Can you characterize those 21,000 leaves that deviate significantly?

For example, do some tiles predominate disproportionately (I would
guess C, G, and/or O, given your CANISTER bonus).

On Fri, Jan 23, 2009 at 5:06 PM, Eugene Deon <[email protected]> wrote:
> I've been trying to tune my leave estimation strategy by solving for the
> exact values of my commonly considered estimation variables as to minimize
> the sum of squared errors between my strategy estimates and all the leaves
> in Quackle's "superleaves" file.
>
>
>
> So far my variables include only:
>
> -single letter values
>
> -double/triple/quad-letter penalties
>
> -vowel/consonant imbalance penalties
>
> -bonuses for number of tiles in CANISTER
>
> -a couple of letter pair values (QU, YY, IY, FF, ING)
>
>
>
> The results aren't acceptable yet.  21,000 of 148,000 five or fewer letter
> racks have errors over 5.  But I would like to keep the number of terms
> managable if possible.
>
>
>
> Has anyone else done a similar analysis?  Are complete details of Maven's
> letter pair values available somewhere?
>
>
>
> Thanks,
>
> Eugene d'Eon
>

Reply via email to