Thrill-Data defined?

[Opie100]

Pardon my ignorance, but what is the definition of the following thrill-data times?

Typical wait
Posted wait
Disney forecast

In other words, what should I use when planning for the most likely actual wait time? The day before posted wait? Today’s forecast or typical?

 

[lentesta]

I don't think thrill-data has actual wait times (i.e., the time that people actually spent in line).

Typical wait looks like it's the average wait for the attraction based on all previous wait times. For example, the "typical wait" at 2 pm for Smuggler's Run looks to be 61 minutes. So the average wait at Smuggler's Run at 2 pm is 61 minutes, regardless of day, month, or season.

Posted wait looks like it's whatever My Disney Experience is reporting as the most recent wait time.

Disney forecast is the internal wait-time estimate that Disney shows in the My Disney Experience wait-time bar charts for the attraction.

The difference between what thrill-data shows for "Disney forecast" and what MDE shows is that MDE typically just shows the bar graph without providing a Y-axis label for the minutes.

For example, here's how MDE is showing Disney's forecast for Rise of the Resistance for today, starting at 8 a.m.:




Note that the bar chart doesn't display a y-axis label to indicate what wait times those bars represent. (That is, there's a blue bar at 2 pm, but we don't see anything that indicates how many minutes that is supposed to be.)

However, the internal data feed inside My Disney Experience *does* say that it's minutes. Here's part of what that internal data feed looks like:

"forecasts": [
{
"percentage": 68,
"timestamp": "2022-01-01T19:00:00Z",
"accessibility12h": "2pm High",
"accessibility24h": "14:00 High",
"forecastedWaitMinutes": 170,
"timeDisplayValue12h": "2pm",
"timeDisplayValue24h": "14:00"
},
{
"percentage": 68,
"timestamp": "2022-01-01T20:00:00Z",
"accessibility12h": "3pm High",
"accessibility24h": "15:00 High",
"forecastedWaitMinutes": 170,
"timeDisplayValue12h": "",
"timeDisplayValue24h": ""
},
{
"percentage": 66,
"timestamp": "2022-01-01T21:00:00Z",
"accessibility12h": "4pm High",
"accessibility24h": "16:00 High",
"forecastedWaitMinutes": 165,
"timeDisplayValue12h": "4pm",
"timeDisplayValue24h": "16:00"
},
{
"percentage": 72,
"timestamp": "2022-01-01T22:00:00Z",
"accessibility12h": "5pm High",
"accessibility24h": "17:00 High",
"forecastedWaitMinutes": 180,
"timeDisplayValue12h": "",
"timeDisplayValue24h": ""
},

So that data feed says that the 2 pm bar in MDE represents 170 minutes of standby wait. So does the 3 pm bar.

The 4 p.m. bar represents a standby wait of 165 minutes, and the 5 pm bar is 180 minutes.

You can calculate what Disney thinks the maximum wait is by dividing any of those numbers by the "percentage" field divided by 100:

180 minutes / 72% = 250 minutes

So Disney thinks the max wait for Rise is somewhere around 250 minutes. I think that might change daily - I haven't looked at it in a while.

And yes, it looks like Disney overestimated the wait at Rise at 1 p.m. Disney's forecast was 170 minutes and the posted wait is 135.

 

[Opie100]

Great post, thank you! Do you think typical wait is the average of (a) posted wait times or (b) Disney Forecast? Also, what generally do you use when planning?

 

[lentesta]

Do you think typical wait is the average of (a) posted wait times or (b) Disney Forecast?

I think the internal MDE forecasts are generally less accurate than the current posted time, when it comes to trying to figure out how long you're really going to wait in line. The MDE forecasts seem to be created once per day, and don't seem to change during the day, even if the ride breaks down.

In general, a rough, ballpark estimate is that you should wait somewhere between 60% and 110% of the posted wait time. Less at the beginning and end of the day, I think. Rise is harder to predict because it breaks down so often.

Also, what generally do you use when planning?

This might sound funny, but I'm Len Testa. I co-author the Unofficial Guide to Walt Disney World and started a website for this kind of thing. My Masters thesis is on computer algorithms for minimizing waits in line at theme parks. I wrote the touring plan software for the Guide and TP. I usually use that, unless I'm testing someone else's advice. Or, you know, just winging it.

 

[Opie100]

Awesome, thank you! I’ve been generating plans this week (and over the years) and love your software.

Related question: Today (1/1) at 7:45am thrill-data had a posted Flights of Passage wait of 50 mins. For my plan on Monday (@ 7:40am) the TP app says est. wait is 9 mins for FoP. Not the same exact day, but is the difference largely due to Disney overestimating their posted time? Trying to decide if we should buy an ILL pass.

 

[lentesta]

We had an actual wait today at Flight of Passage at 7:48 AM of 29 minutes, off a posted of 75 minutes. I'm not sure where the difference in posted wait times early in the day between us and TD comes from - I thought we were all getting it from MDE. We all seem to agree on posted wait times at 4:30 p.m. (i.e., now).

Monday should be somewhat slower than today - it's the first work/school day after the holiday, so lots of people will be back home. And I don't think the Marathon crowd arrives quite that early.

While I'm not sure I'd bet my favorite pair of socks on the wait being exactly 9 minutes, my guess is it's going to be 20 or under, and that's going to be close to the best you can do on FoP for that day. If you plan to get to the park before opening and ride that first thing, I wouldn't buy an ILL for it.

Also, I suggest re-optimizing your touring plan immediately after your first ride. Once the park opens, we switch the plan's wait-time forecasting to a different, intra-day model that uses "today's" wait times to make forecasts for the rest of the day. So even if the days-in-advance estimate is off, the day-of estimate starts to correct once we begin getting waits in the park.

 

[Opie100]

Sounds good, thx again! This is very helpful. FWIW: TP anticipates that AK is a 7 overall today and a 10 on Monday.

 

[lentesta]

Sounds good, thx again! This is very helpful. FWIW: TP anticipates that AK is a 7 overall today and a 10 on Monday.

Man, I shouldn't second-guess our stats team. Crowds have been somewhat lower than we expected, primarily Omicron-related I think.