So I ran freebayes 1.3.1 with all the default parameters on dog dna samples. I am getting this result for one of the samples (although it seems to be occurring in other samples too):

37 20134611 GCCGGGGTGCCGGGGGCC GCCGGGGCC,GCGGGGGTGCCGGGGGCG,GCGGGGGCC,GCGGGGGTGCGGGGGGCG,GCGGGGGCG,GCCGGGGTGC GT:DP:AD:RO:QR:AO:QA:GL 3/3:24:0,22,0,0,0,0,0:0:0:22,0,0,0,0,0:418,0,0,0,0,0:-50.6989,-42.8808,-10.0438,-81.5228,-45.9661,-5.53744,-73.8782,-38.3406,-12.4611,0,-81.5228,-45.9661,-12.4611,-12.4611,-5.53744,-81.5228,-45.9661,-12.4611,-12.4611,-12.4611,-5.53744,-73.8782,-38.3406,-76.9605,-69.3158,-76.9605,-76.9605,-47.6417

I'm only showing the relevant columns. So the genotype call is 3/3, which is NOT correct from looking at the alignment there:

This implies that most of the reads should go to the 1/1 genotype call. And, the AD field looks like this:

0,22,0,0,0,0,0

There are 22 reads that contribute to the 1st alt allele and none to the others. So shouldn't the call be 1/1?