Crowd Calendar 2.0: Smarter, More Accurate And Not Scary At All
Exciting things are happening at TouringPlans.com headquarters. Recently we launched a redesigned version of the site. Today, we are eager to give you a sneak peek at the new version of our Crowd Calendar.
The Crowd Calendar is designed to help you answer the following questions:
- What is a good season to go?
- What week should I go?
- What parks should I visit on each day?
So what are we changing?
The “Mountains” scale was great, but we can do better
Our previous approach was to use the peak wait times at the Magic Kingdom Mountains (Space, Splash and Big Thunder) to predict crowd levels. This was popular and well understood, which helped us explain how the old calendar worked. The main disadvantage with the Mountains is that Disney can add and subtract ride vehicles at any time, varying the capacity of the ride and thus the amount of time people have to wait. Comparing the wait times of rides with variable capacity isn’t the best we can do.
We’re now using all attractions to measure crowds
Walt Disney World has over 100 attractions and now we’re using every one of them to produce a single, all-in-one crowd level estimate.
We have a LOT more data
Thanks to our fabulous Lines users and our crack team of researchers, we collect wait times from every Disney World park, every hour of every day. These data – tens of thousands of individual wait times – are analyzed by a professional statistician, who looks for trends and patterns in the data to see how WDW crowds are behaving.
Our new crowd calendar recommendations are made to match those patterns and trends we see happening in the parks. For example, we’ve observed that Tuesday, Wednesday and Friday had the lowest wait times in the Magic Kingdom over the past year, so our crowd calendar recommends those days more than others.
“So what do we do with this NEW Crowd Calendar?”
The Crowd Calendar will look the same, but there are some important distinctions. The biggest one is this: Crowd Calendar 2.0 is based on a percentile rank.
Imagine you have a box of 100 marbles of different sizes. If you sort the marbles from smallest to largest you can divide them into ten groups of ten, according to size. The largest ten marbles get put into Group Ten. The next ten largest get put into Group Nine, etc.
In our case, we are sorting all the possible crowd sizes at Walt Disney World into ten groups. The highest ten percent get a rank of ten; the next highest ten percent get a rank of nine, etc.
| Crowd Level | Old Calendar | New Calendar |
|---|---|---|
| 10 | Peak wait at the Magic Kingdom Mountains is about 100 minutes or more | These are the most crowded days of the year. Wait times for all attractions are at their highest. |
| 9 | Peak wait at the Magic Kingdom Mountains is about 90 minutes | If 10 is the highest, wait times for this day rank a 9 out of 10 |
| 8 | … about 80 minutes | … an 8 out of 10 |
| 7 | … about 70 minutes | … a 7 out of 10 |
| 6 | … about 60 minutes | … a 6 out of 10 |
| 5 | … about 50 minutes | … a 5 out of 10, an average day |
| 4 | … about 40 minutes | … a 4 out of 10 |
| 3 | … about 30 minutes | … a 3 out of 10 |
| 2 | … about 20 minutes | … a 2 out of 10 |
| 1 | … about 10 minutes or less | These are the least crowded days of the year. Wait times for all attractions are at their lowest. |
“O.K. Mr. Smarty Pants, then what does a ‘7’ out of 10 actually mean to me?”
The new index combines the wait times for all attractions but we can give some examples of how the crowd level translates into wait times:
| Crowd Level | Peter Pan’s Flight | Soarin’ | Toy Story Mania | Kilimanjaro Safaris |
|---|---|---|---|---|
| 10 | 75 minutes | 125 minutes | 130 minutes | 55 minutes |
| 9 | 75 | 110 | 120 | 50 |
| 8 | 70 | 105 | 110 | 45 |
| 7 | 65 | 95 | 105 | 45 |
| 6 | 65 | 85 | 105 | 45 |
| 5 | 65 | 85 | 100 | 40 |
| 4 | 65 | 85 | 100 | 40 |
| 3 | 60 | 80 | 95 | 40 |
| 2 | 55 | 80 | 95 | 35 |
| 1 | 55 | 75 | 90 | 35 |
On average, a ‘7’ translates into a 65 minute peak wait time at Peter Pan’s Flight, 95 minutes at Soarin’, 105 minutes at Toy Story Mania, and a 45 minute peak wait time at Kilimanjaro Safaris. These peak wait times combined with the peak wait times for every other attraction contribute to the crowd level ‘7’.
“What does it mean if the crowd level changes for my trip?”
Don’t panic. Our estimates for crowds still remain the same; we’re just expressing the results in a different way. Suppose one of your trip days was a ‘7’ and becomes a ‘4’. The ‘7’ indicated that the Magic Kingdom Mountains will peak at 70 minutes, the ‘4’ indicates that the combined wait times at all attractions rank a 4 out of 10.
“What about the best days recommendations, are they going to change as well?”
The new calendar is more accurate for recommending parks too so we felt that it was important to make some changes. Here is how we will recommend parks.
- Avoid a park if it has Extra Magic Hour morning. Unless you are a resort guest with park hoppers, it is usually not worth it.
- Avoid a park if it is hosting a Special Event. For example: Grad Nights, Gay Days, or Star Wars Weekends
- Avoid the Magic Kingdom on Holidays. Other parks are almost always a better option.
- If the crowd level is lower than the 7-day average then recommend the park.
- Some special cases:
- If there is no recommended park then recommend the park with the lowest crowd level.
- If there is no park to avoid then avoid the park with the highest crowd level.
- If there is a 7-day stretch where a park is not recommended then recommend it on the day with the lowest crowd level during that 7-day stretch.
“How do these new rules affect the recommendations in the old calendar?”
There are 1500 park recommendations on the calendar. Less than ten percent of these are changing from a “park to avoid” to a “best park” or vice versa. Here is a summary of the changes.
- There are 7 cases (i.e., less than one percent of all recommendations) where a park will change from “recommended” to “avoid” during the first six months of the calendar. In the second six months of the calendar there are another 27 cases, but those may change once Disney announces the park schedules.
- Nine percent (9%) will change from “avoid” to “recommended”.
- About 40% will change to or from neutral (13% best to neutral, 13% neutral to best, 11% avoid to neutral and 3% neutral to avoid).
- Fifty percent (50%) of the recommendations will remain the same.
“I was planning a visiting a park because you recommended it. Now it is listed as a park to avoid!”
We strongly believe in giving you the most accurate and up-to-date information possible, even if that means going against previous recommendations. If one of your days went from “best park” to “avoid”, bring along Lines, our Cheat Sheets, and a good Touring Plan. All will help you minimize your time in line. And please contact us with any specific questions.
Coming Soon
We are super-excited about this new calendar because it will give you a broader picture of the crowds at Walt Disney World. Crowd Calendar 2.0 will appear on TouringPlans.com sometime this week. Let us know what you think.
UPDATE!
The new Crowd Calendar has arrived. Go check it out!
Happy planning!
You May Also Like...
-
We have had a few weeks to observe the wait times at Universal Orlando Resort since the parks reopened in June.…
-
Everyone's number one priority right now is to help slow the spread of COVID-19 by practicing social distancing. Disney did their…
-
Our latest Universal Crowd Calendar update was incomplete. We have now updated the dates that were missed. For a description of…
-
Wait times at Universal Orlando Resort are up, especially at the Universal Studios Florida park. We have seen crowd levels at…
I just looked at the WDW crowd calendar for the next 12 months (March 1, 2012 to February 28, 2013) and counted 49 days ranked 10. Shouldn’t there be only 36 or 37 days (10% of a year) ranked 10?
Generally I do not learn post on blogs, however I wish to say that this write-up very forced me to take a look at and do it! Your writing taste has been surprised me. Thank you, very nice post.
i miss being a kid 🙁
Is there a way to determine best times to visit water parks and Disney Quest based on the per park and overall crowd calendars?
Love the new calendar. Been relying on it for years. With that said, what about Universal? If not a formal calendar, are there best days? Or is there any way to guage best days at the 2 Universal parks based on trends between particular Disney parks (e.g, bad days at MK = bad days at IOA).
The are no obvious trends between crowds at the Disney Parks and crowds at Universal. With the addition of the Wizarding World of Harry Potter, Universal’s Islands of Adventure is pretty much packed every day. We have seen evidence however, that Mondays are better than Tuesdays and Wednesdays and the weekends tend to be more crowded in general.
Is the December 17, 2010 Epcot hours correct? TP is showing evening EMH but the Disney website is not. Hope TP is right and love how each park is broken down in the new crowd calendar. Thanks for all the great info!
Not to “toot our own horn” but I would suspect we’re probably right. We use several sources besides the Disney website because we find that it often has mistakes, usual in the form of slow updates. It never hurts to call and ask, however, especially if your plans depend on it.
Toot your horn….Disney website updated today and touring plans had the correct calendar first!
Thanks Adrienne, good to know!
I am with the last couple of posters…there are a few days during our vacation where one park is the lowest crowd- example MK on Thursday, November 11th is 0.6 (the lowest of all the parks!), yet it is listed as a park to avoid. Not sure if I should follow the general calendar or the individual. Clarification would be appreciated! Thanks!!
I think I have figured it out now. Turns out, there was a lot of discussion about this very item when the new per park crowd calendar was first introduced. See this post:
http://blog.touringplans.com/2010/07/23/individual-park-crowd-levels-now-available/
As I understand it, the per park crowd level reflects the line wait time for rides. The overall park rating considers the crowd level in the park as a whole. For example, EPCOT may be very crowded with people attending the food and wine festival with people who are mostly interested in eating, so this might be indicated as a park not recommended. At the same time, there mat be very few people in line for the rides (test track, Soarin, etc.) so the per park indicator may be a low number.
For our family, its all about the rides and so we are using the per park crowd level as our main indicator.
I hope this is correct, and I hope this helps.
Hey Lionel, glad it is starting to make sense – it only took us about 3 years to figure it out 🙂
Just to clarify, all the indexes use wait times as their basis for estimating crowds but our recommendations also take into account touring conditions (things like park hours and park schedules). For an explanation of all that, check out this article.
I’m in the same boat as Burt. I am palnning a trip OCT 2-9 and the overall crowd calendar seems to conflict with the individual park calendar on a couple of days. I’ve tried to sort through the older posts to make sure I didn’t miss an answer already provided, but can’t find anything.
Can you explain the discrepancy?
I can’t make sense of this. Consider Monday, August 23. Hollywood Studios is listed as an “Avoid” on the calendar. But the estimated crowd level is 0.8, which suggests it will be a ghost town. What am I missing?
The new calendar is totally confusing and not that helpful. Maybe you should go back to the old one, which seemed to work well for us.
Please help me understand why the Crowd Calendar says avoid HS on Aug 2, when the Per-Park rated it at 5.1, the lowest for the day? Likewise, why is AK listed as avoid on Crowd Calendar for Aug 1, when it’s rated 6.7 and HS is rated 9.7? Which chart should I follow?
I did a little more surfing of the Crowd Calendar site and found the individual park crowd calendar. On Feb 1st HS is a 10, while the other parks are low. So we should be fine as long as we don’t hit HS.
P.S. Still curious over why HS so high…..maybe more Disney cutbacks and it is the only Fantasmic show for the month of Feb :-).
I have been plannng our next Disney vacation for the end of Jan 2011 (Jan 25th to Feb 2nd). I had used the crowd calendar to pick the best block as we are somewhat flexible. We are getting ready to book and I saw that the crowd levels have changed for our visit. Specifically Feb 1st has changed from a 1 to 8! (Yikes).
Can you let me know the reason for this? I am worried that other days in our block will jump up as well. We might be better off going one week earlier if other days are going to be just as bad. Thanks.
I have a question for August 3rd and 10th they have days that are predicted a 10. I know that summer can busy but how can it become as busy as Christmas day, or a holiday? There is no promotions/holidays on those dates.
Thankyou very much!
I have used the Unoficial Guide on my last two trips to WDW. I planned my park visits based on the crowd calendar. I will be going to WDW in a few days and thought I would printout the crowd calendar when I found it was new and improved. One thing I couldn’t help but notice was the explanation of the recommended days contradicted the actual calendar. You wrote: “we’ve observed that Tuesday, Wednesday and Friday had the lowest wait times in the Magic Kingdom over the past year, so our crowd calendar recommends those days more than others”. I noticed that the Magic Kingdom is recommended on Wednesday, Saturday and Sunday and listed as a park to avoid on Tuesdays. If Tuesday’s have historically had the lowest wait times over the past year, why would it be the day to avoid that park?
with a month into the new calendar, do you find the values for the days to be accurate?
how come morning EMH impacts park recommendations but not evening EMH?
Hi. With the new crowd calendar, once you determine the actual crowd conditions for a particular day, do you then replace your “predicted” crowd levels with the “actual” crowds that day? If so, do you then have to make adjustments to the crowd level estimates for the remaining days for that year, in order to ensure that an even number of days are allocated to each crowd level? It seems that in order to have an equivalent number of days allocated to each crowd level you would have to either: 1) only use the estimates (and therefore not “replace” a day’s crowd level rating with the actual conditions that occurred), or 2) only use actual crowd levels (which would mean that the final crowd levels could only be assigned retrospectively).
Thanks for any insight you can provide here.
Was the new calendar a last minute thing? Or did I miss a prominent heads-up? We almost canceled our vacation because of this!!! The change came up just before we left for WDW- we based on plans on the old calendar, which seemed to indicate roughly May 26-June 3 was the last, best week before Summer Crowds we could do. The new calendar showed our planned week was the WORST WEEK to go. Memorial Day weekend now had 9-10’s. I hate mob crowds and this looked like a disaster. This Calendar change was easily the most stressful thing for me related to the whole trip! Maybe I missed the clues the new calendar was coming… but having bought the book and subscribed to the site I was mighty upset my attempt at using your information for planning went so awry.
Uhhhh so I hope you were able to relax and have fun! I think a sensible rule of thumb is that any US federal holiday with a 3-day weekend when kids are out of school is going to draw crowds.
Sorry for adding stress to your vacation John. Our new calendar presents the crowd predictions in a different way that we felt was more relevant. We didn’t make these changes lightly however because we understand that many readers like you rely on our recommendations. In this case, we felt strongly that the new information provided would, in the long term, be an improvement. We did our best to advice our readers of the upcoming changes and presented a comprehensive explanation in our blog.
Memorial Day weekend does rank low in the list of crowds during Holiday weekends behind Christmas, Easter, Thanksgiving, Presidents Day and a few others so you made a good choice. Summer crowds are starting to pour in already.
Thanks Fred. I agree the new calendar is much better, and people will see the benefit in the long run. Easy to overlook a blog link- I probably saw it but couldn’t realize the significance to my plans.
Actually the trip worked out okay, as we (mostly) stuck to Best Parks each day and (mostly) followed the Touring Plans. Longest real wait was 45 minutes for Soarin (1st ride), and we saw 90% of what we really wanted to over the 6 days.
I used the Lines app on Windows Mobile cellphone a little, and added a few times also. This is great as it effectively lets you add tons of “volunteer researchers” to feed you data. I’m sure you’ll update next years book to include this.
Do you really see the ride times as roughly equivalent or weighted across all parks (except as impacted by EMH and events)?
One improvement would be either a crowd level for each park, or some annotations you could use to flag “Parks to Avoid”. The notes are good (special events at DHS for example) but otherwise we can’t tell WHY a park should be avoided on a certain day. That makes evaluating the relevance more difficult.
Right. But as many schools are still in session until June, and the old calendar showed little or no increase in level (7-8’s) it was hard to anticipate how much Memorial Day would have a real impact.
So which would rank higher on your decision- the guesstimate that a 3 day weekend draws more crowds or the (old) Touring Plan Crowd estimator which showed it would not? That was my problem.
With the much better data in the new calendar, the impact you mention was clearly seen. I didn’t see the new calendar until we were in the car however! Ouch.
under the new crowd calendar, do the recommendations for the water parks change from Saturday and Sunday?
No Julie, the water parks recommendations still remain.
In the Crowd Calendar explanation, you state the MK is typically the park to avoid during major holidays. Yet it is the recommended park for Thanksgiving Day. What’s the story? Thanks!
You are right Paul, this was inconsistent with our general recommendations. This was changed during our most recent update.
On 9/26/10, the crowd calendar recommends Magic Kingdom. However, Disney’s website says there are extra magic hours.
Similarly, on 9/27/10, the crowd calendar recommends Hollywood Studios. Disney’s website says there are extra magic hours.
Is the recommendation correct? Or Did Disney change the EMHs?
You are right Cynthia, this was inconsistent with our general recommendations and was changed during our most recent update. Disney has not changed its EMH schedule for September.
Wow, I just noticed that Sunday of Memorial Day weekend jumped to a 10. That is higher than July 4th. I have heard a lot of comments saying that the Memorial Day holiday was not a bad time to visit. Is there something in particular that has changed that for 2010? Thanks
I like the new calendar… makes sense to me.
We’ve been using the calendar to plan Orlando visits for several years. The times when crowds were off from projections in the past couple of years, I noticed that the bulk of the folks there were non-US visitors. As the dollar moves up or down versus the Pound and Euro, people jump on the plane to visit.
If you are planning a visit, you can use off-site hotel prices to confirm the projections. The second and third week of Sept (our next planned visit) shows lownumbers in Calendar. There are also a *lot* of motels in the area for under $30 – 40 per night.
We don’t stay on property, so I can’t speak on how Disney prices rooms. I do know that when you can get a Comfort Inn for $29/night, the park crowds will be small.
And having used Lines since the beta for Blackberry, all I can say is that it is great. Saves a lot of walking thanks to all of the users that report actual wait times.
If you recall that commercial where the kids were so excited about going to Disney that they couldn’t sleep? That’s US about the Crowd Calendar park-by-park breakdowns. 🙂
yes yes!!!
I was wondering if you guys still incorporate the showing of Fantasmic as to whether or not HS is a best park or park to avoid. And this year, do you think the Summer Nightastic will pull crowds away from Fantasmic? I was curious as to why there weren’t 2 showings of it (Fantasmic) in July and thought that might be the reason. Thanks for all you do!!!
Hi Frank, Really glad to see you’re able to incorporate all the lines data to further quantify the park ratings. Do you think you’ll be coming out w/ a per park rating soon? As mentioned above, we “know” that one or more parks will be more crowded than others on any given day due to EMH or other factors, which means by the forumla’s very nature it is overestimating the crowd level for some, and underestimating at others. I.e. MK=7, AK=3, DHS=4 and EP=3 you then weight the data somehow and post the singular aggregate crowd predictor as maybe a 5. But w/out seeing the 4 individual park rankings, we are limited in interpreting that number … basically using your posted best and worst recommendations as our only guide. I hope you can post the per park rankings as that would be really helpful. (maybe for members only w/ all sorts of caveats about it being new and perhaps on a trial basis?)
Sorry, that’s Fred not Frank!!!
Hi there, Fred! First of all – GREAT job on the new calendar!! 🙂 I love it!
I have a couple of questions, though.
In Sept the crowd score on the 1st is a ONE (woo hoo!), but on the 2nd it goes to a 6 and then back down to a 3 on the 3rd. I was confused because the 2nd is a Thursday and I can’t figure out what would bring crowds on a Thursday that would not be there on Wed and leave by Friday? I am just curious. 🙂
And Thanksgiving Week…is MK really the best park on Tues, Wed, Thurs and Friday? Can we get some info on which are the 2nd best?
I love the new setup!!
I was just going to ask the exact question about Thanksgiving week. The crowd numbers make sense to me. Just not the Best Parks/Worst Parks for Tues, Wed, Thurs and Fri.
Thanks!
Looking forward to the updates. I love fact-based information!
Wishing that the crowds for Dec 18-25 will somehow be diminished.
Also wishful here….Will you be updating the Universal Studios and Islands of Aventure touring plans with the opening of the Harry Potter areas? I know that Disney World and the crowd calendars are your babies…but this is the website for the Unofficial Guide.
Thanks!
One curious thing is the chart titled “What does the crowd level mean?”, that translates crowd levels into actual wait times, approximated. It we look carefully at the chart, it seems that there’s really not much difference between a 4 and a 7 on the new scale. Maybe this means that the distribution is really wide and fat near the center, more like a super-Gaussian than an actual Gaussian. (Sorry for the math geek terminology!)
We were there mid-Sept 2009 and found the estimates MUCH higher than actual wait times. But, it was a welcomed surprise!
Always grateful for improvements to an already astounding product. Lines, for example, is the best of its kind and the crowd calendar has helped us tour on the absolute best days.
The only thing that I see is that the calendar still doesn’t use 1-3 at all. It’s like a bell curve in Geometry in which no one fails the class.
Except in our case the 1-3s would be the teacher’s pet 🙂
There are no 1-3s yet becasue the new calendar won’t be launched on the site until later this week.
So, what does “later this week” mean??? Today? Tomorrow? Saturday? You know we have NO patience out here in cyberspace. I need to know!!:)
It’s hump day! I’m with ya Tim. Anxiously awaiting the new calendar due out “later this week”!
Might be interesting to see the calendar as a percentile system. For example the most crowded day of the year would be the 100th percentile, the median day would be the 50th, so on and so forth. It’s very similar to the system you are going to now, it just breaks it down a little bit further. Plus it sounds like you might already have the data for this.
In fact Charlotte, that is what we have done. In the interest of maintaining the look and feel of the calendar we decided to present it in the same way (using 1-10). Our next step is to integrate more information into the calendar using the percentiles themselves.
Awesome. Disney geeks everywhere are thrilled.
March 31st is a holiday but the only day we will be there that shows less than an “8”…..so although it says avoid MK on holidays, since it is a 5 or 6, isn’t it better than the next day when crowds are an 8? Thanks so very much for the well thought out information ya’ll provide. Even though doesn’t always prove to be exact, still better than no idea at all. Thanks.
Forgot to click followup comments box. ( :
Question.
Does the Crowd Calendar take into account things like whether Christmas is on a Wednesday or a Saturday or whether Thanksgiving comes early or late in November?
Not directly, Allan.
There is no evidence that these subtitles affect the wait times. Since 2002, Christmas and Thanksgiving have always had fairly equally busy crowds.
I really like those pictures in the front of the Unofficial Guide that show the crowds in Fantasyland at 9am, 11am, and 1pm. They really do a good job of concretely showing what one might expect.
It would be great to see illustrations like this of the crowd levels with different pictures for each ranking, 1-10. Personally, I would like to see a picture that represents park opening and another that represents the peak time of day. It would really help me visualize what kind of crowd to expect with the new rankings.
I bet a level 10 day would just be a picture of Len hiding in the tequila cave at Epcot.
I know I’d be happier in La Cava de Tequila on a level 10 day more than any place else in WDW! And I have to say that tequila does sometimes have the power to take a frown and turn it upside down! 😉
What an interesting idea Amy!
We’re also working on a textual representation of the park experience for each crowd level. Stay tuned…
Love this idea – excellent!
I think it’s a really interesting idea, but I’m sensing a chance for confusion. Looking ahead the next year, here are the number of days of each type:
10: 20
9: 13
8: 85
7: 24
6: 63
5: 50
4: 110
1-3: 0
In the new system, it sounds like there will be an equal distribution of every number (about 36 of each number). This means that all the 10s, 9s, and a couple of 8s will turn into 10s. And the old 8s will be spread into four categories (three 10’s, 36 9s, 36 8s, and eleven 7s).
My concern is twofold:
First, if there’s really a difference between an “old 10” day and an “old 8” day, does it make sense to lump some of them into the same category?
Second – if there really wasn’t much difference between the “old 8” days, then does it make sense to now put them in four different categories?
I guess my concern can best be said this way – if there truly is a “lumpy” distribution of the types of days (which appears to be true – lots of “4” and “8” days, not many “9” or “7” days), does it make sense to distribute these days evenly into ten equal-size groups?
I see what you are saying Clark but remember, the previous scale was simply an expression of the peak wait at the MK Mountains. The new scale represents the rank of a particular day compared to all others. Comparing the two scales directly doesn’t really work. the exact same crowd may be an 8 in the old system and a 5 in the new one.
The new site looks great!
Thanks Chris!
I was in in WDW this past March from the 14th through the 17th and the parks were more crowded than I’d seen them in 27 years of Disney visits at various times of the year. All of those days jumped up to level 10’s quickly, which was higher than originally predicted by not only the crowd calendar, but also by Disney (they increased MK park hours mid-week).
Any idea why this seemingly random increase? Do you think mid-March 2011 will really stay at a 6 or 7 as predicted by next year’s calendar, or should I expect a similar leap?
I would love to know the answer to this as well. The difference in a level 7 vs. a level 9-10 is a big deal to me.
We’ll look more closely at this, its a good point. The past two years have seen a lot of last minute adjustments, likely due to the economic uncertainty.
Thanks!
I appreciate the crowd calendar very much. We were there May 4-11, and I felt like the crowds were heavier than predicted. Not just with wait times, but also just the overall number of people that seemed to be in the park. I will say, though, peak wait times vary so much within a single park. We were at Hollywood Studios on Mother’s Day. The wait time for Toy Story exceeded 200 minutes (that’s 4 hours)! But, the wait time for Tower of Terror was only 20 minutes. Star Tours was a walk-on, and the single rider line for Rockin’ Roller Coaster was also pretty much a walk-on. It seems to me it would be hard to make a judgement there about the crowd level. I still don’t quite understand the general recommendation to avoid EMHs. I feel like that is part of what I’m paying for by staying at a Disney resort (and I want to get my money’s worth). Plus, we’ve been in early May both this year and last. The morning EMHs at Magic Kingdom were great, allowing us to ride Dumbo, Peter Pan, and Pooh with essentially no wait. We put off the mountains until later in the day, using Fast Passes for them. The evening EMHs at Animal Kingdom are also fantastic! We would never try to ride the “carnival” type rides in Dinoland unless it was EMH (my five year old loves the dino version of Dumbo), and the wait for Expedition Everest is practically non-existent. We may be one of the few families that loves AK, though, spending two days there each trip. Anyway, just my two cents…
I think the recommendation to avoid the park with EMH morning is for people that won’t be able to be there for it. If you are going to get there for the morning hour and take advantage, it is a good idea to do it because you get so much done in the first hour and the following 2 hours.
However, if you aren’t going to be there for EMH because you choose not to or are staying off property, it isn’t a good idea to go to that park. The reason being is that the lines for the attractions have had an hour to grow exponentially. So while it may be 9am, lines and fast pass return times are already what it would be like at 10am on any other day.
I agree Chris, thanks!
Katherine, do you like the March or the May visits better? I know March is busy with Spring Break, but the weather is supposed to be so nice at that time. In May, we find the crowds manageable, but the heat can get to you. What do you think?
Sorry, that was a typo. We were there May 4-11 this year!!!
To answer your question (sort of), I really like visiting in May. First of all, it really isn’t as crowded as other times (even though I did feel like it was more crowded than predicted). Second of all, the room prices are less expensive, since it’s not a peak time to visit. The weather can be hot…we had a few nice days the week of May 4-11 (high 80s, breezy), but there were also a few very hot, very humid days. When we went in May 2009, we were there during the monsoons and it rained heavily almost every day. This year, we only had rain one day. So, overall, for crowds, weather, and costs, I think May is a great option.
I completely agree about the crowd calendar in early March. I felt like it was consistently higher crowds than the calendar predicted and the “Best” parks were not best.
Interesting Steve, thanks for the feedback. We’ll see if the new rules bring the recommendations more in line with people’s experience.
Major typo here…we were there May 4-11 (not sure how that got past me). But, I still felt like on a few days the parks were more crowded than the calendar predicted.
Great job, Fred (and team). I’m looking forward to using the new calendar on my next trip in December.
I do have one quick question/comment. Does the new method you use of distributing the days over 10 groups end up “forcing” days together together in the highest and lowest categories? A typical distribution would assume some outliers, so I wonder if the distribution of wait times within the “1” or “10” category will be wider than say the “5” or “6” categories? If this is true, wouldn’t the worst “10” day be quite different from the best “10” day? Or are the variances in wait times confined to a small enough distribution that there isn’t much of a difference.
I was also wondering about this with my question. In other words, does the new scale end up distoring a bell-ish curve?
Great points Chris. Looking forward to Fred’s reply.
From the description (so far) it seems like we could think about it as deciles – all the 10s would be at the 90th percentile (of days) and above for crowdedness (overall WDW, or for that park, if there was a park-by-park breakdown). Correct?
Hey Chris,
Thanks for the post. Fred will be along shortly to answer your questions, and he loves this kind of talk!
One thing to keep in mind is that the calendar must naturally have some simplifications to it, as it’s used by hundreds of thousands of people per year, many of whose skills lie outside of math.
I think that beyond a certain threshold, the additional complexity needed to differentiate the far ends of the scale doesn’t help these families. A “10” means “prepare for the worst” and a “1” means “Did we miss an evacuation drill? ‘Cause no one’s here.” I think most people are looking for just that.
I was having the same thoughts as these other folks, Len. The old scale seemed to work on a standard deviation type of situation, with 5 being average, etc.
It will be interesting to see how the percentile ranking shuffles out. In my head, 36.5 days at a level 10 seems horrific, but in actuality it may not be too far from the current number of 10 days on the “old calendar.”
On a side note, it would be a dream come true to be in the park and feel a “Did we miss an evactuation drill?” moment. I dream of open spaces and no lines… 🙂
Great point Chris. At the higher end the variability is greater. The difference between a high ten and a low ten is larger but in our minds they’re both a ten 🙂
Avoid ’10’s whenever you can.
Thanks for the feedback, Fred and Len. Your answers are exactly what I expected. The goal of the calender, though based in statistics, is to guide people when stay away or best to go, not give them a full statistical breakdown of the year or a math lesson. I wasn’t trying to poke holes in anything, just making sure my stats knowledge was up to par!
Anyways, thanks again. The insight/analysis you provide is extremely helpful and much appreciated!
Thanks Chris. We do plan on taking the crowd calendar to the next level however. For those who want more information about the crowds we are working on a way to integrate it into the calendar. Stay tuned…
So if I understood this correctly, approximately 36 days will be 10s, 36 days will be 9s, etc? Does that mean that if there is a change in one day (maybe due to a promotion), another day has to also change to balance it?
Doesn’t this assume that the scale of least to most crowded is an approximately linear progression? Is that the case?
I’m not trying to grill you, just interested. 🙂
The crowds, in general, follow a normal distribution (bell curve) with most of the days near the middle of the scale and a few extreme days at both ends. The 1-10 scale we have is just a starting point to help our readers plan. We hope to add future details which will break down how a particular day compares to others.
By the way, love getting grilled on this stuff, it represents the better part of my spring 🙂
Great news! I am leaving in less than two weeks so can’t wait to see how the new calendar compares. I have been curious because the lowest number during our visit will be on Memorial Day so we wanted to visit MK that day, but it is the park to avoid. Don’t want to waste a low number on an easier to negotiate park so wasn’t sure what to do!
Pat –
I was in WDW last year Friday-Monday over Memorial Day weekend last year and it was great – crowds didn’t seem bad at all. Out of the entire weekend, we found Sunday to be the MOST crowded, and Monday (morning) at the MK to be nearly empty. This could have been due to the fact that MK had EM nighttime hours on Sunday night though. Have fun!
Great to hear that Boots! This year’s calendar has been predicting higher crowds on the Sunday than the holiday Monday itself. I’ve been responding to emails about this for weeks. Nice to hear that my models are actually working as they should. If you don’t mind, I’d like to use your quote in a short blog post.
You can absolutely quote me. My guess is that Monday is a travel day for a lot of people, or a rest day before heading back to work on Tuesday. I can honestly say that the morning I spent at the MK on Memorial Day was one of the most pleasant, least crowded times I’ve spent at that park. (again, I think part of the reason was because the MK had been open until 1 AM the night before…) Saturday of Memorial Day weekend was also far less crowded days I’ve spent at WDW in mid-March over spring break. I would never hesitate to do a Memorial Day weekend in Walt Disney World.
Terrific – thanks guys. I will be there for sure this year and will let you know how it worked!! Boots, I hope your experience is a good omen for us!
Great job Fred! And great timing, Crowd Calendar 2.0 will be most helpful on our upcoming trip. As far as I’m concerned, from a fellow math geek, whenever you can mix a math lesson and Disney, its a winning combination!
Ha 🙂 Living the math dream.
Thanks for the hard work and the great explanation! I’m looking forward to checking out the new Crowd Calendar 2.0 later this week. I think it will really help characterize the highest attendance days much better–before there were many 10’s, some of which were actually more like 13’s. Now only the highest 10% of days are 10’s, and so forth. It seems like this will help quite a bit.
Great point, thanks Elisabeth!
This is great! I do have some reservations about recommending a park because it’s at it’s lowest point for the week. If that’s the case, I think it should be noted in some way. I can say that in my last couple trips (March), it appeared that the parks that were neither “Best” or “Avoid” were the ones with the lowest crowds. Looking forward to the new calendar!
Thanks Steve, great comments. This new calendar is the first step. We hope to integrate this with other important data that will help our readers plan. Showing reasons WHY the calendar says what it says will be a big part of the future.
Thanks Craig. Yes, one of the things this new method will allow is an index by park. This is one of the features that we are working on. Stay tuned…
This would be a great feature and was exactly what I was thinking was needed as I read the article.
Fred, do you have a feel for how much park-to-park variation we might see on a given day, compared to a single overall number?
Given things like EMH, special events, general touring trends, one might think it would be significant. But the data might tell us otherwise!
Thanks for making a good resource better.
I would love to see a crowd scale per park!
Our data shows that the park-to-park variation is actually quite large in some instances. EMH, events and other factors make some parks less than average (2s, 3s and 4s) while others can be above average (7s and 8s). Our “best day” recommendations take this all into account.
One think I’d like to see is a crowd calendar option to show the crowd level for each of the 4 parks, not just a roll-up. That way I could see what the estimated difference is between a recommended and not recommended park. So if the recommended is a 6 but the not recommended is only a 7, I’d still consider the 7 park if it fit better into my overall schedule. But if the not recommended was a 9 or 10 I’d make sure I stayed away from it and rearrange my schedule for the week.
I absolutely agree with this comment. We have the MK scheduled for the 26th of October and the CC listed the crowds as a 1 but it also says to avoid MK because of MNSSHP. If the CC gave levels for specific parks, I would have a better idea of whether or not I should actually avoid MK. Despite that minor critique, I love the new system, genius!
Fred, you’ve outdone yourself here! Absolutely fantastic!!!!
Is the new calendar actually up, or will it be put up soon?
The old calendar still remains. The new version will be up later this week.
Great explanations!
While I’m glad to see an enhancement to the crowd calendar, I’m also a little nervous. We leave Friday for a trip to The World. How comfortable should I be with the recommended parks from the old calendar through the 31st?
Thanks for awesome resource!
Hey Jeff,
You should be very comfortable with them!
Thanks!
Whew! I was starting to think that no one had answered me because the crowds were suddenly going to be much greater with 9s and 10s every day. ha! (like that would ever happen)
BTW – with the statistical discussion exploding around here, I’m glad I paid (some) attention during probability & stats last semester. 😉
Great app just got better! I like being to see other peoples posted times as well. I’m keeping a journal of day time and crowd level for those fairies. Its to determine the best time to see them.
Fred,
EXCELLENT explanation. Very well done. Thanks for all your hard work. Statistics and Disney?! You have the dream job. 🙂
-Jessie
I know! I encourage everyone to use me in that inevitable conversation with the kids: “But why I do I need math? I’m never going to use it.”
Anything that makes the crowd calendar more accurate for every park is great news. Yay!
Yes, we agree. It will be more accurate for each park but it may take us all a while to interpret the new version.
Question about avoiding a park during EMH days – if you’re staying on Disney property do you recommend hitting the morning EMH and then bailing out to a different park in mid-morning? At what point in the day do crowds surpass non-EMH days?
Lets put it this way: the only way that we recommend doing morning EMH is to bail around 11:00am and take a nap/break or hop to another park. Once you get to the noon hour crowds are higher than a non-EMH day.
Thanks for the tip. We did EMH in the MK during peak Easter days and were able to run right into a few of our favorite attractions, which was def. worth getting up early for! But that was easier when staying on the monorail. We’re going back for a cheapie visit (All Star Motel) in August and I’m planning very early mornings to escape the heat, but I’m just wondering about the breakeven point of taking the time to travel to another park.
I completely agreed. We did Hollywood Studios EMH morning on Saturday, May 8. Went on Toy Story twice, Rockin Roller Coaster Twice, ToT, Indy Stunt show, Great Movie Ride and left by about 11:30 a.m. or so, took a nap and went to MK at night.
Yeah, I try to line up a sit-down meal in late morning in whatever park I’ve planned to visit or in a nearby monorail resort as a way to extract the kids from the parks by a certain time and have a little wind-down time. Late Lilo and Stitch breakfast works great for all of us!
No doubt that EMH mornings work great for resort guests with Park-hoppers but those who stay off-site or have single park tickets make up a good portion of our users. We try to make the best recommendations for the most general case.
I just want to see SOME days that are 1s, 2s, or 3s! (And visit then!)
Us too! The ones, twos and threes have always been there but our present scale doesn’t let them show. This new scale, by definition, will have an equal number of days for each scale. The lowest days get the ones and twos and the highest days get the nines and tens.
Speaking of 3’s I see that June 2 is a 3. Is that just because everyone stays an extra day after Memorial day and then leaves? Or is this the day they flush out all of the restroom plumbing and we’re the last to know.
We are arriving in the World on Sat the 29th. Our original plan was to sleep in and hang around the resort on June 2 but I also noticed the crowd level had dropped to a 3. Is that truly the prediction? If so, I will change our plans by taking a break on Tuesday to enjoy the low crowd on Wednesday!
We were at the parks at end of Jan 2010 and felt the score was too high.
Interesting, thanks.
Too low.
Ok, thanks. We are working on other ways to express the crowd levels. We know that people experience the crowds in different ways. Some people visit on an ‘8’ day and say it wasn’t too bad while others will feel overwhelmed. We just try to present a frame of reference.
Thanks Todd. Would you say that those times when it is ‘wrong’ the number is too low or too high?
Yeah the times are always under it seems, never over. The same could be said for ride times on the same days when the crowd estimate is low.
To clarify (can’t edit) – I don’t mean every single current crowd estimate is under, just the ones on the days where it just feels wrong. There are plenty of days where I feel it’s spot on.
This is a nice change. Including all the rides will give a much better feel – there are times with the “old” system where I feel the number is just wrong, hoping this changes that. Thanks!