Attraction Rankings

[jackdude101]

Here is an attempt I've made to grade locations with roller coasters, steam railroads, and classic carousels using objective data sets to see which places are the best. Typical weather conditions are also taken into account. There are 592 locations in the USA and Canada on the list. I'd add other countries, but it's difficult to get all the data necessary to compile the scores for every country, especially if the country is impoverished or non-English-speaking.

Presently, the roller coaster data is made up of roller coasters worldwide, the steam railroad data is made up of steam railroads in the Anglosphere (USA, Canada, UK, Ireland, Australia, and New Zealand), and the classic carousel data is made up of classic carousels in the USA and Canada. If anyone happened to know of a good online database for classic carousels outside the US and Canada, I'd easily be able to expand the Attraction Rankings master spreadsheet accordingly. I'll post details about how the scores are calculated if asked.

I should also note that when discussing this scoring system in the past, people have gotten bent-out-of-shape because I referred to it as "scientific," so I won't refer to this project as "scientific," even though the data used is objective.

 

[BigBad]

I'll post details about how the scores are calculated if asked.

Go 538 on us!

(Nate Silver predicted a bunch of US elections a few years ago. He runs a website called 538 that goes into the statistics behind a lot of life, ranging from figuring out how many people have ever died to why the Packers should have gone for 2 against the Cardinals.)

 

[jackdude101]

Roller Coaster data has six criteria:
Original opening date (even if it has been moved to a different park). Scores increase the older or newer it is, and decrease as it gets closer to the average age.
Track height (taller the better)
Track length (longer the better)
Max. Speed (higher the better)
# of inversions (more the better)
Ride comfort. This is a more complicated criteria that mainly targets the rough rides of old tech wood roller coasters (the older and faster, the more rough the ride will tend to be and hence the score will be lower). Steel roller coasters and new tech wood roller coasters are included in this, too, though their roughness due to age will increase at a much slower rate.

Steam Railroad data has five criteria:
Original opening date (older the better). Here, I am referring to the track bed, not the track itself. Sometimes you may be riding a heritage line that was built only a few decades ago, but is built on a former track bed of a line that was built over a century ago.
Main line length (longer the better)
Track gauge (wider the better)
# of operational steam locomotives in service (more the better)
Build year of oldest operating steam locomotive (older the better)

Classic Carousel data has four criteria:
Original opening date (older the better)
# of rows of horses/animals (more the better)
# of operational band organs (more the better). These were standard issue on many carousels a century ago, but are rare now.
# of brass ring dispensers (more the better). These are even more rare, and the carousels that still have them tend to be true-blue treasures in the amusement world.

Weather data has two criteria:
Annual average temperature (higher scores are the ones closest to 72 degrees F [22.2 degrees C])
Annual total rainfall (higher scores are the ones closest to 31.5 inches [800 mm])

Each individual data set is graded on its own bell-curve and then converted to a 0-10 scale. The total score for each type of ride and the weather equals the average of their respective data sets (average of the six roller coaster data sets, average of the five steam railroad data sets, average of the four classic carousel data sets, and the average of the two weather data sets). They are then totaled by park. On the Attraction Rankings spreadsheet, the master overall score is the sum of the ride type scores and weather scores for each location (the ride type total scores by location are converted to a 0-10 scale prior to the overall score calculation). This effectively treats a location's roller coaster assortment, steam railroad assortment, classic carousel assortment, and weather quality equally; and it "penalizes" a location for not having a good variety of different rides of these types. Under this scoring system, the two American Disney Resort properties (when all of the scores for their respective parks are added together) are #1 and #3 on the list. These also happen to be the #1 and #3 most-visited amusement areas on Earth (#2 is Tokyo Disney Resort, BTW).

That's mainly what I was going for, to see if I could explain a park's popularity in terms of numbers related to key rides and weather, and not more common basic systems that rank parks based solely on arbitrary poll numbers or the sometimes biased opinions of critics.

 

[Hyde]

Interesting, and thanks for walking through some of the methodology.

On weather, are you controlling for when a park is open? Since seasonal parks are obviously not open during the cold months, you may be inadvertently penalizing them for weather that occurs when they are not operating. For instance, taking a weather data set from April to October would better reflect realistic temperatures during park operation.

I also wonder if there may be an opportunity for incorporating roller coaster rankings into roller coaster scoring. Obviously Golden Ticket, Mitch Hawker, etc. have their own methodology and flaws; but it would also help capture the essential good (and bad) of a roller coaster that goes beyond statistics. E.g. While Millennium Force trumps Maverick statistically, many would agree Maverick is the better roller coaster. Obviously would need to think through how to account for bias in polling, but still worth broaching, especially depending on the sample size.

Also glad that you are taking into account ride type. Will be interesting to see what kind of return this factor has.

 

[jackdude101]

Before adding the weather score, I was planning on making "days open per year" one of the criteria, but that quickly became problematic as not all parks provide that information, or at least not thoroughly. For instance, with Six Flags Magic Mountain, they only list their calendar dates six months into the future. Also, with the Henry Ford Museum & Greenfield Village, they state that they are closed during December, except "on certain dates" for special events (this is referring to their old-timey Christmas setup). They of course don't say what days those are or how many days. Plus, there is also the issue of how many HOURS during a particular day that a park is open. A park could be open 360 days a year, but only open from 12 PM to 4 PM, for instance. There are also a number of extremely small places on this list that may not post anything at all.

Anyway, the temperature data is a sort of "substitute" for the days open data. If the average annual temp. is lower it implies that the park is not a year-round destination, and the lower score from its temp. stat will reflect that. Also, they are only penalized once for having cold weather, as the rainfall score only includes rainfall and not snowfall (apparently, the people who collect this data are able to tell the difference).

As far as including popular polls like Golden Ticket Awards and Mitch Hawker into these rankings is concerned, I'm going to have to decline. The primary purpose of this ranking system is to at least attempt to create an alternative to the very subjective "which ride is your favorite?" types of rankings. I suppose a good comparison would be the difference between rating a movie based on the number of awards it has received versus its box office performance. My stuff is more closely related to the latter.

Lastly, I've ridden all the Cedar Point roller coasters at one time or another, except their tiniest one (because I'm too tall to ride it, lol). Millennium Force was my favorite, but I actually only partially liked Maverick, due to its tendency to smack your head against the sides of the restraints. Its sister coaster at my home park, Busch Gardens Tampa, does not do that and is much better, imo.

 

[jackdude101]

I was able to find a semi-legit online resource for British carousels (http://www.fun-fairs.co.uk/forum/69-gallopers/) and one for Australian carousels (http://carousels.org/australian.html). That was the one piece missing from my data that was preventing the non-North American English-speaking countries from getting on my master Attraction Rankings spreadsheet. They are included now in the new Top 30 below.

 

[jackdude101]

...and here's the Top 30 for the UK & Ireland only. The steam trains tend to dominate this list, but that's just because they are the type of attraction that the Brits seem to excel at the most. I can say with great confidence that Britain is the King of Steam.

 

[ECG]

Taller ≠ better, faster ≠ better, longer ≠ better and that Cedar Point coaster ranking proves my point. Using that criteria to rank coasters might not be not subjective, but is flawed - regardless of the other criteria you brought into the equation. You can't take some of the most liked coasters ever built and ignore their popularity.
There's a reason that Maverick has a constant 2-3 hour wait and it isn't due to capacity or technical issues (like TTD, for example). It's because it's one of the top 2-3 coasters in the park. And comparing the ride comfort to Cheetah Hunt at Busch Gardens in Tampa is ridiculous. If Cheetah Hurt had a similar layout it would feel the same, but Cheetah Hunt is a family coaster and Maverick is anything but. :roll:

 

[GuyWithAStick]

Yeah, stats usually mean nothing when it comes to roller coasters. Kingda Ka, the tallest and America's fastest, is usually considered crap. Phoenix at Knoebels, which is tiny, is said to have the best airtime of any coaster ever. You really can't rank something subjective. It's like ranking the best car, it's different for everyone.

Sent from my VS820 using Tapatalk

 

[jackdude101]

Cheetah Hunt is a family coaster

You lost me at that point. A roller coaster going 60 mph is not a family coaster. Anyway, rating roller coasters based on stats isn't perfect, but they are a strong "forecast" for how good they are.

 

[jackdude101]

Yeah, stats usually mean nothing when it comes to roller coasters. Kingda Ka, the tallest and America's fastest, is usually considered crap. Phoenix at Knoebels, which is tiny, is said to have the best airtime of any coaster ever. You really can't rank something subjective. It's like ranking the best car, it's different for everyone.

Sent from my VS820 using Tapatalk

I have a setup in place that considers outliers in the data when calculating the scores. If a piece of data is 5 standard deviations above or below the mean or more, it gets capped. Currently, for the height, the point at which 5 standard deviations is reached is ~299.93 feet. Kingda Ka obviously gets the max score for height, but because its score is capped at a certain point, it doesn't get the #1 overall score (it's currently #11). The one that does hold the #1 spot for roller coasters currently is Fury 325.

Also, in terms of the Phoenix, it does sound like a great ride with all the airtime moments that the hardcore amusement park travelers keep talking about (if there was consistent widely-available data on airtime for every roller coaster, I would include that, too), but I have a strong sense that there is a lot of additional positive bias for this ride, due to its refurbishment in the 1980s sparking a roller coaster relocation/renovation renaissance of sorts. That's the whole point for why I did this data scoring system for roller coasters and other rides: to cut through all of the bias and subjectivity amongst ride enthusiasts and high-profile awards like the Golden Tickets, which are barely based on anything other than emotion. Think of the scores I've developed for these rides as a "blind taste test" to see if they live up to the hype or fall flat.

In terms of it not being appropriate to rank these rides with this data, I strongly disagree. The emotions you feel during the ride are indeed subjective, but the physical ride is not. I'm ranking the roller coasters themselves, not the rider's emotional state during the ride.

 

[Smithy]

I think you have too much time on your hands.

 

[LiveForTheLaunch]

This is confusing but I love that Greenfield Village/Henry Ford and Crossroads Village are on the list! Amazing places to go but not sure how you'd put them up against a theme park.

 

[JoshC.]

In terms of it not being appropriate to rank these rides with this data, I strongly disagree. The emotions you feel during the ride are indeed subjective, but the physical ride is not. I'm ranking the roller coasters themselves, not the rider's emotional state during the ride.

So essentially you're ranking roller coasters as an ornamental piece?

By ignoring a rider's emotional state, you ignore the point of why people ride a roller coaster. In a vague, generic, sense, people ride roller coasters to 'feel' something; fear, fun, excitement, etc. Of course there are other reasons people ride them, like to increase a cred count, or for bragging rights or whatever, but the fundamental reason people ride roller coasters is because it puts them in an emotional state which they like.

So, ignoring that means you aren't ranking coasters on how "good" they are in the sense of how much people enjoy riding them. Instead, you are simply ranking them by how good they sound as ornamental piece of sorts, by means of looking at their stats. Such stats give no indication whatsoever as to how it will trigger your emotional state.

Some things simply can't be ranked scientifically!

 

[jackdude101]

This is confusing but I love that Greenfield Village/Henry Ford and Crossroads Village are on the list! Amazing places to go but not sure how you'd put them up against a theme park.

Outdoor living history museums often have things like steam trains and carousels (which Greenfield Village and Crossroads Village both have), which in a way makes them resemble theme parks. Greenfield Village was the primary source of inspiration for putting the Main Street, USA section in Disneyland (only tweaked slightly to resemble the hometowns of Walt Disney and one of the chief designers), don't forget.

That was another one of my goals with doing this: to highlight lesser-known places that an amusement park visitor might enjoy, but may overlook because they don't have roller coasters and other thrill rides.

 

[ECG]

Cheetah Hunt is a family coaster

You lost me at that point. A roller coaster going 60 mph is not a family coaster. Anyway, rating roller coasters based on stats isn't perfect, but they are a strong "forecast" for how good they are.

Family coaster might be the wrong term, but the park built the coaster to be "family friendly" - something Cedar Point didn't do with Maverick.

 

[SaiyanHajime]

Cheetah Hunt is a family coaster

You lost me at that point. A roller coaster going 60 mph is not a family coaster. Anyway, rating roller coasters based on stats isn't perfect, but they are a strong "forecast" for how good they are.

Family coaster might be the wrong term, but the park built the coaster to be "family friendly" - something Cedar Point didn't do with Maverick.

I'm convinced that family is nothing but a last minute panic all inclusive marketing tool for parks who fear they're excluding the most profitable theme park market. Instead of actually designing stuff for families, they slap on the family buzzword and all is well.

My two favourite examples of this lunacy include Skyrush and Detonator's Angry Birds retheme. And before you ask, yes, the park claimed Skyrush was family friendly. It's that winged seating, you see, it offers guests a choice and makes it suitable for everyone! Here's a quote from the park... “Because Hersheypark is a family park, we wanted a ride that offered choices to the riders. The wingedseats meet those specifications…”

**** off.

Maybe it's possible Skyrush wasn't intended to be this ridiculously intense thing. I see increasing evidence that such design is little more than a fluke side effect of the commutative requirements a park specifies for the design of a coaster... But then surely you'd cover it in trims? But Skyrush is 200ft. Like, seriously? "Family"? 200ft must have been in the brief, so either "family" is an after thought, or it's badly designed! Mad thing to say about such a great ride but hey... Intention matters!

I mean, sure, families with older kids visit theme parks. Of course they do. But IMO " family ride" should mean "suitable for the majority of age groups" and Skyrush is only suitable for the "adult" theme park market, which starts at about 12 years. It's not even suitable for older adults, because if anything is gonna cause heart problems, it's Skyrush!

The main reason Cheetah Hunt is not a family coaster, regardless of what the park says, is the height restriction. It's going to be around 1.4m, because of the restraint and seating (same with Skyrush). That equates to about 10 years old for most kids. Again, failing to be inclusive of the wide range of age groups.

Height restriction should be the definition, imo. 1.2m and under and I'll accept it as family, regardless of ride experience, within reason, but I can't think of an example that would be an exception.

Tame adult rides are fine... But, its all about intention. I question the logic of Cheetah Hunt when they could have gotten something very similar with a lower height requirement. That's just bad design imo, get a ride type your target audience can ride, for a start, and make sure the ride experience matches that expectation!

So, yeah, Cheetah Hunt isn't a family ride imo, its just a tame adult coaster. And that's fine. It can be that. But the park is just trying to tick boxes with buzz words when they call it family, worried about turning parents off from coming.

Sent from my MotoG3 using Tapatalk

 

[nadroJ]

^It's a good 'starter' coaster though, so, sort of family in that aspect?

This is weird by the way. Cool idea but to gain the actual information you'd need to make such a judgement would take a lifetime. And will never be accurate because opinions are a thing.

 

[Hyde]

Something to emphasize to everyone is that such statistical models do not come to fruition on their first pass. The published piece I worked on for my masters degree was what effect electric vehicle incentives have on buying behavior in California. I just did a count for fun on how many regression models I developed over 4 months of work: 36 models until we landed on our beautiful, simple model that showed utility subsidies and front end tax credits encourage the greatest EV purchasing. Glad you've been able to make some more advancements with this regression in adding carousels Jack, and am interested to see more!

Looking over the brief pictures you've shared, it would appear on the surface that roller coaster count carries a strong correlation with overall ranking. If possible, I'd heavily encourage you to run a statistical regression of the data, to see to what effect each factor plays into the rankings - it could very well be that one factor, such as number of roller coasters or roller coaster height, trumps other factors in the rankings. There are a number of ways to control for this, such as taking the natural log of an input, which might allow for a more even keel. (Natural log is often a best recommendation for factors that are far outliers in statistical regression)

I have a setup in place that considers outliers in the data when calculating the scores. If a piece of data is 5 standard deviations above or below the mean or more, it gets capped. Currently, for the height, the point at which 5 standard deviations is reached is ~299.93 feet. Kingda Ka obviously gets the max score for height, but because its score is capped at a certain point, it doesn't get the #1 overall score (it's currently #11). The one that does hold the #1 spot for roller coasters currently is Fury 325.

So your standard deviation is 59.9 ft.? Fascinating, I'd have ventured a larger deviation. How does the standard deviation look for speed, length, inversions, or other factors? Again, if you are finding height plays too overbearing a roll in the regression, it might be worth venturing other controls, such as taking natural log, or lowering the cap of standard deviation. EDIT: Actually, thinking more, don't lower the standard deviation just yet - you are wanting to focus on the differences in height of rides and how they affect park ranking; capping the roller coaster heights means you are assuming that at a certain point, more roller coaster height does not matter. This is not really the case, as it can be well argued and reasoned that a 300+ ft. roller coaster will show greater draw and popularity than a smaller roller coaster.

The #1 challenge of any statistical regression is being able to quantify the qualifiable. Saying you love something is easy enough - but it needs to be put to numbers in order to be made accountable. How much do you love something? 4 out of 5 times?

Looking over our results, we know we are getting close to a good calculation, but are not quite there. For instance, Knott's Berry Farm carries a larger return than Six Flags Magic Mountain. Cedar Point is also showing a massive, outlier return. I believe it would be safe to say that this is actually not the case, so there must be some additional factors we are not currently accounting for.

And as we have discussed before, there are a number of very popular roller coasters in the world that are statistically inferior. Maverick, Intamin Mega Lites, and RMC are just a few examples of this - accounting for this popularity would lend greater robustness to the model, and lend a more reliable output.

I would encourage a consideration of using rider survey data as an easy way to account for some of these "other" factors that are not being captured. It is simple enough to scale up or down the impact rider survey would have on the model, but with a large sample of rider survey data already available (Mitch Hawker Coaster Poll), it is worth the time to see if this data could lend further insight.

So recap:

- check statistical significance of each factor to see what level of influence they play in the model. If one factor, such as coaster size, is found to be vastly more significant than other factors, this will make the model very correlated to that factor, and would require more control.
- Good regression controls: take natural log of factors, square the factors. I'm happy to talk more on other possible controls worth considering.
- Check the correlation of roller coaster count to park ranking. If there is a high correlation, the model is only reflecting that one variable.
- Consider using other variables to account for other factors. Using ride ratings could be one such way to easily account for other qualities that are not quantified, such as ride smoothness, asthetics, etc.
- Remember that no model is perfect. What is perfect is the interpretation of the output, and using proper citation and logic for using the output to influence your findings.

I'm happy to talk more about this with you offline, and help in running some of the model if you are willing to share. I do still have licenses to STATA and SAS, and would love help out!

 

[ECG]

Don't mean to hyjack the topic but...

I'm convinced that family is nothing but a last minute panic all inclusive marketing tool for parks who fear they're excluding the most profitable theme park market. Instead of actually designing stuff for families, they slap on the family buzzword and all is well.

This is something that I spoke to the BGT park president about during media day saying that it looked like the coaster had been toned down from the original animation of Cheetah Hunt the park released. He assured me that wasn't the case and that the coaster was designed from the ground up to be more family friendly because they felt like the park already had too many coasters, like Kumba, ShreiKra and Montu, that weren't.

My two favourite examples of this lunacy include Skyrush and Detonator's Angry Birds retheme. And before you ask, yes, the park claimed Skyrush was family friendly. It's that winged seating, you see, it offers guests a choice and makes it suitable for everyone! Here's a quote from the park... “Because Hersheypark is a family park, we wanted a ride that offered choices to the riders. The wingedseats meet those specifications…”

That's strange because during the Q&A we had while waiting for our Skyrush ERT we were told that the coaster was specifically designed to be the most intense in the park with thrill seekers like us in mind. I thought you were there for that.