Norm Wattenberger: REKO Update

[Norm Wattenberger]

1. I finally got around to adding an 8 deck version of REKO. The webpage has been updated.
2. The REKO Surrender strategy has been tweeked for a bit of an increase in performance. I believe REKO now outperforms KO Preferred in LS games.
3. New REKO sims for 1, 2, 6 and 8 decks for 16 rule combinations and all reasonable penetrations have been run and posted in the CVCX archives.
4. The REKO strategy files have been updated and included in CVBJ, CVData and CVCX updates.
5. There is still only one index value in REKO.

The REKO Strategy

[MJ]

For single deck, the webpage states IRC = -1. However, if you look at the Playing Strategy for single deck, it states IRC = 0. Make up your mind already! :-)

MJ

[MJ]

What percentage of SCORE do you think the player would lose by not surrendering 8,8 and 7,7 vs 10?

MJ

[Norm Wattenberger]

Thanks. Fixed.

> For single deck, the webpage states IRC = -1. However,
> if you look at the Playing Strategy for single deck,
> it states IRC = 0. Make up your mind already! :-)

> MJ

[Norm Wattenberger]

Depends on decks and penetration. At high penetration little difference. Poor penetration a few percent.

> What percentage of SCORE do you think the player would
> lose by not surrendering 8,8 and 7,7 vs 10?

> MJ

[fatcat519]

1) The linked website IRC's, except for six decks, are all one point different from those in the KO book. Is there a reason for this?

2) I've been using RA indices from BJA3. The 10v10 Hi-Lo RA index is 7, a much greater change from the EV-Max index than any of the others. In an eight deck game, I estimate this 7 to be about KO IRC+41, which would put it at RC=13 with IRC=-28. This seems a long way from the reKO general index of 2. Should I remove the 10v10 for RA play?

3) For eight decks, is there any harm in keeping the 12v4?

Thanks, fc

[Norm Wattenberger]

> 1) The linked website IRC's, except for six decks, are
> all one point different from those in the KO book. Is
> there a reason for this?

Either + or - 1. This was necessary to make indexes all the same value for any number of decks.

> 2) I've been using RA indices from BJA3. The 10v10
> Hi-Lo RA index is 7, a much greater change from the
> EV-Max index than any of the others. In an eight deck
> game, I estimate this 7 to be about KO IRC+41, which
> would put it at RC=13 with IRC=-28. This seems a long
> way from the reKO general index of 2. Should I remove
> the 10v10 for RA play?

No, the indexes selected were tested for optimal SCORE which takes into account risk.

> 3) For eight decks, is there any harm in keeping the
> 12v4?

It slightly reduces SCORE for 8 decks. Not enough to worry about.

[MJ]

[MJ]

> Depends on decks and penetration. At high penetration
> little difference. Poor penetration a few percent.

Wait, do you mean a few hundreths of 1% or a couple full percentage points?

MJ

[Norm Wattenberger]

SCORE drops by a few percent.

[Norm Wattenberger]