Question for MathProf, DD, or Stealth: KO or HI-LO?

Hi-Lo seems to pretty much be the industry standard these days. For those of us who play shoe games, which system would you guys suggest using between KO and Hi-Lo? Let us assume the counter plays in AC, Foxwoods, and Mohegan 6 deckers.

KO seems a lot simpler given it can be used in RC mode and there are hardly any index values to memorize. All one has to do with KO is memorize how much to bet at what RC. On the other hand, Hi-Lo seems to be much more complex involving deck estimation and a TC conversion. The idea of a dealer waiting for me to place my bet in the circle because I am staring into space calculating my bet concerns me. The question on my mind is whether or not the added effort (and additional errors) are justified by any potential theoretical gain. Simulations comparing the 2 systems (I-18 Fab4) for shoe games with typical rules show SCORES which are quite similar. Hi-Lo does do better in theory, although not by much. Another added benefit of Hi-Lo is shuffle tracking, but how many players can honestly say they shuffle track with good results? How many casinos even have shuffles that are trackable? I understand that some casinos (although I personally never played in them)cover the shoe as well as the discard rack to deter counters. Hi-Lo cannot beat this game but KO would work.

Card counting is tough, especially in a casino enviornment. Perhaps simplicity is the best way to go. I heard that even legendary player Ken Uston employed an unbalanced count at the tables.

MJ

You need to have it down to being automatic.. so you can observe, listen and think of all the bs going on.

overprepare

>>>> I heard that even legendary player Ken Uston employed an unbalanced count at the tables. <<<<

Ken Uston primarily played 1&2 decks using Revere RPC which is a balanced system.

This is probably the most debated question here and most things have been said one day or another, so I'd like to stress a point that is hardly spoken of: I mean flexibility.
Hilo can easily adapt to any rule set,since the advantage situation is always clear at any point in the shoe. You can even play a game with poor penetration if there are outstanding rules to compensate for.All you need to do is adjust your bet ramp.
KO can hardly be played with poor penetration because you simply don't get enough action.Adapting the key counts would require some form of true counting but that's exactly what you wanted to avoid.
Another thing I like about hilo is the fact that I don't have to think for a second when I'm faced with the situation 16v10 or 12v4 which come up all the time.The fact that there are not many indices to learn with KO is one thing, but the playing efficiency will suffer from such crude RC-indices.
But if you have troubles with TC-conversion,you better play KO. Just look for good pen and spread.

Francis Salmon

Uston used the SS count for awhile. It's an unbalanced count.

I think my view on this issue is fairly well-known on BJ21, at least among Green Chippers (I hope.)

The key issue here is what your most comfortable with. If you find it easier to play with KO, and you have more confidence when you use it, then by all means do so. If you fell more comfortable with High-Low, then do so.

I think there are some advantages to High-Low, and Francis has outlined a few of these. I also think the difficulty of true-counting is also over-stated. BUT these considerations pale compared to the consideration of usability.

If you will make fewer errors with KO, then by all means use it. If you have more confidence in your count with KO, then use it.

Both KO and Hi-Lo get the money. Both systems have strengths and weaknesses, and people can spend a lot of time debating the relative value of both. In the end, I agree with MathProf: use whichever system with which you are comfortable. If you are playing shoes, the differences are so negigible that it doesn't matter.

A comment on a couple of things you said:

The idea of a dealer waiting for me to place my bet in the circle because I am staring into space calculating my bet concerns me.

If a player has to stare into space and think about calculating his bet to the point where a dealer has to wait for him to make the bet, chances are, he isn't ready to play in a casino.

I understand that some casinos (although I personally never played in them)cover the shoe as well as the discard rack to deter counters. Hi-Lo cannot beat this game but KO would work.

There are other ways to estimate the number of cards left in play, so one could still use Hi-Lo in such situations. However, the other ways to make those estimates can require more mental gymnastics and might affect one's efficiency.

You seem to suggest that Hi-Lo can compensate for poor penetration given favorable rules. KO can nearly outperform Hi-Lo under similar circumstances. Even if your math is perfect, your deck estimation is not and you will have an imprecise TC as a result!

As far as 16 vs 10 and 12 vs 4, even if you played them incorrectly over your lifetime the difference in EV would be negligible, at least when the minimum bet is out. The fact of the matter is that KO has only 13 indices for shoe games, most of which occur at one index number. Placing bets and making index plays in RC mode is hard to beat. I'll bet the errors most players make with Hi-Lo exceed the gain in using it over KO. There is just sooooooo much room for error in the TC conversion. Besides, KO can outperform Hi-Lo using EV maximizing indices for the Catch-22 and a good bet spread for the shoe game. Do you disagree?

MJ

First,why do you ask a question if you think you know the answer.
The problem with sims is that they always assume optimal betting.In practice this is impossible to do with KO unless you resort to ...guess what - deck estimation!
Don Schlesinger gives 16v10 as the second most important index play after insurance and I have no reason to question that.
TC-conversion might be prone to error but no conversion is an institutionalised error!

Francis Salmon

The problem with sims is that they always assume optimal betting

This is totally false.

In practice this is impossible to do with KO unless you resort to ...guess what - deck estimation!

None of my sims use deck-estimation with KO. So your two comments are contradictions.

TC-conversion might be prone to error but no conversion is an institutionalised error!

You seem to be the only person in the world that believes this. And yet you constantly repeat it as if everyone knows it.

Look, almost no one agrees with you. So, why do you keep bringing up my name when I'm just one out of all the people you disagree with on this subject? Your post should be titled "Sounds like Everyone" instead of "Sounds like Norm Wattenberger."

I don't follow all this "true conversion implies errors" stuff.

I played lots of tennis in college. I practiced until I could return balls with the skill level necessary to compete with the players I encountered regularly. Why would I go onto the court knowing that I could not even hit the ball in the "sweet spot" nearly every time? Why would I not practice enough to at least not make gross technical errors like giving my opponent a soft lob and then rushing the net so he can return it in a way I have no chance to reach?

21 should be treated the same way. Anyone can buy CVBJ and practice until they can play for hours with no errors of any kind, which means those "true count conversion errors" are not happening as CVBJ will instantly grumble at you for an incorrect bet or an incorrect BS departure.

Bottom line is that these errors should not be taken for granted and the assumption that they are simply going to happen is not true. You are a good case in point in that you seem perfectly at ease computing the TC to 1 decimal place, where I would most likely produce many errors until I practiced that enough. But for TC conversions to whole numbers, I simply fail to see the problem. In fact, if you practice enough the conversions become unimportant. In DD games, with 1/2 deck played, and a RC of 3, I don't stop to think 3 / 1.5 or 3 / (3/2) or even 3 * .66, I just know to bet and play properly without knowing anything else other than running count and decks left. I believe that is the result of lots of practice, so that the TC conversion is done subconsciously. And since CVBJ has convinced me I can do that for many one-hour sessions without an error, I'm happily not worrying about errors, and worrying more about the lousy cards I often get. :)

Now if you want to play, without practicing, then perhaps KO is a better alternative. Of course, I know tennis players that play without a decent backhand as well. They run around the ball to use their forehand stroke, and it works, but puts them in awkward positions at times. I'd prefer to enter a casino prepared to do battle, not partially prepared, hoping for the best..

When asked I suggest either KO or HiLo 90% of the time. The performance is very close. For some time I have tended to favor KO over HiLo when giving advice (even though I was not allowed until recently to include KO tables in CV.) I look at the issue of TC conversion a bit differently. Yes, once you are well practiced it comes almost naturally and the difficulty may well be over-stated. But I get calls from beginners and my fear is that no matter how much you try to instill in them the wisdom of avoiding the casinos until counting is automatic - I know full well most won't have the patience or discipline to wait. Such people will never make it to the point that TC conversion is natural. Now maybe they shouldn't be counters at all. But, I think KO requires a substantially lower hurdle in the early part of an AP career. Of course even if I recommend KO, I still suggest they read a book like PBJ, or lately Rick Blaine's new book, as the KO book ius a bit skimpy on side issues you run into as counter.

"In fact, Hi-Lo is notably superior to both Red Seven and KO in one and two-deck games, and from many professional players� perspectives, where accurate bet-sizing according to Kelly principles is important, Hi-Lo is also far superior in shoe games."
From an article in Blackjack Forum 1999

Still think that I'm alone with my opinion?You better continue the debate with Arnold.

Francis Salmon

It should be noted that in most comparisons between HiLo and KO (as well as Red7,) compromise indexes are used for the unbalanced strategies. That is KO Preferred instead of KO Full. Obviously there is some loss when using compromise indexes; particularly in pitch games. If you compare HiLo with KO and use accurate indexes for both; they are extroadinarily close. KO appears to have a tiny advantage in most cicumstances.

Where did Arnold say "TC-conversion might be prone to error but no conversion is an institutionalised error!"? I can't imagine him saying anything like this. He suggests Red7 to this day.

Your quote answers a different question. And that was seven years ago when people were still arguing over Kelly betting and before SCORE. I can't find the article; but probably also refered to compromise indexes. You recently said that people go to Cacarulo for accurate numbers. For a more recent comment - last month he said KO-Full outSCOREs HiLo using Kelly betting. As usual, my sims agree with Cac's sims.

I thought it was clear to everyone what I meant with "institutionalised error".
When you use a fixed number for something that's constantly shifting throughout the shoe you have an inbuilt error.This goes for playing as well as for betting.Accuracy is not possible without deck estimation.

Francis Salmon

and no one agrees with you. So why do you keep saying I am the only person that disagrees with your belief that KO can't work. It does work. And Arnold would be the first to agree. And one of the most successful teams uses it. You think KO is a failure on some sort of principle. But, you never back this up with any data. Francis, KO is a great strategy. Every simulator says so.

You do realize that "deck-estimation" is built-in to unbalanced startegies? That is the entire point of such.