lagavulin62: ego aside, here's my attempt tko preferred

[lagavulin62]

here are the results I get using tko preferred with all category A and B indexes with some exceptions. 16 vs 10 was taken at TC equivalent of 1(-3). since in the book insurance is taken one below pivot and I wanted to keep it simple I only took insurance at the pivot(0). so that users will know how I set cvcx to calculate the tc(in case I made a mistake) I set the irc at -24 and checked flooring for tc calculations, and made sure it calculated per full deck flooring on the sim page.

1 billion rounds, 4 players, 4th seat, 6d, 75% pen, das, doa, no surrender

spread ev sd/hr ror di c-score No

1-20 1.06 67.4 14.8 5.15 26.54 37,677
1-16 .96 57.6 12.0 4.89 23.89 41,871
1-12 .82 47.4 12.3 4.47 19.98 50,059

I noticed those scores in cac's sims were with a pen of 5/6 and for playing alone. I think the parameters I set are more realistic so if the numbers seem low thats a big reason why. of course this is not meant to be a comparison so understand I'm not trying to prove any system is better. but several people had wondered about how tko would sim so this is my contribution(however flawed it may be) for now and hopefully others will run with this idea. I have been experimenting with generating indexes for this but I'm not ready to embarass myself with that just yet. so criticize away

[Norm Wattenberger]

Between KO-P true counted and Full KO running count only, which is easier and which more powerful.

[Cacarulo]

> Between KO-P true counted and Full KO running count
> only, which is easier and which more powerful.

This is a good point which I have already simulated. The used conditions are always the same: 6D,S17,DOA,DAS,SPA1,SPL3,NS,5/6,heads-up,5000 million rounds

1) TKO + C22 (exact floored EM-indices)

play-all | rounds played = 100.00% 

 spread   ev/h   sd/h   ror%      n0    di  score      ekb   avb    unit  kelly 

 ------------------------------------------------------------------------------ 

 1 -  4  0.758 23.506  13.53   96053  3.23  10.41   728.51 1.666  13.727  1.000 

 1 -  8  1.846 35.253  13.53   36486  5.24  27.41   673.38 2.096  14.851  1.000 

 1 - 12  2.806 46.245  13.53   27170  6.07  36.81   762.27 2.481  13.119  1.000 

 1 - 16  3.629 55.570  13.53   23450  6.53  42.64   850.96 2.797  11.751  1.000 

 1 - 20  4.430 64.904  13.53   21462  6.83  46.59   950.83 3.115  10.517  1.000

2) KO-FULL with 95 EM-indices

 play-all | rounds played = 100.00% 

 spread   ev/h   sd/h   ror%      n0    di  score      ekb   avb    unit  kelly 

 ------------------------------------------------------------------------------ 

 1 -  4  0.679 22.318  13.53  108068  3.04   9.25   733.67 1.544  13.630  1.000 

 1 -  8  1.770 34.376  13.53   37720  5.15  26.51   667.63 1.954  14.978  1.000 

 1 - 12  2.728 45.462  13.53   27781  6.00  36.00   757.74 2.303  13.197  1.000 

 1 - 16  3.609 55.856  13.53   23950  6.46  41.75   864.40 2.624  11.569  1.000 

 1 - 20  4.457 65.994  13.53   21928  6.75  45.60   977.24 2.935  10.233  1.000

3) KO + C22 (exact RC EM-indices)

 play-all | rounds played = 100.00% 

 spread   ev/h   sd/h   ror%      n0    di  score      ekb   avb    unit  kelly 

 ------------------------------------------------------------------------------ 

 1 -  4  0.651 22.192  13.53  116043  2.94   8.62   755.97 1.548  13.228  1.000 

 1 -  8  1.728 34.199  13.53   39179  5.05  25.52   676.93 1.966  14.773  1.000 

 1 - 12  2.664 45.066  13.53   28621  5.91  34.94   762.41 2.316  13.116  1.000 

 1 - 16  3.531 55.353  13.53   24575  6.38  40.69   867.73 2.642  11.524  1.000 

 1 - 20  4.352 65.203  13.53   22446  6.67  44.55   976.87 2.950  10.237  1.000

Note that using an exact TC calculation (with decks remaining estimated to the exact card) is 2.1% better than using the KO-FULL version and 4.8% better than KO+C22.
Think that nobody can be that precise in the TC calculation and the real percents are actually lower than that.
Now the question is: Is it really worth learning those TKO incarnations?

Sincerely,
Cac

[koolipto]

> Note that using an exact TC calculation (with decks
> remaining estimated to the exact card) is 2.1% better
> than using the KO-FULL version and 4.8% better than
> KO+C22.
> Think that nobody can be that precise in the TC
> calculation and the real percents are actually lower
> than that.
> Now the question is: Is it really worth learning those
> TKO incarnations?
> Sincerely,
> Cac

Cac and Norm
I have read most if not all the threads on KO/TKO as well as TKO/A on this and the main forum for well over a year. A lot of what Cac has previously posted on TKO has been systems comparisons (TKO to Hi-Lo, UZB etc). Never once have I detected a hint that true counting KO may not be worth it. As for deck resolution, there have been a number of recent threads suggesting that precise deck resolution did not matter that much.

If KO is pigeon holed as a very easy system for the masses and for people who do not wish to explore addtional power, then I can understand the comments (juxtaposed against reKO for example). But I would think, by definition, anyone seeking to true count does not fall in to that category. Is there any reason to think that the attempt to true count KO is subject to more mistakes or deck estimation error than any other true counted system?

[Norm Wattenberger]

> Is there any reason
> to think that the attempt to true count KO is subject
> to more mistakes or deck estimation error than any
> other true counted system?

The nice things about balanced strategies is that the running count has a relatively small range and the true count is always the same sign as the RC. Much of the time you don't need to calculate the TC at all. I think true counting an unbalanced strategy is a bit more difficult than true counting a balanced strategy and any increase in effort is likely to increase errors. Just my take on the difference.

[ToAnyOne]

There are valid reasons other than simplicity for wanting to use an unbalanced count. Unbalanced counts are generally designed with the purpose of alerting you at the EXACT moment a certain determining condition is attained regardless of any other considerations.

i.e.
R7: as soon as edge requires bet raise
KO: when to push the max bet
Ten C: when to take insurance

In my experience true counting unbalanced counts is harder than a balanced count of equivalent level and that added difficulty will bring no improvement whatsoever in multi-deck games. What I found the most difficult with unbalanced counts in general is knowing when to wong-out.

I also think that people put too much in the belief that the only reason why someone would want to use an unbalanced system is to ease their mind; the fact that the majority of people do not take pleasure at putting their mind to work is not a reason to discourage experimentation and belittle the possible benefits of doing so.

That being said I do not use an unbalanced count, but for someone that does and wants to maintain the simplicity, I think R Harris's solution is the best, I would maybe add WO if RC is less than IRC after a few rounds, for infrequent WO's.
Nevertheless I would encourage anyone to try different systems or even types of systems because every system brings a little bit of a new perspective on the game, which is definately something we can never have too much of.

Maybee I went off a little too much on a tangeant, but that's my opinion, I hope you like it,

TAO

[Trapper]

> What I found the most
> difficult with unbalanced counts in general is knowing
> when to wong-out.

The KO book gives 4 RC wong out points for decks played (6 DK). You could add wong in points as well which would mean memorizing 8 numbers. Not too hard.

How much of the 4.8% increase in SCORE for TKO over KO C22 (which I assume would perform very similarly to KO preferred) would be achievable in the real world? Wouldn't learning and applying WI / WO points while playing standard KO Preferred or reKO outperform the other variants discussed on this thread by a big margin?

[koolipto]

[AutomaticMonkey]

> There are valid reasons other than simplicity for
> wanting to use an unbalanced count. Unbalanced counts
> are generally designed with the purpose of alerting
> you at the EXACT moment a certain determining
> condition is attained regardless of any other
> considerations.

> i.e.
> R7: as soon as edge requires bet raise
> KO: when to push the max bet
> Ten C: when to take insurance

> In my experience true counting unbalanced counts is
> harder than a balanced count of equivalent level and
> that added difficulty will bring no improvement
> whatsoever in multi-deck games. What I found the most
> difficult with unbalanced counts in general is knowing
> when to wong-out.

> I also think that people put too much in the belief
> that the only reason why someone would want to use an
> unbalanced system is to ease their mind; the fact that
> the majority of people do not take pleasure at putting
> their mind to work is not a reason to discourage
> experimentation and belittle the possible benefits of
> doing so...

My opinion is very similar, and I can't think of a good reason to use an unbalanced count in a shoe game. My deck estimation skills are poor but they don't have to be that good to be effective.

That said, I can't think of a good reason to use a balanced count in a pitch game! Especially if you are playing a lot of indices, or sidecounting for a sidebet, recalculating after each card can be onerous and doesn't buy you all that much. You're generally not Wonging in or out of pitch games so that drawback is minimized.

So my solution is to learn two similar counts, a balanced one for shoe and an unbalanced one for pitch. The combo I use now is RPC and UBZ. They differ only in the weight of the ace so the confusion is lessened, and each system is mighty powerful. Other similar combinations are RPC/BRH-1, High-Low/KO and High-Low/Red 7.

[Trapper]

If you were using KO in pitch games would it really make sense to learn HI-Lo for shoes for straight counting? All the sims I have seen show the difference in SCORE between the two counts as small to non existent. I suspect you are a rare counter who can switch between counts without accumulating errors. I know it would take me a long time to stop counting sevens. Switching back and forth between counts with different tags would be out of the question.

[Parker]

> My opinion is very similar, and I can't think of a
> good reason to use an unbalanced count in a shoe game.
> My deck estimation skills are poor but they don't have
> to be that good to be effective.

> That said, I can't think of a good reason to use a
> balanced count in a pitch game! Especially if you are
> playing a lot of indices, or sidecounting for a
> sidebet, recalculating after each card can be onerous
> and doesn't buy you all that much. You're generally
> not Wonging in or out of pitch games so that drawback
> is minimized.

> So my solution is to learn two similar counts, a
> balanced one for shoe and an unbalanced one for pitch.
> The combo I use now is RPC and UBZ. They differ only
> in the weight of the ace so the confusion is lessened,
> and each system is mighty powerful. Other similar
> combinations are RPC/BRH-1, High-Low/KO and
> High-Low/Red 7.

Another obvious combination would be UBZ2/Zen. It is my understanding that George C. himself does this, or at least did so at one time. The only difference between the two counts is the tag for the 3.

That being said, I have enough trouble learning one count, let alone two. In addition, it would be necessary to learn two sets of indices.

The first thing anyone considering this should do is invest in some software and run some sims. Run the games you normally play at the stakes you normally would play, using the bet ramps you anticipate using. Then factor in the number of hours you expect to play, so you can come up with a real-world dollar figure of how much you will actually gain on a monthly or annual basis by using two systems.

My guess is that, at this point, most people will abandon the project.

[ToAnyOne]

When I was waking up this morning, I just had an interesting idea to get the best of all the first level counts.
If you are wondering weather to count KO or hi-lo or red seven, you could have it all and more by just counting hi-lo RC and keeping track of the amount of 7's seen.
Then in order to know when to first raise your bet you add the hi-lo RC plus half of the 7's count; using IRC of 0 you raise your bet at +12 in 6 deck. Here you have the benefits of red 7, but vastly superior in precision because all 7's are worth .5, which is much more representative of the EOR than counting half the 7's as 0 and the other half as 1.
Then in order to know when to put out max bet, you just add both counts, and you got KO, max bet at +24 for 6 deck.
When you want to know when to WO, you just use the hi-lo and ignore the side count.
Then you can add multi-parameter tables such as stand on 14vT if the pack is short 1 1/2 sevens/deck. This can add up to .118 on the PE. Slight increase in IC as well.
Lastly, combine this with a super 7 side bet and you are rocking.
If you feel this is too sensitive information, feel free to bust the post, if not, all comments would be appreciated.
Thanks,

TAO

[Parker]

Actually, the concept of counting all the 7's as .5 has been around almost as long as Red 7. In fact, most simulations actually do it this way (and give slightly inflated performance results) simply because the simulators are not set up to keep track of suits (although Norm has probably figured out a way around this by now).

The only problem with counting 7's as .5 is that, strictly speaking, it transforms Red 7 into a level 2 count, with added complexity.