Apply suggestions from code review

Co-authored-by: Peter Ralph <petrel.harp@gmail.com>

Author: apragsdale

docs/stats.md

@@ -751,7 +751,7 @@ by [Zhao (2007)](https://doi.org/10.1017/S0016672307008634).
 
 Briefly, consider a pair of sites with {math}`n` alleles at the first locus and
 {math}`m` alleles at the second. (Whether this includes the ancestral allele
-depends on whether the statistic is polarised.) Write {math}`f_{i,j}` as the
+depends on whether the statistic is polarised.) Write {math}`f_{ij}` as the
 statistic computed for focal alleles {math}`A_i` and {math}`B_j`. Then the
 weighting schemes are defined as:
 
@@ -838,6 +838,8 @@ prod_branch_lengths = np.outer(total_branch_lengths, total_branch_lengths)
 print(ld / prod_branch_lengths[0:4, 0:4])
 ```
 
+To compute the average {math}`r^2` for a uniformly chosen pair of mutations, we also
+weight by tree span:
 As with the `"site"` mode above, we can specify the row and column trees
 separately.
 
@@ -933,7 +935,19 @@ One-way statistics are summaries of two loci in a single sample set, using
 a triple of haplotype counts {math}`\{n_{AB}, n_{Ab}, n_{aB}\}` and the size of
 the sample set {math}`n`, where the capitalized and lowercase letters in our
 notation represent alternate alleles.
-
+::::{warning}
+This means there are two common situations in which many or all LD values will be nan.
+These are:
+
+1. A branch-mode ratio statistic computed on less than the full set of samples
+    will always be nan, since part of the trees are ancestral to none of the samples.
+2. A site-mode ratio statistic will be nan at any sites at which there are alleles found
+    in the entire set of samples that are not seen in the provided sample set.
+
+This behavior **may change in the future**,
+because possibly more natural behavior not currently implemented
+would be to ignore the branches/alleles not ancestral
+to any of the provided samples.
 (sec_stats_two_locus_sample_two_way_stats)=
 
 #### Two-way Statistics