All of them would use this format, although some data don’t apply to some scenarios. For example, we will only have predictions (no observations or same-day adjustments) for future dates.
When we make same-day updates, we update for the remainder of the day (so current/future hours). We don’t predict the past, although the same-day prediction line will show the latest-forecasts-before-the-time-passed for times in the past.
An intra-day update forecast can be triggered be a few different things (like a wait time submission from the Lines app). There’s not a limit on how frequently it can run, although I’m sure there’s a point where that might overwhelm our systems. So don’t try it!
I guess it looks nicer? The dots emphasize the uniqueness of the user-submitted, measured wait times, but you’re right the data behind all the lines on the charts are a bunch of “dots.”
That vertical line indicates a time when the posted wait time was observed to increase by more than 30 minutes. That could indicate that something strange was going on that boosted/dropped wait times.
The Offline band is translucent and should not obscure anything. During offline periods, we should not have posted wait times (that’s how we know it’s offline), so there may be gaps in some lines during offline periods.
Predicted wait times are not generated using a donor date in the past. Historical data are the basis for our forecasts, but we look at lots of different characteristics when calculating forecasts for a given attraction on a given date. The first forecasts we share are a year in advance, so I guess that’s their point of “conception.” Obviously we create them a bit before we make them live.
We constantly monitor how well we’re predicting wait times, and we make adjustments when we believe we can improve our predictions. There’s not a specific schedule for updates. It could be triggered by unanticipated changes to park hours, but it also could be that we’ve got new data that suggests wait time trends have changed for a park or attraction, or that Len found a clever adjustment that improves accuracy.
I realize that although I’ve provided answers I may not have given you the context you’re looking for. To be honest, I struggle to see utility in these charts (how will folks use it to impact their trips?), and that’s one of the reasons we’re asking for feedback.