yourfirstvisit.net

OVERVIEW: TOURINGPLANS.COM’S “CROWD CALENDAR 2.0”

The crew behind the Unofficial Guide—creators, among other things, of TouringPlans.com, Lines, and The Unofficial Guide to Walt Disney World 2011, have released a new version of their venerable daily crowd calendar.

The new release takes a great resource and makes it even better.

While it has many new features, the most important one is that the crowd predictions on the TouringPlans.com daily crowd calendar now match the most typical reason people look for crowd calendars. That is, the calendar now represents an estimate of general Walt Disney World theme park crowds on a given day.

Users of this site’s guidance on when to go, or its crowd calendar, or its itineraries, don’t need the touringplans.com daily crowd calendar.

But even so, it is a great and very affordable resource to use to double check my thoughts, valuable for all, and especially valuable for those designing their own itineraries.

MORE ON CROWD CALENDAR 2.0

This crowd calendar has long been a staple for Disney World visitors, first in The Unofficial Guides, then as a free service on touringplans.com, and now as a paid–but well worth it–resource on that site.*

While a staple, anyone who talks to other Disney fans or spends time on the Disney message boards understands that there could be confusion between what the old version of the crowd calendar actually predicted, and what many of its users thought it was predicting.

The old crowd calendar predicted average peak wait times for three rides at the Magic Kingdom.

It then converted these from minutes to an index of one to ten–a day rated “7,” in the old calendar, meant an average peak wait at these three rides that day on the order of 70 minutes, a “6” on the order of 60 minutes, etc.

However, many people interpreted the crowd rankings as applying to the Magic Kingdom overall, or to the theme parks generally–e.g. to crowding at Epcot.

This mismatch led to some potentially confusing and frustrating outcomes:

• Times when the crowd calendar was analytically trivially incorrect (for example average peak waits for the three rides may have been 70 minutes rather than the predicted 80, because more ride vehicles had been added than expected), but profoundly correct for how users were using it (yes the parks were pretty crowded that day)
• Times when it was analytically correct, but “incorrect” for a user who was using it to, say, predict crowds at Epcot–which it was not designed to do.

The new crowd calendar solves most of these user and communications issues by explicitly forecasting what many users are looking for–overall crowd levels, based on expected waits for a hundred different attractions in all four parks.

There’s still potential for confusion when parks have different levels of waits on a day, but the gang at touringplans.com have said that they plan to add individual crowd forecasts for each of the four parks as well! This would be a great addition.

UPDATE: as of 7/24 individual park forecasts are available as well!

While making this major and welcome change to what they are forecasting, touringplans.com has also modified the labeling of days–the scale of 1 through 10 they use to indicate a day’s projected crowd level.

• In the old model, each increment (5 to 6, or 3 to 4) represented 10 more minutes of forecast average peak waits for the three Magic Kingdom rides.
• The difference between numbers was very understandable, and veteran users learned to dread the appearance of 8s, 9s and 10s in their planned weeks.
• On the other hand, although 1s, 2s and 3s were theoretically possible, they hardly ever showed up…as peak waits rarely were projected to be that low! People who did not understand this did not understand that, in the old calendar, a crowd rating of 4 was about as good as it got!!

The new calendar takes a different approach to labeling days. Bigger numbers are still worse than smaller numbers, but now the numbers 1 through 10 are equally distributed among the next 365 days.

• The 36ish days with the lowest forecast wait times get a 1
• The 36ish days with the next lowest forecast wait times get a 2, etc, all the way up to
• The 36ish days with the highest forecast wait times getting a 10.

This change enables visitors to make much sharper distinctions among the less crowded days of the year.

On the other hand, in my judgment, it pushes too many days into the group labeled 10, with almost double the number of days in that group as the old calendar had labeled 10.

This matters because the very worst days at Walt Disney World are much worse than the ones that are, say, the 34th and 35th worst days of the year. On the other hand, you should avoid days ranked 8, 9, or 10 under either the old or the new calendar, so maybe it doesn’t matter after all!

(Note that you can’t directly compare the old and new day ratings, because of the old one applying only to 3 rides in one park, and the new ones being based on a hundred rides in four parks.)

What’s lost in this scheme is the easy comparison of numbers–e.g. 3 vs 6.  In the old numbering, this meant 30 more minutes of average peak wait time for the three forecast rides.  In the new model it means something like “moving from the middle of the bottom half of wait times to just above median wait times.”

However, I don’t have a clearly better approach to suggest.  The addition of other parks and many more rides is what makes this particularly challenging…

The approach I take on my own crowd calendar, if used by touringplans.com, would be something like

• Summing all the minutes of waits forecast for a day,
• Substracting from this sum the equivalent number for the the lowest-wait day of the next 365 days, and
• Dividing the result into the highest-sum day of the year (after it also had the lowest sum subtracted)

This results in an index for every day of between zero (lowest of the year) and 1 (highest of the year.)  However, because of forcing the lowest days to zero, you are still missing the simple proportionality of the old day rating model…

The new crowd calendar also includes the traditional “best parks” and “parks to avoid” suggestions.

One of the items added to the decision rules for what park to recommend is the new park-level crowd data.  This is a step forward.

Even so, I’ve never been keen on these “best parks” and “parks to avoid” suggestions, as they do not apply equally well to different types of visitors, and this is not clear from the crowd calendar page.

• For example, they work pretty well for guests without park hopper tickets, without access to Extra Magic Hours, and staying 7 nights or longer…
• But work less well for those with hoppers, with access to Extra Magic Hours, and/or on shorter trips.

For many visitors, individual park level crowd forecasts–when they are published–will be more help in making  decisions about which park to go to when. UPDATE: as of 7/24, individual park forecasts are available. (Or you can use the itineraries on this site!)

Overall, the revision represents a significant step forward, and makes the best daily crowd calendar on the web even better!!

*The first 30 days of the crowd calendar are free–you have to pay only to see the rest of the upcoming year; paying these folk also gives you access to other cool resources of theirs like Lines.  There’s also lots of other great free stuff on touringplans.com besides the next 30 days of the crowd calendar.

MORE ON WHEN TO GO TO WALT DISNEY WORLD