Info on DHME count

**dalmatian** · 01-07-2025, 12:41 PM

Hi Everyone,

I'm looking for more information on this DHME count that counts low vs middle vs high cards. Does anyone have any links or literature that explains how to use this count along with indices?

Initially I wanted to expand on the high low count and track low cards used in hi low (2,3,4,5,6) vs intermediate cards (7,8,9) vs high cards (10,A). This would add that extra dimension of intermediate cards to help improve playing decisions (PE) and even help in betting (BC) WITHOUT actually changing the hi low count (you could revert to hi low at any moment). However, I noticed that the DHME groups the 6s into the intermediate cards and removes the aces from the high cards and makes that a separate count. Is this really that much more powerful? Or would making my own custom count (not sure if anyone has claimed credit for this yet) and doing low cards (2,3,4,5,6) vs intermediate (7,8,9) vs high (10,A) be worth the effort.

Thanks for input!

**Freightman** · 01-07-2025, 12:54 PM

I don’t know what DHME stands for, however your description closely (not exactly) mirrors FBM ASC Advanced where 6’s are included in the 6789 grouping. I refer to this as grouping as aiding QTC - Quality of True Count. Aces are reckoned with halves but are side counted to improve various playing decisions. Naysayers question, but yes, it’s very powerful.

Regarding your secondary thought of hi lo with 789 side count - also worth the effort - Secretariat can add some interesting thoughts.

**dalmatian** · 01-07-2025, 01:11 PM

@Freightman

DHM Expert. {PCS/L1}. (D. Howard Mitchell aka Edward Gordon) (c1973). DHM Associates, Arroyo Grande, CA. This system requires the player to count four different groups separately (low 2-5, middle 6-9, high 10, and aces). The DHM systems were once available through the Gambling Times Systems Shop. Originally based on the Gordon Count.

**dalmatian** · 01-07-2025, 01:19 PM

Originally Posted by Freightman

mirrors FBM ASC Advanced where 6’s are included in the 6789 grouping. I refer to this as grouping as aiding QTC - Quality of True Count.

so if the ratio of high card to low cards is large BUT there are also a large amount of intermediate cards left to be played, this is a "lower quality true count" than the same high card to low card ratio but with little intermediate cards left to be played?

**Freightman** · 01-07-2025, 01:34 PM

Originally Posted by dalmatian

so if the ratio of high card to low cards is large BUT there are also a large amount of intermediate cards left to be played, this is a "lower quality true count" than the same high card to low card ratio but with little intermediate cards left to be played?

Essentially, yes.

The idea here is a variable single ramp or a dual ramp system. You are in your lower ramp when intermediates surpass high cards. Play decisions influenced include insurance, 10v10, 11v10, 99v7 general basic doubles and splits etc etc. Totally unsuitable for a team approach part of todays posts.

Check out this link, though I’ve modified a couple of things.
https://www.blackjacktheforum.com/sh...ine-by-request

I’ve thought about though not yet acted on simplifying (for a couple of reasons I won’t disclose) of simply counting 98 since clearly more valuable vs 76.

**Secretariat** · 01-07-2025, 03:37 PM

Hello Dalmatian and Freightman.

Regarding the expanded 3-column HiLo card system it has the potential to bring HiLo to/or above the basic Halves level. Not sure if it can surpass Freightman’s modified Halves system. Depending on some adjustments, HiLo modified can lead to perfect insurance decisions or pretty close. Knowledge of 789s helps efficiency with stiffs. A surplus of 789s suggests a lower betting ramp but also indicates a weaker dealer 7 as upcard. This can be capitalized on. How to determine the exact indices is another ball game. Cacarulo and I worked on it a bit but we never went all the way through.

**Secretariat** · 01-07-2025, 03:41 PM

Regarding the Gordon Count, it has a lower betting efficiency than Hi Lo but if you add the 6 as a side count, and oppose it to the ace, you have the exact same running count as HiLo. I think it would be best to oppose 67s to 89s as it improves both betting and playing efficiencies. Or just count the 67s and use your eyes to see if they are above/under normal. Another approach would be counting 89s like Freightman mentioned.

Note that Gordon Count with double counting tens would be exactly a reversed Tarzan Count, the top PE system with built-in perfect insurance. However, for an experienced player like you it would probably be ill-advised to change completely to a count-up-and-zero-down approach.

Here's a link for the Gordon Count. Hope it's OK Norm,
https://groups.google.com/g/rec.gamb...m/Wxi9TwT3uxIJ

**dalmatian** · 01-08-2025, 02:24 PM

Originally Posted by Secretariat

Regarding the Gordon Count, it has a lower betting efficiency than Hi Lo but if you add the 6 as a side count, and oppose it to the ace, you have the exact same running count as HiLo. I think it would be best to oppose 67s to 89s as it improves both betting and playing efficiencies. Or just count the 67s and use your eyes to see if they are above/under normal. Another approach would be counting 89s like Freightman mentioned.

Note that Gordon Count with double counting tens would be exactly a reversed Tarzan Count, the top PE system with built-in perfect insurance. However, for an experienced player like you it would probably be ill-advised to change completely to a count-up-and-zero-down approach.

Here's a link for the Gordon Count. Hope it's OK Norm,
https://groups.google.com/g/rec.gamb...m/Wxi9TwT3uxIJ

@secretariat

in your opinion, would "gordon count" be feasible on an online blackjack game where you can utilize a computer but the pen is 3/6 or 4/8??

**Secretariat** · 01-08-2025, 03:26 PM

Originally Posted by dalmatian

@secretariat

in your opinion, would "gordon count" be feasible on an online blackjack game where you can utilize a computer but the pen is 3/6 or 4/8??

@Dalmatian

The Gordon/Tarzan counts separate from the rest at deeper pens because of PE.
At 6D or 8D and 50% pen, and assuming near perfect play (with your computer) I don’t think you would get much of an advantage over HiLo (with near perfect computer play too) but this would have to be simmed and then you could determine if it's worth your time. Not a great game for sure and you know that.

Important factors would be rules, side counts used, and number of players.
Main factor would be the possibility to wong in and out.

Your computer could probably achieve perfect betting and perfect insurance, but how close to perfect play can you program it?

**dalmatian** · 01-08-2025, 03:36 PM

Originally Posted by Secretariat

@Dalmatian

The Gordon/Tarzan counts separate from the rest at deeper pens because of PE.
At 6D or 8D and 50% pen, and assuming near perfect play (with your computer) I don’t think you would get much of an advantage over HiLo (with near perfect computer play too) but this would have to be simmed and then you could determine if it's worth your time. Not a great game for sure and you know that.

Important factors would be rules, side counts used, and number of players.
Main factor would be the possibility to wong in and out.

Your computer could probably achieve perfect betting and perfect insurance, but how close to perfect play can you program it?

I see your point. I'm thinking of creating a python program that when input the rank of cards depleted uses the Gordon count to determine the play on the next hand as well as amount to bet. Kind of as a mini project. Maybe also show how much better it is than hi low. Will be interesting. Sadly, not very practical......

**dalmatian** · 01-08-2025, 03:38 PM

Does anyone know if modern computing is capable of taking all the cards depleted, then taking the current players cards, running several million rounds simulation using precise rules and penetration to determine the perfect play, and spitting out the result via vibrations to the player in a matter of seconds? Probably not, haha......

Or is there a better way to get perfect play?

**Secretariat** · 01-08-2025, 04:06 PM

Originally Posted by dalmatian

Does anyone know if modern computing is capable of taking all the cards depleted, then taking the current players cards, running several million rounds simulation using precise rules and penetration to determine the perfect play, and spitting out the result via vibrations to the player in a matter of seconds? Probably not, haha......

Or is there a better way to get perfect play?

Take a look at KC's simulator at BJSTRAT
Does not include players' vibrations though.