basing on average price of keys at each sale:

nov 19 8am: 2 sold at $21.48, key price was $2.12 (10.132 keys)

nov 19 9am: 2 sold at $21.05, key price was $2.11 (9.976 keys)

nov 19 11am: 2 sold at $24.57, key price was $2.11 (11.645 keys)

nov 19 3pm: 1 sold at $23.09, key price was $2.10 (10.995 keys)

nov 19 4pm: 1 sold at $24.57, key price was $2.09 (11.756 keys)

nov 20 1pm: 1 sold at $22.50, key price was $2.13 (10.563 keys)

nov 21 10pm: 1 sold at $24, key price was $2.15 (11.163 keys)

nov 22 11am: 2 sold at $23.11, key price was $2.15 (10.749 keys)

nov 22 10pm: 1 sold at $21.11, key price was $2.15 (9.819 keys)

nov 23 5am: 1 sold at $21, key price was $2.14 (9.813 keys)

average price of SCM sales: 10.661 keys

i don't think the difference between 10.47 and 10.661 is significant enough to warrant adjusting the lower margin, considering the two recent sales were both much closer to 10 keys than 11

as for SCM sales being the only proof for the lower margin, the lower bound is also based on the 9 key buy order which can be used as primary proof according to the suggestion guidelines (there tends to be a gap of a key or two between buying and selling prices for items with trading volume this low, as can be seen with 9 vs 12.34 keys in the listings snapshot, so it makes sense that the lower bound for an item with a thus-far unsuccessful 9-key buy order could fall around 10 keys). i would find more sales if i could but it looks like most of the volume for this item is on SCM because that's where the reasonable sellers of this item are