This is a good example for the so-called “stable error”.

According to the evidence, this character is only used for the 反切上字 of 𨺙 U+28E99.



Yi Bai provided two variants of 𨺙 U+28E99, one is 陧, the other one is 倪. 陧 is a 入聲字, that is not related to 計 here; the 韻 of 倪 and 計 are both 齊 and the tones are both 去聲. The evidence shows the word 陴𨺙 that must be the variant form of 俾倪. The most common 反切 related to 計 for 倪 should be 五計反/切, but the submitted character must not be the variant of 五, so it is better to keep as-is in main set. At least, we can know the 聲母 of the submitted character is 疑母, that matches the rationale for the phonetic element, so the radical assignment is also right.