Why do the offer numbers for Tokyo differ between Mr. Maezawa’s figures and the back office?

Maezawa here with an additional answer. The discrepancy was caused by different filtering criteria. My figures were based on the number of dates women who joined in April had during their first month (the month after registration). The other figures were based on the number of dates all female members had in April (April 1–30), regardless of when they joined. That’s why the numbers were different. Sorry for the confusion.

Maezawa here. Huh? You’re right! Hmm, I wonder why. This is still just speculation, but maybe the conditions that were set or the way the data was extracted were different. That said, I don’t know exactly what differs at this point, so I’ll ask what settings were used to produce those figures. I expect the answer will come tomorrow, so please bear with me until then. For reference, the settings I used were to extract data for one month starting in April. I defined “one month” as follows: For example: - Someone who joined on April 1 was counted through May 1. - Someone who joined on April 25 was counted through May 25. I set up the function to extract each person’s data accordingly. So, people who joined in the latter half of April don’t actually have a full month’s worth of data yet, since it’s still early May.