[QUOTE=Simkie;6859346]
Question for you, JB: if I have a horse that tested Aa through Animal Genetics, is it possible that she could be AAt? I ask purely for my own curiosity :)[/QUOTE]
Nope
“At” tests as just plain old A with Animal Genetics/UC Davis/Any standard Agouti test. So does A+ (wild bay) as well. All three are versions of Agouti, and the standard Agouti test just tests for the presence of Agouti.
So it’s possible the mare could test as Ata rather than Aa with PetDNA… Unless she’s phenotypically bay, in which case she’s definitely Aa 
DancingFoal, as Simkie said, there is no way possible that the foal has a 33.3333% chance of ANYTHING. It’s just not possible. The calculator is wrong. The foal has a 50% chance of being red based, period!
Me personally, I’m either looking at a 100% chance of brown or 75% brown/25% black. Mare is tested EE, and she’s brown, so EE At-. Her sire is registered as bay and has produced black offspring, so if he was truly bay, he was Aa, and could have only contributed his “a” to the mare. So most likely, without having her tested for Agouti, she’s EE Ata. Stallion is by a homozygous black (EE Ata as well) sire, and is brown himself. His dam is brown, out of two brown parents…she’s obviously Ee Ata as well as she’s produced both chestnut and black. Thus stallion could be EE AtAt, EE Ata, Ee AtAt, or Ee Ata. Interestingly, he had his first foal last year and I’m almost positive it was wild bay (out of a chestnut mare)…So, if wild bay is recessive to brown, that means the stallion MUST be Ata because that foal would have to be Ee A+a in order to be wild bay.
So you can often research enough to figure out the genetics. If you’re crazy at least 