On 00:20 of Day N I record the KOBIS PS, which are now all for day N and forward. Divide CGV PS by a ratio related to the estimated final CGV ratio of the day (a bit higher, since I assume as a main chain CGV has a higher concentration of presale than end of day business) to estimate overall PS for day N. Then you can divide estimated Day N overall PS by KOBIS to see what % went to a Day N, and subtract from KOBIS to see what quantity is for Day N+1 forward.
In this case, Sat 00:20 KOBIS was 357k, and with 128k CGV I estimate roughly 265k overall PS for Sat. So about 74% of PS for Sat forward were for Sat exactly, and there were roughly 92k presale for Sun forward. By the time of 00:20 Sun, Sun forward presale are 229k, a growth of roughly 150%. About 95% of them have gone to Sun exactly in the form of 114k CGV Sun PS= approx 217k overall Sun PS. That leaves about 12k right now for Mon forward.
Unfortunately, I only added this KOBIS tracking to the sheet starting a bit of the way through F2’s run. If I wanted to use that fact to try to forecast a Mon number before seeing any direct Mon data, I would have to compare to the behavior of F2’s 2nd Mon, and I don’t think that would come out quite right (just for kicks, getting 14k CHV PS and 105k day — Don’t take that too seriously here). The idea is to eventually have a bank of data for KOBIS/overall PS behavior of movies, and then perhaps we can estimate a day or two further ahead for movies in late 2020 or 2021 with decent results.