Crowd Calendar prediction levels VS. Historical Crowd levels, which is more accurate in predicting?

I’ve been really noticing the last month many inaccuracies in the data when looking into crowd levels. Bare with me here…

I’m a big believer in doing your own homework and not relying on others. Here is a great example of that. Using the Historical Crowds data, it’s always good to look at your planned trip week 1 year ago and 2 years ago.
Example: Looking into visiting WDW the week of Sunday February 7, 2021. (This is the week before President’s Day week) I’ll focus on the historical crowd levels just on Sunday. (first full day in park)

This is the TP Historical Crowd Level Data for Sunday February 10, 2019:
Across All Walt Disney World Theme Parks:
What we predicted: 4 out of 10
What we saw: 5 out of 10
Not bad right? But look closer, these were the levels at each individual park:
Magic Kingdom: what we predicted: 4, what we saw: 4
EPCOT: what we predicted: 4, what we saw: 5
Hollywood Studios: what we predicted: 4, what we saw: 9
Animal Kingdom: what we predicted: 4, what we saw: 10
So, is the 5, “What we saw” level across all parks an accurate average?
the average of 4,5,9, and 10 = 7, hmmmm…

Now let’s look at the same data for 2020, which would be Sunday February 9, 2020.
The Crowd prediction for 2/9/2020 was level 5 across all parks.
Note, only in the last week or two did TP up the level to a 7 across all parks.
These were the individual Park Levels:
Magic Kingdom: what we predicted: 4, what we saw: 9
EPCOT: what we predicted: 7, what we saw: 9
Hollywood Studios: what we predicted: 6, what we saw: 9
Animal Kingdom: what we predicted: 7, what we saw: 10

Now all this data is from only ONE specific day. But there are MANY days similar to what happened here during the past couple weeks. For example the weeks of Jan 26 and Feb 2 2020, even weekdays that were predicted 4’s (across all parks AND individual parks) turned out to be 6’s or higher. The funny part is, The actual crowd levels on these two weeks nearly exactly the same as 12 months prior in 2019.

Conclusion: Historical Crowd levels at individual parks don’t lie!
Problem 1 is TP may not get serious with their crowd predictions until it’s too late
Problem 2 is TP does not give you the real average of “across all parks” even after the fact
Are you still with me?? :face_with_raised_eyebrow:

1 Like

I started to notice this too when I began planning my Sept 2020 DLR trip. I saw a big discrepancy on predicted CLs for Mondays & Tuesdays.

It appears, in my research, that far in advance they are just using the historical data from that calendar DATE and not adjusted for the day of the week.

For example, my trip starts on a Monday. That day was a Sat. in 2019. So I see a CL10 right now for this date because in 2019 that “date” was a weekend even though in 2020 it is a Monday. If you look at the actual CLs on the “what we saw” during the same week in 2019, it’s indicative of a typical Monday / Tuesday in the fall.

I’ve reached out multiple TP statisticians / staff and none have replied / answer my questions about it. (Even though we “chat” daily on social media about non-TP related stuff)

I love TP but any CL # that is more than 30 days out I no longer trust. Wait and see… it’ll fluctuate. TP only updates once per month and copies & pastes the previous year’s historical info into the farther away future dates.

1 Like

Wow, that’s crazy if true and very careless. You and me could punch in those accurate numbers in 30 minutes. They should hire us!

As many hours as I spend here answering questions & reviewing ppl’s TP, it would be no extra work to do it. :stuck_out_tongue_winking_eye:

When I see new users ask about CLs for 6 months away I have to tell them it’s a prediction, not definite. I always recommend people look at the historical calendars for at least two years worth of data to see when there are slower / better weeks in the months they are wanting to travel, instead of exact dates.

One thing to hear in mind is that the overall CL is not the average of the park CLs. One of the reasons for this is that a MK 5 is not the same as an EP 5, so calculating an average would not be a valid statistic. IIRC it is calculated much in the same way as the individual park numbers, but using a basket of attractions across all the parks.

5 Likes

Agree! If you do it the other way, it’s an average of an average, which is no bueno.

I’m also one who really likes looking at the historical crowd calendars and then adjusting by day of the week/holidays/etc. I always make a big spreadsheet to see how accurate things have been and what to predict for my own selected dates. Just one thing to mention when using those historical CLs is that since CLs are determined by wait times, it is important to take a look at the downtime for the park. If EP was a predicted 3 but 20% of ride capacity was offline and the actual CL was a 7, TP probably wasn’t that far off on their CL prediction, it just couldn’t predict big downtime for rides.

1 Like

I make a spreadsheet too. You kind of have to to see what’s going on. Sure good point, thanks!

PLEASE fix this in the next version of TP or Lines. We booked this trip in February looking for CLs around 4-6, we ended up with 8-10 on nearly every day of a six day trip. Our one day at CL 6 was wonderful, at Epcot.

1 Like

Ultimately, I think this is why I’ve been using Undercover Tourist’s crowd calendar over TP’s when it comes to planning which days to hit which parks many months out.

I think where TP excels (for Disney) is in predicting wait times for rides in the park on any give day as your trip approaches…but when planning the bigger picture, it isn’t quite as helpful.

I do wish there was a better indication of crowds instead of wait times, though. I really do care more about crowds in general, than specifically wait times…particularly at EPCOT, where wait times can be quite low even when crowds are horrible.

But, such data doesn’t seem to be directly available.

I also question the decision to base CL levels on just a small window of park hours, which when people decide to show up can vary. I think average wait times throughout the entire day instead of just, essentially, the afternoon gives me a better picture of how busy a park will be overall. HS predictions are suffering from this now, where a lot of the crowd has shifted to the morning hours due to ROTR, making the CLs seem lower than they are overall.

2 Likes

ryan1, as long as UTs predictions correlate with TPs historical crowd data for specific days at individual parks. I’ve followed UT also, they have some inaccuracies too and to me also not dependable. I’m with you about predicting actual crowds, that’s what I’d like to see too. Good point about the hours used to predict wait times. Does it seem to you more and more are rope dropping than years past in general at all parks?

Sounds like we were there the same time. We also had crowd levels of 7-10 each day (even though predicted much lower). We also had the one much more manageable day in Epcot!

We still got lots done and had a great trip, but am a teeny bit sad the crowds were often so oppressive it took away from just enjoying the general atmosphere.

1 Like

We are looking at these same dates for 2021, and I’m a little nervous now given how crazy crowded it was this year. I know it used to be a historically calm time at WDW, but I’m wondering if there is any reason that has changed. I would do September for a better guarantee of low crowds, but I HATE the heat :slight_smile:. Was there something special happening when you were there that made it busier than usual? Thank you for any insight you might have!

I’ve noticed the exact same thing and on my last trip, I did the same, looked at the past 2 years for the same day (=weekday, not just the date) and it was more accurate for sure.

1 Like

Welcome ichigo! You’ve come to the right place for lots of info. So far overall in Jan/Feb 2020, crowds have been slightly higher than in Jan/Feb 2019. This was to be expected with Galaxy’s Edge being brand new. More people planned winter vacations than other years because of GE, which can also be a factor in higher crowds in every park. If you stick to weekdays for WDW in 1 year, you will be fine, especially with a good plan. I would never go to WDW in September either, mainly because I already work outside and don’t need to be sweating on vacation! I don’t believe anything special happened yet this Feb. But sometimes, leading up to a big holiday weekend, crowds can start filtering in on Thursday already. (I am eyeing up Feb 23-25 next year) Can’t wait!

The past few weeks have been really busy, it seems like. I imagine this is due to a combination of 1) this time becoming more popular for families, people have less of an issue with taking kids out of school to go during the winter; 2) The Pro Bowl in Orlando on 1/26; 3) The Super Bowl in Miami on 2/2; 4) Galaxy’s Edge/RotR opening; 5) MLK weekend falling a bit later this year (was late last year also but previous several years was closer to the middle of the month.)

Has anyone seen any indications that capacity was reduced the past few weeks at all? I think this seemed to be the case about a year ago, resulting in some big discrepancies in the CC.

I don’t think there was anything special, other than Star Wars opening.

I feel like some of January had closer to expected crowd levels, and I was kicking myself, since we debated between the first week of February and the 3rd/4th week of January.

Like I said, we honestly still had a great trip, and did a lot - so it in no way ‘ruined’ our trip. I can just imagine if the crowds were a little less how fun and enjoyable some of the details and atmosphere could have been. My kids just wanted to run a bit free (within reason), and it would have been nice to not have to have them stuck to us like glue walking through hoards of people.

One of our days at animal kingdom the fast pass line for expedition Everest was literally running over the bridge to Dino land. That’s crazy to me lol.

We did two PPO breakfasts which were great to at least get a glimpse of the parks without the masses.

I used TP to plan our trip, we were at WDW on Feb. 9, 10, 11. I had spent many hours on TP to plan our trip for 9, had it all planned out. TP was WAY off on their crowd level predictions, so our plan was “thrown out the window” as we waited in long lines. I wish I had done more research on other sites to determine when to go. The long lines, when we did not have a fast pass, and inability to add fast passes as they were unavailable, spoiled this trip. I realize the lines were not TP fault, but I had relied on their predictions. I found out there were 22,000 people there as part of the National Cheerleader competition being held and sponsored in part by DW. I would think TP would know this. My advice, go to several sites to see predictions, and look at history, and events happening in Orlando when making plans. This was my 5th trip to DW, and my last, I can find better things to do then stand in a 1 to 2 hour line for a 4 minute ride. So disappointing! Pam Ebel

I also look at historical crowds and wait times rather than the crowd calendar.
I did this for my trip over new year and had a wonderful time. I did a lot of attractions everyday (i would say a minimum of 10 per day, including all the headliners and with 7 rides on FoP in a day and a half at AK with only 2 FPs). I did NYE at magic kingdom and did 22 attractions (never waited more than 20 minutes), 2 firework shows, lunch at BoG and dinner at Columbia. Predicted lines were way underestimated, so planning the day in advance with actual waits from past years helped having a wonderful day.
For the following days I toured based on past wait times again and had a great time. Never waited more than 20-30 minutes (except for FoP - but i don’t care cause i love it and wanted to do multiple times- na’vi and everest - this one only because we were with first timers and wanted them to see the whole line, and it still was only 60 minutes).
So, I completely agree with you! :smiley:

1 Like

The cheerleaders are there every year (multiple times during the year) and generally do not have that much of an impact on crowds. 22,000 people is not that many at WDW. There were several factors in play this year late-Jan through the first couple of weeks of Feb, but the cheerleaders weren’t really one of them.