Quality Control (QC)

I use Nextclade CLI on my own computer to analyze my data. I tried to use two different reference datasets, nextstrain/flu/h3n2/ha/CY163680 and nextstrain/flu/h3n2/ha/EPI1857216. I found that I get different QC scores for my datasets. The nextstrain/flu/h3n2/ha/EPI1857216 got more bad and mediocre qc.overallStatus. Will the qc.overallStatus be different when I use different dataset? If yes, how do I choose my reference sequence? I would appreciate any insights into these questions. Thanks a lot!