So essentially any sire's and dam's that perform below the overall average are simply not graded and discarded. and then from there the gradings progressively increase with Grade 1 beving average and Grade 6 being exceptional?

Sounds like a reasonably simple approach but as you've said it could be a time consuming exercise to bring together the results of all the sire's/dam's "children". You'd have to have a database that would allow you to do so, or to complete it manually would be time consuming but rewarding once you have it in place.

The way you've approached this is similar to how i'm approaching my current backing endeavour, find a niche thats not readily available or easily constructed for the public and get it into a profitable methodology.

On a side note, have you or do you utilize the gradings to change your staking?

Just as an example if all race/horse filters are met:

Grade 6 sire / Grade 6 Dam: Back for 12 points
Grade 3 sire / Grade 5 Dam: Back for 8 points
Grade 4 sire / Grade 1 Dam: Back for 5 points

etc etc or some variation of this?

Would be interesting to see (if you haven't already) if this increases your profitability and/or what Gradings offer the best POT. For example a horse that has grade 6 sire/dam may be a little overbet due to its expectional parents and therefore have a lower POT compared to say a horse with grade 4 sire/dam as they are a little less underrated possibly by bettors and offer more value?

Sorry for all the questions, just some ideas i picked up