On 2/25/22 10:09 AM, Fred. wrote:
> By the way, the data shown is average IMP gain /per hand./ I'd expect
> the measure of the effectiveness of a NT range to be its IMP gain
> multiplied by the frequency with which a hand in that range it is
> opened 1NT.
That's a good point, too. The other side of it is you have to subtract
the loss when hands in different ranges are not opened 1NT. The lower
the range, the more often 1NT will be opened, and as far as I can tell,
all you have to do to get what Fred want's is multiply each range by the
relevant dealing odds and add or subtract accordingly.
A significant loss for 10-12 will be that you lose _both_ 12-14 and
15-17. That's another way of saying if you open such weak balanced
hands, all your other ranges are necessarily wider, and that has to hurt
at least a little bit.