This is a great TGIF promotion, but these pre-ordered blocks of cards (e.g. Advent calendars and Feb. 2019) play havoc with the ratings. For example, the average rating of Chloe Lamour's cards within a pre-ordered period is almost a full point lower than those outside of the period.
So, here's my question for members: Would you like @Totem to provide a "kindred spirit" system that predicted your rating for a card based on other users who have given ratings similar to yours in the past? For example, let's say I was considering buying Chloe's "Warm Glow" card (currently with a 3.3 rating?!?). The system would average ratings for that card only from users who had purchased "Warm Glow" and had given similar ratings to mine in the past. Ratings from people who strongly dislike cards I like would be filtered out from that calculation. A card's rating would be different for each person and would be a better indicator of whether they would enjoy it. In fact, this would allow iS to recommend cards to each user based on the cards they had already purchased and rated high. Thoughts?
Totem can just put them on sale randomly...use this algorithm?
1. 0% chance to have discount at < 20 days after release
2. Then 20-90 days they have a 5% chance to show every hour...once a day
3. 90-180 days after release they have 2% chance to show every hour
4. 180-360 days after release they have a 1% chance to show every hour
5. >360 they have a 5% chance to show each day.
So there is a hotspot when people will look for certain cards on sale...after 3 months then most people that were think about the card have ignored the fact it has been on sale nearly 70 times so it goes on sale less and newer card start to take up the slots...so cards with low ratings will still appear and totem doesn't have to worry about artificially low ratings.