Looking for dates for a future trip and noticed that there are no Level 1s on the crowd calendar? By definition, shouldn’t there be 10% or 36 days with 1 levels. There are also around 70 days with predicted Level 5 crowds. I know that the days are defined by wait times but how are they divided into representative numbers. Perhaps I am understanding this all wrong?? Help from those in the know please.
Originally the numbers worked the way you indicated; there were roughly 36 days of each number. About 6 months or so ago they completely changed the algorithm and quite frankly I have no idea what the numbers “mean” anymore except big is bad. I THINK it forced things into a bell curve; the result should be more days in the 4-7 range and fewer at the two ends. But it feels like black box math that we have to accept in good faith.
Yes, it’s more like a bell curve distribution. And we’re coming up on the 1-year anniversary of Crowd Calendar 4. Here’s the original announcement, which explains the new format.
Ahh… I got it. Now I understand the math or at least the theory of the math. But still, why are there NO #1s. Shouldn’t some days be the least crowded or is there a threshhold for 1s based on minimal waiits that is never reached anymore bc it is much more crowded in general? Just curious.
This may be helpful: http://www.disneytouristblog.com/when-to-visit-disney-world/
I don’t think that it is important whether the lowest value is 1 or not (or 42, for that matter), as long as the predicted wait times for any given number are correct. When CC4 was recently rolled out for DLR and UOR @fred made a Blog post that had some good analytics of CC4 accuracy at WDW http://blog.touringplans.com/2014/08/23/crowd-calendar-4-launched-universal-orlando-disneyland-resort/
Totally get your point and what the OP is saying. Guess the thinking is a bit different w the new cal? Whereas the old cal was a true 10 scale, guess the new cal tries to let users know how crowded the parks feel. So, there aren’t too many 1 feels although of course- relative to the year- there will be days that are a factor of 10 less crowded. Something like that I suppose.
Well, the old CC wasn’t really a true 10-point scale with a meaningful difference between numbers - the lowest 10% were 1s, the next 10% were 2s, etc. The big problem with this was that the true wait time profile was more of a bell curve, so that there wasn’t much of a difference in the wait times of the numbers in the middle of the scale but there were comparatively large differences in the wait times at ends of the scales. Recalibrating the CL assignments so that they more accurately mirror the wait time profile results in more accurate (and stable) predictions, and the differences between numbers would actually be more meaningful.
However, I still feel that there should be an 11 for Christmas and Thanksgiving…
When the change was made last year, were the previous years converted to the current scale? That is, when I look at the historical crowds to compare future crowds to my previous trips, is it a fair comparison? I assume so, as there are not decimal numbers in the historical crowd calendar. Trying to compare Thanksgiving Monday, which is a 9 prediction, to a previous Thanksgiving week trip a couple of years ago when we “felt” like we had the run of the park on the Monday and Tuesday before Thanksgiving before crowds became busier on and after Thanksgiving.
I asked Fred about this once. He said Christmas would be like a 14.
I believe it was converted. @fred can confirm.
Yes, I can confirm, the historical calendar reflects the current methodology so that comparisons are fair.
And yes, Christmas needs its own identifier, specifically Dec 28 through Jan 1. Wait times for those dates stand alone as the highest of the year, by far. They dwarf wait times at Thanksgiving and Easter. We toyed with the idea of labelling those days as “EXTREME Crowds” and using a 1 to 10 scale for the rest of the year.
I’m telling you, you need to take Nigel’s advice and crank them up to 11. Other touring plans can only go to 10, these go to 11.