Crowd Calendar 4.1

We’re making two small adjustments to our Disney World crowd calendar. First, we’re going to use the last 24 months of wait times to determine what each 1-to-10 number represents on our scale. (We currently use all the wait times we’ve collected since 2010 – keep reading for the challenges this presents.) For example, in a crowd calendar update on April 1, 2018, the crowd levels will be based on data from April 2016 to March 2018. We will make Crowd Calendar updates monthly.

We made a similar change to the Crowd Calendar in early 2017. Readers have told us that they prefer faster, smaller updates to those big once-a-year changes. We agree.

The impact of the adjustment will, on average, affect crowd levels less than 1 point from month to month over a twelve-month span. In the short term, that is, you won’t notice much.

Second, we’re going to provide a separate 18-month view of the calendar for those folks who want to start their planning that far in advance. Of course, a lot of things can change in 18 months. But so many people have asked us for this kind of data to get a general sense of trends that we think it’ll be useful to many more.

Why These Changes are Needed

Our crowd calendar measures crowds by looking at the average posted wait times at popular rides during the busiest time of the day. For example, here’s how we measure Space Mountain’s crowd level using the posted wait time outside the ride.

Average Posted Wait Time
11 am – 5 pm
Space Mountain
Crowd Level
Under 42 minutes 1
42 to 50 minutes 2
50 to 58 minutes 3
58 to 67 minutes 4
67 to 75 minutes 5
75 to 90 minutes 6
90 to 107 minutes 7
107 to 124 minutes 8
124 to 140 minutes 9
140 minutes and over 10

These ranges are based on wait times we’ve collected at Walt Disney World between March 2016 – February 2018.

One advantage of this method is that it’s objective and easy to check. Even if you’re at home in Minnesota, you can use the My Disney Experience app to see what the wait time is at Space Mountain and check our prediction.

Growing Attendance and Reduced Capacity

There are a couple of disadvantages to using posted wait times, too. One is that as attendance grows, there’ll naturally be fewer days with low crowd levels and more days with higher crowd levels.

Another disadvantage to using wait times is that they’re susceptible to changes in ride capacity. For example, the number of people who can ride Space Mountain in a given hour depends on how many rocket cars are running on Space Mountain’s tracks. On very busy days like New Year’s Eve, Disney runs the ride at full speed – every car that can fit on the track is on the track, and Disney has enough cast members working at Space Mountain to ensure that every car is filled as fast as possible with as many people as can fit.

However, it’s expensive to run Space Mountain at full speed – it costs money to pay the cast members to run the ride, and Disney has to pay for maintenance, wear, and tear on the ride vehicles. Disney could save money by running only one rocket car at a time on Space Mountain, but the lines would be enormous and customers would be unhappy with the wait.

On days with fewer people in the park, however, Disney tries to find a balance between the cost of operating Space Mountain and the satisfaction of guests who have to wait in line. They do this by reducing the ride capacity, using either fewer cast members, fewer cars, or both.

Since late 2015, we’ve seen Disney World’s wait times increase much faster than (we think) attendance has grown. We think this is because Disney is decreasing ride capacity to save money and improve their corporate earnings. That is, Disney’s betting that it can save more money on ride operations that it will lose from dissatisfied customers who waited 35 minutes for Space Mountain instead of 25 minutes. (We’d bet that Disney is using advance FastPass+ data to make these staffing decisions, because your FastPass+ reservation tells Disney which park you’ll be in and which rides you’ll ride, at specific times of day, two months before you get there. That’s plenty of time for Disney to set the weekly work schedules of its cast members.)

A long-term program to reduce capacity means that our crowd estimates will go up even if the number of people visiting Walt Disney World stays the same.

That’s not good, either. We think our 1-to-10 scale is easy to understand. And we want there to be enough days throughout the year with crowd levels of 1, 2, and 3 to balance out the days of 8, 9, and 10.

A “10” day at Epcot

 

The TouringPlans statistics team takes its work seriously.  We review our accuracy on a daily basis. We even report every day’s performance on the Historic Crowd Calendar pages. These changes will be included in the next Crowd Calendar updates. The Walt Disney World Crowd Calendar will be updated this week. Disneyland and Universal Crowd Calendars will be updated in March. Our 18-month Crowd Calendar will start showing up in March.

OK, we have done our homework to secure the science behind the Crowd Calendar. Now we would like your help with the artistic side. Check out the examples below and select the color scheme(s) you like the best.

Loading ... Loading ...

Steve Bloom

By helping TouringPlans.com continue to reach the most accurate crowd level predictions, Steve finally found a way to meld his training in statistical analysis with a lifelong passion for Disney. He first visited the Magic Kingdom in 1972, just a few months after it opened. Now he enjoys frequent trips with his two kids. At age four his son insisted on wearing cowboy boots to reach the height requirement for Test Track, and his daughter believes that a smoked turkey leg and Dole Whip make a perfectly balanced meal. Even though she doesn't quite get it, Steve's wife is supportive of his Disney activities.

52 thoughts on “Crowd Calendar 4.1

  • February 12, 2018 at 10:15 am
    Permalink

    To me…Red means stop, green means go, yellow proceed with caution! Love that one!

    • February 12, 2018 at 10:16 am
      Permalink

      edit: yellow & orange

  • February 12, 2018 at 10:24 am
    Permalink

    I like the first one, but I would switch green and blue. To me, green means go. But even as it is, it’s great and easy to read

  • February 12, 2018 at 10:25 am
    Permalink

    To me it doesn’t make sense to have white as a middle option, not sure why but my instinct is if you use white at all then it should be the lowest number

    • February 12, 2018 at 1:48 pm
      Permalink

      I agree. I like the use of white; it’s a bit easier to read. However, it seems like it should mean low crowds with the colors getting darker as the crowds get higher.

      • February 13, 2018 at 5:35 pm
        Permalink

        Prefer white at the low crowd end, and not in the middle.
        Which the voting so far reflects…all the white in the middle options have the lowest number of votes so far.

  • February 12, 2018 at 10:33 am
    Permalink

    Too many people misunderstand that the Crowd Calendar is like a crystal ball of what will be happening. In actuality, it is the best bet based on the data they have collected. It is a valuable tool but is that, just a tool.

    I love the dedication and transparency of the Touring Plans team. They noticed a problem, went to work and found logical solutions to fix the issue. They didnt try and hide it or make excuses. Quick turn around also. Most people are just starting to realize there was a problem.

    Thank you Touring Plans for making our planning so much easier!

  • February 12, 2018 at 11:08 am
    Permalink

    “Since late 2015, we’ve seen Disney World’s wait times increase much faster than (we think) attendance has grown. We think this is because Disney is decreasing ride capacity to save money and improve their corporate earnings. That is, Disney’s betting that it can save more money on ride operations that it will lose from dissatisfied customers who waited 35 minutes for Space Mountain instead of 25 minutes.”

    In my sample size of 1, Disney has guessed wrong on this. I went a few weeks ago, during what should have been one of the slowest times of year. The lines were the worst of any trip I’ve ever been on. While it was fun, the value prop was lowered. As much money as you spend on a Disney trip, it seems like a rip-off for them to be increasing lines on purpose. Rather than an annual trip, we aren’t planning on returning for several years- maybe 2020 or later.

    Disney is a business. And so when they make decisions to put their customers after short-term profits, I fight back in the only way they will hear- by not spending my money there.

    • February 12, 2018 at 12:49 pm
      Permalink

      We had a very similar experience. We still had a good time, and we ultimately rode everything we wanted to, but wow, we just were not mentally prepared for the way lines started long, right at rope drop, and then got worse.

      We spoke to a few cast members who talked about capacity/staffing issues. Not just at Dumbo and Star Tours, but even at Everest.

      Mobile ordering and modifying/refreshing for 4th and more FPs did help, though.

    • February 12, 2018 at 8:07 pm
      Permalink

      Ditto. It was crazy crowded 2 weeks ago, especially during what is typically such a low time of year. But to me, the park attendance itself (not just the ride wait times) felt like the increased number. It was *hard* to navigate MK and AK on days that were supposed to be a 3/4 but ended up being a 7/8. So even though they may have understaffed some rides making wait times longer, just navigating the park felt as if it was super busy. So I felt like attendance itself was up and reflected in the feel of the park, as well as the wait times. Just my two cents.

      I also had friends that rescheduled their trip to last week from when they were originally supposed to come in the fall and had to postpone because of the hurricane (since this was supposed to be a “low” time of year, just as their original trip was planned during a “low” time of year). Any thoughts that hurricane rescheduling could be impacting numbers during this other “slow” time of year at all? I don’t know … it can’t have been *that* many people … but it got me wondering.

      • February 14, 2018 at 5:19 am
        Permalink

        My experience was also this! In the past four years, I’ve gone in late January (right after MLK day) 3 times. This past trip seemed to have WAY heavier crowds. Not just the lines, but the crowds when walking. It completely changed the way we had to tour. The cast members I spoke to all had the hurricane theory. That also jives with what Mike on Be Our Guest said after the hurricane- that almost all their guests were able to reschedule.

  • February 12, 2018 at 11:44 am
    Permalink

    Huge undertaking but so appreciated by those of us who like to look far ahead. Indeed re-calculations are needed as Disney has made so many capacity changes.

  • February 12, 2018 at 11:52 am
    Permalink

    As a person that works with information display on a daily basis, I get that red=bad but visually it is a really difficult color to “read” when digesting information.

    All of the shades of red chosen in these examples glow/vibrate and make it difficult to read the text placed on it. I would definitely lean towards a color scheme the is more user friendly and readable…which means either choice w/o red.

  • February 12, 2018 at 1:26 pm
    Permalink

    I like the red, yellow, green, blue in terms of ease of reading- it makes sense.
    Would it be possible- in order to get the most data, therefore being the most accurate- to encourage people to time their waits? I recall a thread a few months back on Chat where a poster asked who timed their waits- and the answer was VERY low. I think if you can improve on that, you can improve on your predictions too. Just a thought.
    Love you guys, thanks for your hard work.

    • February 12, 2018 at 3:07 pm
      Permalink

      We would love for more people to time their waits. I would say for every 100 posted wait times we receive we get 1 actual wait times. It is my fantasy that the app would time users’ wait times automatically.

      • February 12, 2018 at 4:09 pm
        Permalink

        I forget about reporting my actual wait time. Can you put this in a very prominent place where its quick and easy to access as I am ticking things off on my touring plan?

      • February 12, 2018 at 4:50 pm
        Permalink

        I would time my waits a lot more often if there were an option to time waits when using FPP. On some trips I’ll ride nearly as many attractions with FPP than I do without it.

        I actually do note the times I get in line and board attractions and my FPP wait times don’t conform to any sort of set percentage of the posted wait (though, other patterns appear to be evident, most likely based on where your merge with the folks in the standby line).

      • February 12, 2018 at 5:58 pm
        Permalink

        I can’t even count how many times I’ve tried to time my wait and failed. I start the timer, leave the phone in my hand to make sure I stop the timer then get distracted enough that I put my phone in my pocket to use my hands to pick up my daughter or something and forget about the timer until I get off the ride or sometimes even after that.
        Something even partially automatic would be awesome. Or if there was a way to have the app check back with me after a bit. It seems like even if the incoming data had a margin of error that could be accounted for it would be better than nothing at all. Maybe a way to indicate that you’ve stopped the timer at a point that isn’t the boarding process, so if I notice that I’ve left it running as I check my phone to see my next step as I get off the ride I can stop it while indicating that I’m exiting, the length of the ride is pretty consistent so my time could be adjusted automatically. Maybe even a more complex option where I could estimate how many minutes it’s been since I got off the ride.
        I always want to “give back” for the work you guys do and try and contribute but fail what seems like more than half the time.
        I don’t know that I’d do any better if you incentivized it by giving membership discounts per x submitted times but some people might.

      • February 12, 2018 at 6:17 pm
        Permalink

        I’d time a lot more waits if I could time FPP and SR line waits. I timed every real wait I had 3 weeks ago. A handful broke, and I couldn’t post due to internet connection. But I honestly waited standby for only 4-5 attractions/ day

      • February 14, 2018 at 1:38 pm
        Permalink

        I wish I was there more often to time my waits!!!! I will do my best on each of my visits!! Something automatic would be pretty awesome!

      • February 15, 2018 at 7:28 pm
        Permalink

        A companion Apple Watch app would be a handy way to time waits

  • February 12, 2018 at 2:31 pm
    Permalink

    On a calendar I associate ‘white’ with being open. If you are going to use it, I would use it for the 1-2 range.

  • February 12, 2018 at 3:02 pm
    Permalink

    One thought on the potential color schemes: Try running them through a color contrast tool to ensure that folks with low vision can also accurately understand these, and also read the numbers for each day on the calendar. This is a good tool one to try, and there are many others just like it: https://webaim.org/resources/contrastchecker/

    • February 12, 2018 at 3:11 pm
      Permalink

      Most of the color schemes are considered color blind safe.

      • February 12, 2018 at 6:37 pm
        Permalink

        I’m not so sure about these configurations for the color blind. My son is color blind. The shades of green will all look brown to him. The if there is brown or shades of red next to it, they will all look like shades of brown. In your above maps I am specifically looking at the second row color scheme as being problematic and possibly the last row.

  • February 12, 2018 at 3:12 pm
    Permalink

    What concerns me with this news is that recent Disney practices has made the crowd calendar less useful. If the calendar is based upon wait times and Disney can artificially inflate those wait times with lower rider capacity, then there will likely be no difference in my wait times on predicted crowd days in a certain range, say levels 3 through 8 (or perhaps 4 through 7…speaking hypothetically since I don’t have TPs data).

    I think the Lines app and custom plan software may have more potential now than the crowd calendar. Perhaps both are needed to function correctly. I don’t know how the algorithm actually works. What I do know is I am anxiously awaiting the email that updates the levels for my June trip even though it only means that the math used has changed and not the actual predicted attendance this summer.

    • February 12, 2018 at 3:42 pm
      Permalink

      I agree with your thought processes. Using a Touring Plan and arriving before the park opens will save you more time than just picking a day based on the crowd level. With the app you can re-optimize a Touring Plan during the day so it can adjust based on the observed condition in the parks.

      • February 12, 2018 at 4:13 pm
        Permalink

        Since the crowd issues from a few weeks ago, I have been adding 3 to noted crowd level for the dates I am going just as a cushion. Is this something that makes sense or since the new changes, I am wasting my time? Thanks for the updates!

      • February 12, 2018 at 4:24 pm
        Permalink

        Sorry. I wasn’t very clear. What I mean is when making my personalized touring plan, I pick a date with the same hours that has a crowd level of 7 if my day is projected to be a 4. Not sure if I am wasting by my time doing that. Sorry. I am new.

        • February 12, 2018 at 4:56 pm
          Permalink

          What you are doing is not a bad strategy to see what a plan would look like under a “worst-case scenario.” I would not spend too much time thou.

        • February 12, 2018 at 5:15 pm
          Permalink

          I’ve been doing this too (making my personalized touring plan on a level 10 day), but make sure you have the same plan on your actual day too so that you can optimize in park if needed.

  • February 13, 2018 at 10:34 am
    Permalink

    I know you guys are tired of hearing me say this, but you really need to come up with a methodology to take into account the increased crowds at Epcot during the festivals. Most people coming to Epcot on a Saturday during Food & Wine aren’t there for the rides. Many times we have been at either F&W or F&G on a “3” day, however it was almost impossible to walk in World Showcase.

    My suggestion is to add the kiosks to Lines and collect user-timed waits for each kiosk, since Disney doesn’t post waits for the food kiosks. It would take a while to get enough data to be meaningful, but it should help make Epcot crowd levels more realistic during the festivals.

    • February 13, 2018 at 2:45 pm
      Permalink

      They are very upfront with stating that their crowd calendar is based on attraction wait times and not how many people walk through the turnstiles. Their main customers are those who want to minimize wait times at attractions and their algorithm caters to that audience.

  • February 13, 2018 at 5:13 pm
    Permalink

    Personally, I would swap the 1-2 and 3-4 colours around on all of them, so that 1-2 is the most lightest, then getting progressively more intensive.

    • February 13, 2018 at 5:52 pm
      Permalink

      Agree!! It doesn’t “make sense” in my mind the way it is. If you are using different shades of the same color, the lighter shade should mean lighter crowds.

  • February 14, 2018 at 11:36 am
    Permalink

    When will the 18 month calendar be available to view?

  • February 14, 2018 at 12:27 pm
    Permalink

    Just thinking out loud: I’m wondering how much of the Touring Plans information is being used by Disney to affect staffing levels? A case could be made that a “Touring Plan 1 Day” could be slowed down enough to seem like a “4” day with little negative effect on the Guests.
    I hate to feel this way; I guess I just need to grow up when it comes to vacations.

    • February 14, 2018 at 12:56 pm
      Permalink

      Probably not very much. As thorough as TP is in collecting data, Disney has collected much more. Think about how thoroughly Disney is collecting daya about traffic patterns through magic band wearers. TP can only dream about acquiring that much info!

    • February 14, 2018 at 1:24 pm
      Permalink

      What TwoBits said! With Hotel Reservations, Dining Reservations, and Attraction Reservations Disney can predict attendance with very high accuracy.

      • February 15, 2018 at 12:22 am
        Permalink

        Ticket prices aside, Disney makes TONS of money inside the parks on discretionary spending (food, souvenirs, clothing, etc..). One would think they would want their guest roaming the grounds and opening their wallets instead of standing in line. No doubt they have massive amounts of data that supports their operating budgets but it seems like this would be a basic approach to generating more revenue.

        • February 15, 2018 at 10:13 am
          Permalink

          And I’m inclined to think about it this way: if you were able to see every attraction quickly, you would likely leave the park and spend money elsewhere, particularly on food. If you are forced to stay all day due to artificially created long lines, then you are eating more overpriced food and drinking more overpriced drinks and giving Disney more money in those regards.

  • February 14, 2018 at 12:47 pm
    Permalink

    Why not a dropdown allowing the user to select their color scheme? Just a few sets of CSS and a value stored in session/DB.

  • February 14, 2018 at 1:42 pm
    Permalink

    Thanks so much for sharing this information. I am SO excited about an 18 month calendar. I realize that it will be subject to change, but I always begin planning 15-18 months in advance.

  • February 16, 2018 at 10:27 am
    Permalink

    What do you do in the case of outliers? My tracked dates for September 2018 keep dropping with these updates, which makes sense if the new system is based on numbers from the past 24 months, but September 2017 was ravaged by hurricanes and this that wait time data is likely much less than historical average and not predictive of September 2018. What do you do in these cases? It would seem smartest to throw out such outliers and use 2015 and 2016 data.

    • February 19, 2018 at 7:35 am
      Permalink

      Modern modelling techniques like the ones we use for our wait time predictions are not as influenced by outliers. In the case of the low wait times we saw during the hurricanes you could argue that they aren’t outliers at all since we can easily classify them as “Hurricane-level wait times”. Once we label them in our database the models know not to recreate those low times unless similar conditions are met, namely, there is a hurricane. That being said, we suspect hurricanes may continue to be a factor each year.

  • February 21, 2018 at 5:20 pm
    Permalink

    Very appreciative of the science behind and dedication from this website. As an engineer myself, and someone that enjoys making plans (even if we have to stray from them day of) it is comforting to have this website, app and books as a resource for planning my upcoming trip. Wish I had know about this when we headed to Disneyland 2 years ago! Thank you all for what you do!

  • February 24, 2018 at 8:04 pm
    Permalink

    Just got back from a trip and had two of the most brutal Magic Kingdom days (Feb 21-22) we’ve ever had. It wasn’t just the crowds, which should’ve begun to abate from Presidents Day weekend… it was multiple headliners all going down simultaneously that killed us. Not only couldn’t we ride those attractions, but all the second tier attractions got swamped with people who now had nowhere else to go.

    Perhaps this is a bit conspiratorially-minded of me, but I’m curious and you’ve got the data:

    MK experienced little to no disruption for Presidents Day weekend itself, when the parks were rated 9-10. But when crowds were projected to come down to 6-7, multiple outages at the mountains, 7 Dwarfs, etc. Even TTA trains were running with 50% empty cars, orange cones on the seats, 10-15 min waits. Is that a coincidence? Or did WDW get too cute with throttling back staffing and capacity?

    Is MK lately more prone to “outages” on shoulder days just before or after periods of really high attendance? Or should I put my tinfoil hat away?

    • February 24, 2018 at 10:22 pm
      Permalink

      Heh. We’re looking at exactly the same thing. Stay tuned.

      • February 25, 2018 at 10:11 am
        Permalink

        Thanks! On a related note, I’ve also wondered how rides differ in terms of staffing efficiency and fungibility. Which attractions give WDW the biggest capacity with the fewest CMs? Can a ride operator from It’s a Small World get reassigned on demand to help out with a backlog at Splash Mountain?

        And when multiple headliners are down, is high capacity actually a curse rather than a blessing for WDW? Were they slow-rolling us on TTA to keep guests in line somewhere… ANYWHERE… so they’d be out of circulation for a bit?

        So many questions… I’m such a nerd.

        • February 25, 2018 at 3:07 pm
          Permalink

          You are amongst nerds here.

          Holding guests at TTC is definitely a legitimate question. It would be nice if info about their crowd and queue control could be leaked by a cast member who has such knowledge.

          • February 25, 2018 at 9:30 pm
            Permalink

            Actually, I was talking about TTA PeopleMover having all the orange cones on the seats and restricting rider capacity… sometimes they filled only one car per train. I’ve never had to wait 10 minutes for PeopleMover, until this past week.

            But to your point, TouringPlans did mention on Twitter at around 10am on Feb 22nd that huge lines were building for monorails and ferries at the TTC, with no buses in sight. I suppose that’s one way for WDW to manage park crowds at MK…

Comments are closed.