A New Metric to Track Theme Park Crowds
I've been analyzing theme park wait times professionally for over two decades. I started at TouringPlans, built crowd prediction models, dove deep into the mathematical patterns that determine whether your Space Mountain wait will be 20 minutes or 2 hours.
Now I'm wondering if there's a better way to represent the wait times that we observe in the parks.
Free daily crowd forecasts for 12 theme parks — powered by the metric you're about to learn about.
The Problem with Crowd Calendars
The crowd calendar that we built at TouringPlans — and many others have copied — uses a 1-to-10 scale, for obvious reasons. It's easy to understand. It closely mimics other rating systems we're all familiar with.
But it comes with some real challenges:
- You need to decide which attractions go into the bucket that drives the calendar
- You need to adjust the thresholds of what each 1-to-10 value means over time
- You have to make sure the values are distributed in a way that makes the calendar usable
These are all valid concerns, and worthwhile to maintain for a crowd calendar. But now that I'm on my own, I'm wondering if it's time to try a different approach.
Introducing the Wait Time Index
The Wait Time Index (WTI) represents the average wait time across all operating attractions throughout the day. One number. No context needed. No thresholds. No bucket of attractions. No changing what the numbers mean over time (that's a big one).
It's an interesting idea, and one that's come together quite well over the last few weeks as I've tinkered with it.
What I've noticed is that the parks become like cities on a weather forecast. 40° in Minnesota means something different than 40° in Miami. Same thing with the Wait Time Index — the range of values you'll see at Magic Kingdom is much different than, say, Disney's Hollywood Studios.
That's because Magic Kingdom does a better job distributing its guests amongst its many more attractions than the Studios does.
What the Data Has Been Showing Me
Once you start looking at parks through this lens, some really interesting patterns jump out.
September 8-22 really is Disney paradise. I always knew this, but seeing Magic Kingdom consistently hit WTI 9-10 during those weeks makes it visceral. Space Mountain averaging 10 minutes instead of 45+. Seven Dwarfs Mine Train becomes doable without Genie+. It's not just "less crowded" — it's a fundamentally different park experience.
Each park has a real personality. Epic Universe maintains more consistent wait times throughout the year — fewer extreme spikes, but also fewer completely dead days. Magic Kingdom swings much more wildly. Universal Studios shows different seasonal patterns than the Disney parks entirely. You can see all of this clearly when the data isn't squeezed into a 1-to-10 box.
The hidden gems become obvious. A traditional crowd calendar might rate two days as "Moderate," but one of those days could be WTI 14 while the other is WTI 22. That's the difference between a great day and a frustrating one — and it gets lost when you round everything to a single digit.
The Year View
I've been mapping full years of WTI data as color-coded heatmaps, and honestly, it's kind of mesmerizing. You can see 365 days of data for any park at a glance on hazeydata.ai.
Blue paradise in September. Deep red zones around Christmas and Easter. Those hidden gems in early February that most crowd calendars undervalue because they don't account for how each park specifically operates during those weeks. When you see 12 months at once, the patterns are impossible to miss.
How Do I Bring This to You?
So now the question is: how do I make this useful for the Disney trip planner?
For now, we're using Discord. If you join the Discord (100% free), you can get information about the crowds that have happened, the crowds happening today, and the crowds that will happen in the future. All of the predictions use actual wait times and the Wait Time Index.
A few commands to get you started:
/today — Current WTI across all 12 tracked parks
/crowd — Detailed forecast for any park, any date
/best-day — Find the lowest wait times in the next 7, 90, or 365 days
/ask — Ask anything about crowds and wait times, powered by AI and real data
Everything is free during this open beta. No subscriptions, no paywalls — just good data and a growing community of people who care about this stuff. The predictions are improving daily as the models learn and the data grows.
I'm not trying to compete with existing crowd calendars or claim I've figured it all out. I'm just exploring whether there's a better way to think about theme park wait times — and so far, the data is telling me there might be.
Explore the data at hazeydata.ai or join the Discord for free daily forecasts.
📡 Data Sources — Our models are trained on data from TouringPlans, Queue-Times, and Thrill-Data. The models, techniques, and predictions are entirely our own.