[blml] Decimal HCP ranges.

[Nigel]

[nige1]
SUMMARY: By evaluating other relevant factors according to your
agreed methods, you can order eligible hands with a given raw HCP from 
worst to best.  Presumably the worst hands with a given HCP are at the
bottom of each *one point range* and the best at the top for example 
12.0 might be QJ2 QJ2 QJ2 QJ32 whereas 12.9 might be ATxxx ATx ATx xx

[Jerry Fusselman]
There is no chance that Tim means this.  Few good bridge players
really believe that all balanced hands with 13 HCP are better than all 
balanced hands with 12 HCP. Using decimal points, I would say that 
your flat 8-quack hand is worth about 9.1 HCP, and the three-ace hand 
with good intermediates and a five-card suit is worth about 14.2 HCP. 
  It would be easy to give some evidence for that assessment with a 
computer study that shows that the better hand yields about 1.7 or so 
tricks more on average.  Anyway, their difference in strength is 
easily shown to be more than 9/10 of jack.

[Eric Landau]
I like the notation, but I would read the fraction as probabilistic
rather than evaluative.  If we start to ask what sort of features of
one's hand might be deemed to be worth 0.1 HCP, or seek examples that
differentiate "14.2 HCP hands" from "14.3 HCP hands", we merely take
the same problems we have with integer ranges to a more refined
level, where they will be even more intractable.  Since such values
are inherently monotonic, there can be no agreement on even the
roughest of scales as long as you can find two hands and two players
who will disagree over which of the hands is better.  Moreover, the
finer the scale, the more disagreements there will be.

But if a CC notation like "14.4-..." or "14.8-..." meant that one
opened 60% or 20% (respectively) of one's 14-counts, that would seem
to be relatively intuitive and provide some useful (if still
necessarily approximate) information.  If that scheme were in force,
writing "15-17" would suggest that one opened over 95% of 15- and 17-
counts and fewer than 5% of 14- and 18-counts.  It works for me, and
also works (by default) for walruses (although I'm not prepared to
argue that it would work for the bridge community at large).

[nige2]

[A] I agree with Jerry Fusselman that, for instance, good players rate 
some 13 HCP hands as worth less than some 12 HCP hands. For example, I 
concede that if your declare 12.0-14.0 as a notrump range, that does 
not imply that Jerry would open 1N on say QJx QJx QJx KJxx.

But that does not prevent the Walrus version of Tim's scale from being 
practically useful, as explained by Eric Landau.

In practice, IMO, it matters little whether you regard the scale as 
probabilistic or as representing a player's subjective ordering of 
eligible hands within each 1 HCP range. (Except in so far as one 
version may be easier to explain than the other).

IMO decimal notation would promote more accurate disclosure for the 
bridge community at large but only if recommended by the law-book.