is going to make you more money in the long run. The best way to tell is by looking at the Gain Per Hand after you play with the spread and try to maximize your gain per hand given your min/max bet.
There is no set spread. After you define your min/max you can really make it whatever you want. I choose to usually up it by the same amount to make it seem like I'm a "progressive player." These are the non-counters whom after they win up their bet by some specific denomination. I like to use this as cover as I slowly work up my spread and before the house knows it I was betting $10/hand and now I'm betting $150/hand because I didn't just jump straight from 10 to 150 but incremented up my spread.
I also want to note that in my example spread I don't have any set increment, it grows geometrically. So first I up my bet by $5, then $25, then $40, then $40, then $40 to my max bet. You definitely don't have to increment by the same amount each time. DO NOTE... My biggest jump (+$40) is first at TC +3 =)... as that's where you start to get in to good profitable hands at more than a 1% advantage.
I could drone on and on about bet spreads, but to be honest I truly believe the best way to develop your bet spread is by defining your bankroll, which defines your max bet, and then you know your min and your max... Play with the in between in some sort of progressive fashion taking note of the Gain Per Hand. The more you get used to these numbers the more you'll be able to tweak them to get the most out of your spread you possibly can. I hope I answered your questions and didn't get too generic. Please let me know if you still have any confusion.
Q: I love the referenced URL for the web, but to the community: is there one iPad app people recommend, for tutoring HiLo count, over another?
I need to practice, practice, practice...
A: Yep, there's so many I couldn't even begin to review/reference them all =p. Most any app that works on the phone also works great for an iPad. A lot of them are free, some of them are a dollar or two (pretty small investment for something that if not done properly could cost you a LOT of money). Overall, just know that the majority of them are the same. Just review and try to find an app that lets you have the most control... It lets you set the number of decks, penetration, keeps the count for you to reference to 'check' yourself, tells you if you make a basic strategy error, etc, etc. Here's a list of about 20 apps that should be fine on either a phone or iPad: http://appadvice.com/appguides/show/card-counting
Q: My local place has a 3-2, 6d, DAS, RSA, resplit up to 3, double anything no surrender. Decent penetration with certain dealers.
Say my BR was $2500, would 1-5 be okay? Red chips? Of course not much money to make money but I intend to play and keep growing my BR by other means of income.
A: 1) Would 1-5 spread be okay on a 2500 bankroll? Let's see =)...
$2500 bankroll, your big bet could be $25, because you'd want to have 100 big bets in your bankroll. Yes, yes, as others might say, you can SIM the game to find a more accurate bankroll for low Risk of Ruin (RoR = going completely broke), but 100xmax bets has been around for a long time and is a good rule of thumb when not running specific SIMs. If you're going to become proficient right now and look to go pro later, I would highly suggest running a SIM of your specific games to better get your bankroll requirements for professional play.
Oddly enough, this would allow for you to have a 1-5 spread while have very low RoR, as 25 = 5xmin bet(5)... So there's your answer to question 1. Yes, you could spread 1-5 on that game with a $2500 bankroll.
2) Is spreading 1-5 profitable/worth it?
I think every beginner counter should make at least a few casino trips using this spread. After all the hours of intense practice, when you think you're really ready to put money on the line in the casino, I think the next step is to deploy the 1-5 spread. Do note this is NOT to make money; this is NOT to build your bankroll... This is to give you LIVE experience actually counting, calculating true counts and actually implementing your bet spread. You might be able to go "Okay, TC is +3, that means I bet $15..." But then you have to reach down, grab the chips, and put them in the circle without looking awkward about it. You learn little things like: When you know the TC is changing this round and you might up your bet next round, if you win the hand pull your ENTIRE bet back while you think of what you want to bet instead of grabbing chips to ADD to the circle. Why? Because if there are chips in the circle and you're "thinking" then the dealer will just start dealing and you'll miss out on betting bigger. Now if you pull your bet completely out of the circle while you decide what to bet then the dealer will wait for you to bet what you want.
These small things to me are the reason one should use the 1-5 spread; not to make money or build a bankroll. In fact, let's look at just how much big money you'd be making at that game with a 1-5 ($5-$25) spread...
1) The Wizards online calculator shows the game has a house edge of: .57% (assuming H17, since you did not specify).
2) Plug in the .57% to the excel doc I showed earlier, and adjust by .5% for each True Count.
3) Update your bet spread to follow the True Count (TC 0 = $5, TC +1 = $5, TC +2 = $10, TC +3 = $15, TC +4 = $20, TC +5 = $25, TC >+5 = $25).
The rest of the spread sheet will automatically update itself if you created it like I explained. Then you can look at your "Total Gain." This is the amount you stand to make per 100 hands you play (which at decent conditions can be taken generically as 1 hour)... Thus, after Steps 1, 2, and 3, one could say you stand to make $1.39 per hour playing that game, with that spread.
If you make a mistake or two per hour (which newer people do) or if you only get 70 hands per hour (instead of 100) you're not even making money at this table. Hell, if you're getting 70 hands per hour AND you make 1 or 2 mistakes an hour, then you're LOSING money at this game! This is the common pit fall most new counters fall in to. The most dangerous thing you can do is "think" you have a winning game and play for hours and hours and days when really you have a losing game! The difference with using the excel sheet like I just did is you know when you have a winning game!
Given you want to play a 3/2 6D game with .57% HE, and you have a $2500 bankroll... If you're planning to build your bankroll you can play off of your theoretical bankroll. I.E. If you're going to put $250/month in to your bankroll until you get to $5,000, then you can play as though you have a $5,000 bankroll right now (so long as you DO in fact continue putting $250/mo in to your bankroll).
Right now my partner and I are playing out of a theoretical bankroll of $32,000. Do we have $32k on hand? Nope. But we usually take about $2k per trip to the casino (which is plenty for a trip and our level/spread), and if we lose it, we'll pull out another $2k for the next trip. If we don't have the money, we'll save for a little bit, then make the next trip. All the math we're doing right now we understand that theoretically we could have to go $30,000 in (and our RoR is <1%). Please be careful with this though... It's not the same as saying "I have a theoretical bankroll of $1 million! Start spreading blacks!" You have to be willing to play up to your theoretical bankroll (by refunding it if you lose what you have on hand).
My honest suggestion is to wait and practice more while you build your bankroll. Then, when you think you're a robot, practice more. NOT just with the cards! I wouldn't advise ANYONE to go play for real money in a casino if they can't even make the simple spread sheet in my OP and figure out what their EV is going to be. You should know if you go play for 3 hours at that game exactly what your EV is. After you can do that, AND you've practiced until you're a robot, then you're ready to put real money on the line.
If you do want to play the theoretical $5k approach... I would suggest the following:
TC <+2 = $5
TC +2 = $20
TC +3 = $40
TC +4 = $60
TC +5 = $80
TC >+5 = $80
Given the game above, this would provide you with an EV of about $8/hour. Okay, so what does this mean...
1) theoretical bankroll of $5k must be funded if you lose the $2500 you have on hand.
2) You're playing with slightly higher RoR by playing 62.5 max bets (5000/80), but I think you should be fine to get a start and understand things while making 'a little' money to try to build your bankroll. While not looking up the actual numbers I'd take my educated guess that by playing 62.5 max bets you're RoR goes up from 1% to something like 5-10% (thus 90-95% of the time you're still safe from going completely broke).
3) Per trip I would take $500 (1/10th your theoretical bankroll). My partner and I take about 1/20th of ours per trip.
4) Understand you're expecting about $8/hour playing this game. You might have a winning trip of $300-$400, but you might also have a losing trip where you dump your trip roll of $500. Just know that if you did that for the rest of your life, you'd average out at making $8 per hour you play.
Any more questions please feel free to ask. Sorry for the long post, there's just so much that comes to mind and I have a tough time filtering it out for 'newer' players because I want you to know as much as possible =).
Q: Also, with the “100 Max bets”, how much RoR does that come out to? Like with my BR of 2500 with the 1-5 spread? Should the RoR be the same as say 1-15 with the 100 max bets red chipping?
A: Risk of Ruin, summed up in to words, is essentially finding your average bet, the number of times you're going to make that bet, and the standard deviations surrounding your action. Basically, it's similar to a bell curve when it comes to standard deviation. You have your expected value, plus or minus some number. Well that number is your standard deviation. 1 standard deviation comes with ~68% confidence, 2 SD's comes with 95%, 3 SD's comes with 99.5% confidence. Without trying to confuse you too much if you're not familiar with the terms, this means you're trying to figure out both your best case (right side of the bell curve) and worst case (left side of the bell curve) scenarios. This is useful for Risk of Ruin because we want to know if we have enough bank roll to withstand the "worst case scenario."
Example: You want to spread 1-10 ($10-$100) at a 6 deck game. I plugged that spread in to my spreadsheet... The average bet is calculated as well. It is the total amount bet (per 100 hands) divided by the number of hands actually played (since we don't play hands less than TC -1). Your total amount bet is Column D * Column C. The number of hands you play is the summation of your "Frequency per 100 hands" column since you're ignoring anything > TC +8 and < TC -1. This comes out to be 84.55 hands per 100 on my sheet I believe (or at least the one I'm looking at now =p).
I'm using the spread (from TC -1 to 8...): 10, 10, 20, 40, 60, 80, 100, 100, 100, 100. The average bet is: $21.27. For example and simplistic sakes I'm going to use an average bet of $100 for my examples below, but note this is no different than your $20 average bet (except x5 =p).
So now I have my average bet. In order to calculate your standard deviations you need your advantage, and the number of events. In blackjack, we take our advantage to be about an average of 1%. Some may have more, some slightly less, but this is a good safe bet (and remember this is the 1% advantage you're playing for in the long run). So now to find your standard deviations you simply need to do some easy math:
Average Bet: $100
Avg Advantage: 1% (.01)
Original Standard Deviation in Blackjack for 1 hand: 1.1*AvgBet (the 1.1 is a calculated number per the game of blackjack done previously by much better math people than I).
The standard deviation past 1 hand can easily be calculated by the following... SD = OriginalSD*Sqrt(NumHands), where OriginalSD = 110 from above.
Standard Deviation (SD)
So let's take a look at a couple things. One, before I get to RoR, let's look at another example of WHY COUNTING WORKS. If you notice above, for 1 hand, with a $100 bet, and a 1% advantage, you'd expect to make $1 (every time you made this bet over and over on AVERAGE in the long run). Now the Standard Deviation of this hand is +- $110, because on any given hand you could lose (-$100), win (+100) or win/lose more pending doubles/splits/blackjacks/etc.
The rest are just multiplications of each other, BUT the interesting thing is as you reach a million hands. A professional actually playing 40 hours per week (not counting scouting, traveling, etc) plays a total of 2080 hours per year. If this pro finds good conditions and is getting 100 hands per hour, then they are playing 208,000 hands per year. This should tell you that it would even take a pro 5 years to reach a million hands... but by the time that they do they should expect to make $1,000,000 (based off the averages above). However, at this point, you'll notice the SD is +- $110,000. So what does this mean? You've effectively erased luck from the game. You might make $890,000, or you might make $1,110,000... but you're making a large sum of money either way! The variance has been effectively minimized and you are left with your expected value of about 1% resulting in $1 per hand you play.
Now for the RoR
Your RoR is based off the number of hands you plan to play. In the example above, using a $100 average bet, you could make $1,000,000 +- $110,000 with a $110,000 bankroll (as this is the most variance you could ever expect to lose in getting to your 1,000,000 hands).
So what does this mean for you? Well, you're 1/5 of the above average bet, and 1/5 of the results... If you divide everything by 5 above, then you could expect your average bet to be $20, $.20 to be your average gain per hand, and +-$20.20 to be your SD. Multiplying out then in 5 years, playing 1,000,000 hands you could expect to make $200,000 +-$22,000. Not too shabby, but you can clearly see why you'd want to re-evaluate this if you were to go "pro" =). From this though you can see that you'd need a bankroll of $22,000 to withstand the possible swings you'd be taking in getting to that point.
Another note, this is for one standard deviation, which is with only ~68% confidence. To get a full SD of 99.5% confidence then you have to do a lot trickier math than I'm willing to write out on here =P... but overall that should be fairly close and if anything I would just expect your bankroll requirement to go up from like $22,000 to something close to $30,000ish (educated guess).
There's a lot of RoR calculators where you plug in your average bet, your gain per hand, and your bankroll and it tells you what % chance you have of going bust (as opposed to doubling your bankroll). Here's the Wizards RoR calculator. Others can be found at qfit, or other places simply by Googling "Risk of Ruin calculator."
One last note: Any level of serious/profession play should be played with <1% RoR. If you're part time or just a fun hobby, I would still suggest keeping it as low as you possibly can, but understandably you might not 'need' to keep it less than 1%.