Should K-O Preferred Indices be adjusted for 2D games with very poor pen.??

Appendix V of KNOCK-OUT BLACKJACK concludes that the indices for the Preferred Matrix are "appropriate for any reasonable level of penetration, between roughly 60% and 90% of the pack." Earlier in Appendix V, it is mentioned that the values for A, B, and C (in the Preferred Matrix) might need an adjustment downward in "an extreme case where a 2-deck game has only 50% penetration." The logic goes as follows: in a 2D game with 50% pen., the average RC = -2 [(-4+0)/2], whereas in a 2D game with 75% pen. (the assumption upon which the strategic matrix entry values of A, B, and C were calculated), the average RC = -1 [(-4+2)/2]. So, if I am interpreting this correctly, the following values should be used in a 2D game with 50% pen:

IRC = -4
C strategy plays = -5 (instead of -4)
Key Count/B play = 0 (instead of +1)
Pivot Point/A plays = +3 (instead of +4)
Insurance play = +2 (instead of +3)

Alternatively, the same effect would be achieved by changing the IRC to -3, and using the standard values for C, B, A, and Insurance (-4, +1, +4, and +3, respectively).

Am I way off base on this, or am I on the right path? I realize that the best advice would be to simply not play in such a game, but the selection of games available to me is not very broad; thus, I would be very appreciative of any insights that can be offered to me on this topic. Thanks in advance.

I use the K-O count and am sometimes am "forced" to play 2 deck games with poor pen, most often when the relief dealer comes in and cuts the deck right in half. Games like this are quicksand.

With lousy penetration you will have precious few high counts- and if you lose on a few of them with your max bet you can be sunk. You have to forget about cover betting if you want to play these games because you will need to take full advantage of all opportunites. Try as hard as you can to play solo with the dealer, with the good counts coming so seldom, don't waste time sharing them with a full table.

I think you will definitely need a more aggressive spread than the 1-5 recomended in the K-O book. 1-8 at least, maybe 1-10 or even more. Consider spreading to a second hand occasionally to get more money on the table on big counts. Also, get out whenever you can on negative counts.

I don't think I would adjust my count to lousy game, but then I'm not stuck with a bad game all the time. Shop your dealers- surely at least some of them will give you a few more cards. Sometimes lazy dealers give the best penetration because they don't want to shuffle as often.

Hope some of my opinons help. One of the best plays with this game is to leave the table.

Sentry

James Grosjean in his book Beyond Counting (pg 39) makes the same point as you: namely that the key count in KO assumes the standard penetration of 75%. He then says that "...in a game with a radically different penetration the average number of decks remaining would be different enough that the key count could change". He also says that different rules (eg; H17 with added 0.2% disadvantage off the top) would also move the Key Count.
By way of example he notes that the key count of -4 (IRC=-20) is for the standard 6-deck, S17, noDAS,DOA,75% RS4, noRSA game.With a change of rules from noDAS to DAS his key count changes to -5. And taking that same 6-deck, S17,DAS,DOA,75% RS4,noRSA game and changing the penetration level from 75% to 50% would change his key count to -8.
So,it seems from what I've read that you are on the right path. If not, at least you are in excellent company.

Based on Grosjean's example that you cited, I am now convinced that the Key Count in the 2-deck game with 50% pen. should be lowered by 1 (from +1 to 0). In that game, the average RC = -2, whereas in the "standard" game with 75% pen., the average RC = -1. Similarly, in Grosjean's example, the average RC in the 6-deck game with 50% pen. = -14, whereas in the "standard" game with 75% pen., the average RC = -11. His lowering of the Key Count by 3 corresponds to the average RC being 3 lower. (In addition, the 2-deck game with 50% pen. that I play allows DAS, which furthers the case that the Key Count should be lowered, based on Grosjean's example.)

What I'm not sure about is if the Pivot Point should be unitarily lowered with the Key Count. Does Grosjean make any mention of that? To me, the information in chapter 7 of KNOCK-OUT BLACKJACK conflicts with the supposition in appendix V that says "the values for A, B, and C (in the Preferred Strategy Matrix) might need an adjustment downward as well...(in) an extreme case where a 2-deck game has only 50% penetration." Since the strategic matrix entry value for A in the Preferred Matrix corresponds to the Pivot Point, that is saying that the Pivot Point may also need to be lowered in that case. However, in chapter 7, it is noted that "at the pivot point of +4, the K-O running count is always exactly equal to a K-O true count of +4; this relationship holds true regardless of the number of decks in play or the number of cards already dealt out, so long as the Knock-Out IRC is calculated as prescribed." It is further noted in footnote 35 that, according to a proof in Peter Griffin's THE THEORY OF BLACKJACK, "the expectation, as derived from an unbalanced count at the pivot point, is independent of the number of cards left to be played." So, what I glean from that is that the Pivot Point should never be tinkered with, which conflicts with the above mentioned supposition in appendix V.

I would sincerely appreciate any additional information and thoughts on this topic. Thanks so much for the helpful information offered thus far.

No, he does not even mention the pivot point. I must confess that I am not a math guru so what follows is just my best guess and I'm more then happy to defer to some others on this board who are much wiser.
Grosjean says in his book that with any running count system the critical index (namely the key count in KO) depends on penetration ie; the number of rounds. And while he discusses various penetration changes affecting the key count he says nothing about moving the pivot point.
Furthermore, as you note Griffin says "the expectation, as derived from an unbalanced count at the pivot point, is independent of the number of cards left to be played."
Trying to put 2 and 2 together, from the above I would assume that while penetration changes affect the key count the pivot point remains constant depending on the number of decks. Thus while a change in penetration to 50% in an 6-deck S17, DOA, DAS game might move the key count to +12 (IRC=0) the pivot would remain at +24. If insurance is taken at pivot minus 1 why would that change if pentration is now something less then 75%? Doesn't the ratio of low cards to high cards depend on the count per decks in play irrspective of where the cut card is placed?
Hope this is of some help to you.

You are correct - the pivot point is not dependent on penetration. This is one of the reasons KO works as well as it does - many strategy deviations happen at the pivot, where the count is most accurate. Also, we want to have our max bet (or close to it) out at the pivot point.

With KO, the pivot point is always equivalent to a Hi-lo true count of +4, regardless of where we are at in the deck.

Should it be lowered along with the key count for games with exceptionally low pen., or should it be left to coincide with the IRC?