• ℹ️ Heads up...

    This is a popular topic that is fast moving Guest - before posting, please ensure that you check out the first post in the topic for a quick reminder of guidelines, and importantly a summary of the known facts and information so far. Thanks.

Alton Towers in 2015 - A Detailed Season Analysis

Sku

TS Member
Introduction

Since the 16th of July I have been collating data from ridetimes.co.uk, and I will attempt to present some of the data in this thread for you to digest. The data on ridetimes is a great resource for getting a general picture of how long queues are on any given day, but it can also give us insight into some other finer details about park operations as well, which I will go into as we progress below.

This is going to be a very long and detailed post, so feel free to skim through to the good stuff if you are not interested in my analysis. I won't be too offended I promise. I will try and go some way to explaining why the data looks the way it does, as most of the peculiarities and patterns demonstrated have sound explanations, which I will try to lay out for you, if you have the patience to stick with me for the duration.

I am writing this primarily to post right here on the Towers Street forum, but a lot of time and effort has gone into collecting the data, producing the graphs, and writing up this analysis. I therefore might reuse it elsewhere as well, such as on a future blog I intend to create, or on other forums, but know that I compiled it and posted it here for you first.

I will try to paint a picture of the season overall, but obviously my data only begins on the 16th of July. We didn't really miss any major attendance events other than the Easter break, which was before The Smiler accident, and therefore represents a completely different state of affairs at the Towers to those which we are left with now. My analysis is therefore focused on where we're left in the aftermath of the accident. Indeed, the accident, and resulting low attendance, is the reason I began to collect these wait times in the first place.

When you collect all the data which is posted on ridetimes, it fairly accurately shows the overall attendance pattern at the resort, and while I can't even begin to estimate actual attendance in figures, we can get a good snapshot of how things have been at the park this year. Of course to both me and you, actual gate figures don't mean very much, what we want to know is how that translates into wait times at the park, and coincidentally, that is exactly what ridetimes allows us to see!

Of course sometimes the wait time for various rides is posted incorrectly by Alton Towers, but the overall pattern remains largely untarnished, once you average it out across all rides, and throughout the entire day. And you can see a clear correlation between the wait times posted at different attractions, when one attractions wait is long, most other attractions are also long. The overall pattern emerges clearly, and mostly unsullied by occasional inaccurately estimated wait times.

How have things been at the park this year?

And so how have things been at the park? With my insight into the average wait times everyday since the 16th of July, I cannot help to have noticed that things have been bad as the season has progressed. Really pretty bad. For almost the entire season, a vast majority of attractions have had wait times lower than 10 minutes all day long, a situation which was reserved for only the very quietest of days in previous seasons, and became a near daily occurrence this year. Even some of the busiest days of the year did not approach the kind of wait times that have been seen in previous seasons on peak days. So not only have ticket sales been hit, but fastrack sales will have been hurt too due to nearly non existent waits, a double whammy for a park reeling from a massive PR disaster, probably beyond compare in the UK themed industry.

It has clearly been grim times indeed for Alton Towers. Attendance at such a resort is largely driven by repeat visitation, when people come back after the next investment cycle, to experience a major new attraction. With The Smiler out of action, and the parks reputation badly damaged, many hundreds of thousands of guests who may have thought about visiting, have simply stayed at home this year, or gone elsewhere. It is difficult to see any recovery in sight without a major new investment in the park. Many have speculated that the parks attendance will be lower than 2 million this year, but it is actually likely to be even worse than most have even imagined. I am mostly numb to it, as I have got used to the kind of low figures you will see below as the months have gone by, but maybe you will find just how bad it was, especially during the summer season, to be quite surprising.

If we average out the wait time of some major attractions over the whole season, you start to see how short those queues have been all year. An average wait time of only 8 minutes for Air and Rita. An average 3 minute wait for Nemesis, and just 2 minutes at Oblivion. You get the picture. For the waits to average out this low, you know it's because for so much of the season, they were posted as 0 or 5 all day long, a sorry state of affairs indeed.

While low wait times might be great for some who are truly mad about roller coasters, a lack of atmosphere really affects peoples overall enjoyment of the park. When the park feels like a ghost town, it completely lacks any excitement or atmosphere. With no wait times, there is no time to build any anticipation for getting on the next big ride.

In fact, very few people who visit as families, or in groups, probably even realise that time spent queuing probably ends up forming some of the most stand out memories of their day. While they are in close quarters with friends, and family, in an increasingly digital world, they probably spend considerably more time communicating with each other, and enjoying each others company than on a regular day. How many photos on your phone when you get back from a day at the park are of your friends and companions taken while you were standing in line? While many people can, and do, use their smartphone in line, it is generally awkward to do so for prolonged periods of time, with the constant shuffling of people around you, and friends close by to keep you company. A day at a theme park is a great way to connect, and share an experience with the important people in your life, and its why so many families and groups of friends love to visit.

Methodology:

Trying to make an actual attendance graph for the entire season proves a little bit tricky. You can't just take the average queue time from every ride, and smosh them together. A 45 minute queue at Oblivion is a far bigger indicator of a busy day, than a 45 minute queue at Octonauts. And as I discovered the attendance pattern for family rides is quite different than that for thrill rides, as they have a different audience, something I will elaborate on further in this post.

For my interpretation of an overall attendance graph, I have weighted the data, to give far more importance to major attractions, than the supporting rides, or the children's rides. I believe that major high capacity attractions are the best indication of how busy the park is overall. Low capacity family attractions can have long lines even on more quiet days, and just don't paint the most accurate overall picture. Further on in this post I break down these different types of attractions for individual analysis, using data I haven't manually adjusted. Only my first overall attendance graph has been somewhat adjusted.

I have selected what I considered to the be the 6 most important thrill rides, and the 5 most important family rides for determining overall park attendance. Please keep in mind I did not have much choice, as more attractions were not added to ridetimes.co.uk until much later on in the season. Water rides are intentionally not included as their popularity is seasonal, and therefore doesn't reflect attendance. Here forth I will refer to these two sets of rides as "Thrill 6" and "Family 5".

Thrill 6 consists of:
Air, Nemesis, Oblivion, Rita, Th13teen, Sonic Spinball, for obvious reasons.

Family 5 consists of:
Runaway Mine Train, Octonauts Adventure, In The Night Garden Magical Boat Ride, Get Set Go Tree Top Adventure, Postman Pat Parcel Post, as these are the 5 rides in that attract sizable queues on busy days.

I only use queue time data between 10am and 4pm for the overall season comparison, to provide the fairest comparison possible between days. These are the core hours the park is always open, regardless of closing time, ERT, or late night events.

Overall Attendance Graph, and Seasonal Analysis

Without further lengthy explanation, I present to you my attempt at weighting the importance of the different rides, into an overall picture of attendance in 2015. Each graph in the post can be clicked on to enlarge it.

K213o2y.png


The first thing I think that most people will notice, is that the first day where I ever recorded wait times, Thursday the 16th of July, was crazy busy compared to the weekend that immediately followed it when the park was almost completely dead. Actually, it was even crazy busy compared to the entire season, with an average wait time of almost 40 minutes across the 11 attractions selected.

It was a midweek Thursday, before the summer holidays had even started, so you will probably be thinking I surely must have made some kind of mistake. But in fact this day really was that busy, as it was the main arranged school trip day in July, with enormous numbers of coaches turning up to the resort full of school children. The days preceding this had also been pretty busy, however I had not yet fully finished my solution for collecting the data at that point, so they are not shown. You can even see this pattern continue somewhat on Monday the 20th and Tuesday the 21st the next week, also with higher attendance due to a few remaining school trips. In comparison, the weekend in between these dates had near walk on wait times.

The explanation for this is quite simple. School trips for July, were all booked and paid for before the Smiler incident happened. Add in attendance at the park being very low otherwise, and you have the result of a major stand out anomaly on the chart, that we see above.

When the summer holiday proper started on Wednesday the 22nd of July, attendance dropped completely flat for 2 weeks, a fact widely discussed on this forum at that time. Attendance is often at its summertime low at the resort right at the beginning of the summer holiday, a tip you can keep in mind for future seasons. However how deep this lull went was unprecedented for attendance in summer holiday time, with many attractions being walk on all day long, and an average overall wait time lower than 5 minutes across all attractions. In the school holidays!

Later in the summer attendance does pick up, peaking towards the end, at bank holiday weekend time. There are a few dips and troughs along the way which generally correlate with bad weather; particularly horrendous weather on the 14th of August is clearly shown. The peak summer holiday attendance day was, not surprisingly, the Saturday of the August Bank Holiday weekend, with average wait times over 20 minutes for the first time since school trips. This week overall was the busiest of the summer holiday.

Once the summer holiday was over, things got very quiet again. Modest attendance on Friday and Sundays, but very quiet for the entire midweek as would be expected at this time of year. A Saturday in September or October is roughly comparable to an average day in the summer holidays.

Things didn't pick up again until the first weekend of Scarefest which was fairly busy, the first midweek gap being very quiet, and the second week being much much busier due to half term. Halloween was the busiest day of the event by a long way due to a combination of factors, for once actual Halloween fell on a Saturday, and also abysmal weather had been forecast on Tuesday, Wednesday and Friday during half term week. Considering the weather was forecast to be so bad on those days, attendance was still pretty good, showing the popularity of the Scarefest event. Also to note is how low attendance is on the final Sunday of half term compared to the rest of the week, this will be down to the fact that people don't want to stay out late the night before returning to work and school the next day, especially when there are many other more favorable days to choose from.

Fireworks attendance on the other hand was pretty low all things considered. The Friday was completely dead, with at least some people turning up later on for the display due to the late arrival deal. With no late arrival deal on Saturday and Sunday, you would think the park would be heaving on Saturday and Sunday daytime, with most people choosing to make a day of it, due to the high cost of admission. But alas, attendance was not even close on the Scarefest half term week levels. This can probably be attributed, again, to fairly bad weather being forecast, but also to changing attitudes in the UK, with Halloween being a far more popular event than Fireworks with a modern British audience. People also may have not felt a firework display alone was enough to justify paying the high admission price to a park which has otherwise been struggling. Firework displays are also of course common place in the UK, and very professional displays can be seen for cheaper, or even for free if you know a good vantage point, elsewhere, but Halloween events like that hosted at Alton Towers are far more rare, and also extremely popular.

So that rounds up my analysis of the season. I am looking forward to being able to have an overview of an entire season next year, especially now more rides are represented on ridetimes, and I intend to monitor wait times from opening day onward.

Average Wait Times for Major Thrill Rides

Now I can start to break down some of the individual data collected for your consumption. Let's start with the actual average wait times for the 6 Thrill attractions, without any kind of weighting this time. I haven't meddled with this data in any way. This is probably a good representation of how busy the park is for thrill seeking visitors, as long roller coaster queues are the main concern when visiting on a busy day. Family attractions are dealt with separately later.

0qajLWt.png


It is a fairly similar graph to the overall attendance one, with the notable differences being as follows:
-High attendance due to school trips more pronounced at the far left of the graph, due to school children wanting to hit all the roller coasters in a short space of time before returning to their coaches.
-Attendance in the summer holiday appears a lot lower than the overall weighted graph. This is because Alton Towers really struggled to maintain a thrill seeking audience in the summer, however attractions aimed at smaller children remained very popular, especially Cbeebies land
-Only really Scarefest was successful in bringing thrill seekers to the park this year.

I wondered if there was a way of showing the seasonal trend when it came to the thrill rides, and wanted to show how each ride contributed to the overall wait time average. I created a 7 day rolling average to even out the weekend spikes, and came up with the following area chart. It stacks the wait times of the attractions on top of each other, to show the overall seasonal trend.

013zpKj.png


While this is great for showing seasonal trend, it is not very accurate, as for instance, in September a Saturday is just as busy as a day in the summer holiday. Also the week before, and the week after, the October half term, look like they were very busy according to this graph, when they are not. The Firework event doesn't even register a blip.

If we don't use a 7 day rolling average, and instead use the data as it is without messing around with rolling averages, the graph looks a lot more messy, but is now highly accurate:

KKxSC0J.png


The first weekend of Scarefest appears appropriately busy now, and the Fireworks event registers at the end of the season too, showing a clearly very quiet week in between those events.

One interesting thing to note is that Th13teen queues were actually longer during the summer holidays than during Scarefest in general. There are 2 reasons for this. Firstly, there were a number of operating problems with Th13teen during those months, with trains being sent empty, and a lot of downtime. Secondly, family based attendance was actually generally higher in the summer holidays than during Scarefest, a point I discuss further later on. Th13teen is a hybrid between a thrill ride and a family ride, I could have really included it in either category, however ridetimes.co.uk shows it as a thrill ride, and it is a SW, so I kept it here.

I have one last graph for the thrill rides, and I think I like it the most just because it looks pretty. It's the same as the last graph, but now in bar form instead, more clearly showing each rides contribution to the overall wait times on any given days that you may be interested in:
SHx8gKA.png


Here you can use the stacked wait time, to see, on average, how long a guest spent waiting if they wanted to experience all 6 of these attractions. On Halloween you would have spent, on average over 250 minutes total waiting in line. However, on the large majority of days, you would have waited around 60 minutes total or less, to experience all 6! Once again this goes to show just how quiet it has been all season. Some people may have waited far longer, or far shorter on these days, as they are of course averages only.

Also, note that the school trip day on the 16th of July had the longest waits of the entire year for Air, Nemesis and Oblivion, even more so than Halloween. The mighty Nemesis managed to attract a 70 minute queue, which is practically unheard of these days. I did not attend to be able to comment for sure but it is possible that 1 train operation could have been in effect on this day, but it's hard to imagine Alton Towers would be this unprepared when the number of guests could have been easily forecast, since all the schools were booked months in advance. Rita's wait time is peculiarly low in comparison to others on this day, as it went down for over 2 hours during the midday peak. Many school children with a coach deadline to make will not have had time to return later on, with the park as busy as it was, and most of them instead choose to hit far away Forbidden Valley rides in the afternoon, according to the wait time data from that day. School children are not great day trip planners, and will likely visit the attractions in the order they arrive at them. Rita was down for the duration of the time they were most likely to arrive in the Dark Forest, which was reflected in Th13teen's queue going to 100 minutes during the outage.

The other extra detail of note here really, is that the 14th of August was very quiet day due to horrific weather and flash flooding blocking transport routes, but Th13teen was unavailable for over 4 hours. This meant the queue was fairly long once it did finally reopen, thus explaining why it seems to have such a long queue on an otherwise quiet day at the other rides.

Average Wait Times for 5 Family Rides

I now move on to the family rides, and we are going to see that the attendance pattern is quite a bit different to what we have seen so far.

6WJAxfm.png


While Scarefest was far and away the busiest time at the thrill rides, when it comes to Family attractions, it was mostly business as usual. The first Saturday of Scarefest for instance didn't register any different, and maybe even slightly lower in attendance than an average Saturday in September or October. Half term week picked up a bit as could be expected, but this is most likely due to the fact its a school holiday rather than because of the Scarefest event, which really, doesn't probably appeal to parents with small children.

Also as could be expected, the busy school trip day in July doesn't really register on the graph either, as these attractions would be of little interest to teenagers. So despite it showing as a busy day in my overall graph, if you were visiting with small children, you were probably fairly unscathed by the long waits.

The Saturday and Sunday of the August bank holiday were by far the busiest days for family attractions. This is likely due to the busy schedules of working families with young children. It is interesting nonetheless, because those two days are not that much of a big deal when you look back at my Thrill 6 graphs, but here they are the clear peak days of the season.

Even more clearly in focus on this graph, is the 14th of August being a very quiet in the summer holiday. This is because flash flooding caused many train cancellations and road delays, and was heavily featured on the news. A lot of people probably didn't even attempt to make the trip given the circumstances and news coverage.

Thrill Ride and Family Ride Comparison

I found the difference in trend between family rides and thrill rides to be pretty interesting, so I went back to my 7 day rolling averages, to plot the seasonal trend of thrill rides versus the family rides.

cr6pa23.png


For almost the entire season, and especially in the summer holidays, you would have spent far far longer on average waiting to get on the family rides, than the thrill rides. This trend was only finally reversed during Scarefest, which attracted many thrill seekers back to the park.

During the first week of Scarefest, the family attendance actually registers as some of the lowest of the year, which leads me to speculate that some families with younger children might actually be put off visiting by Scarefest. This trend changes again during the half term week, but as I mentioned before, this is probably more because it is convenient to visit during school holidays, rather than because of Scarefest itself. Regardless, the summer holiday was in fact busier at the family attractions than during the half term, which is the opposite of the thrill ride attendance pattern.

It is interesting to note, that in previous seasons, we would expect to see average wait times at the major thrill attractions to be quite a fair bit higher than at smaller family rides, especially during the summer holidays, and other busy periods, but in fact we did not see that for almost the entire season. This is really one of the most interesting things I think I can demonstrate that shows the impact of The Smiler incident. Thrill seekers have been majorly deterred, but the popularity of Cbeebies particularly land remains largely intact.

Typical Day Wait Times by the Hour

At this point I think I'm mostly done discussing calendar days, but what does the average day look like at the park by the hour? Well, I have another juicy stacked area graph for you to show just that.

nP1PdXa.png


As we can see the Sonic Spinball queue, as everyone already knows, stacks up very early in the day. The wait times overall reach a natural peak just before the mass 1pm lunchtime, where they suddenly start to drop down again. Had the park been busier this year, and open for longer hours, you would have seen a double peak, with a second bump in waits after lunchtime, and we did see this pattern at Scarefest (Please see my post in the Scarefest thread for my more detailed analysis of that event). With low low wait times, and early closing, there wasn't really any mad rush to hit the queues after lunch, with more people likely choosing to hit the exit having already finished with everything.

The only ride that really starts building a slightly longer queue later in the day is again Sonic Spinball, due to it's low capacity and proximity to the exit. Also to note is that Air, hidden all the way at the back, takes the longest amount of time to start building up it's queue, which is then quite stubborn and doesn't decrease much towards the end of the day.

Another way of looking at this same data, is to map the absolute percentage share that each ride has of the wait times compared to the others for each hour of the day, which further reinforces the trends we have just discussed.

54yigZa.png


Here most interestingly, we see the Sonic Spinball effect once again, attracting a very large queue, and very rapidly at the beginning of the day. We can also note Oblivion with generally fairly low wait times, but a defined mid morning bulge that tapers off later, also fairly close to the park entrance. Things don't really pick up in Forbidden Valley until the afternoon due to its distance from the entrance.
 
Last edited:
Ride Downtime

The other, and unexpected addition type of data, I was able to collect, is about how often, and for how long, various rides are not in operation. This data is only so-so accuracy wise. Very short downtime, and brief shutdowns are unlikely to be posted to ridetimes. Especially since the data only updates once every 30 minutes.

Ongoing Downtime

There are of course a couple of major ongoing closures we are already aware of, but I will list them here for the sake of completeness.

-The Smiler:
Closed on the 2nd of June, through to the end of season.

-Nemesis Sub Terra:
Closed at 11:30am on the 7th of August, through to the end of the season. It did open for the morning on this day, which suggests the original reason for the downtime was not planned, but in fact incidental. It's downtime only became extended indefinitely at a later stage, when it was confirmed by Alton Towers that it would not reopen some weeks later.

-The Flume:
Closed from the 10th of October through to the end of the season.

All other rides which are listed on ridetimes, were operational every single day of the season, at least for some part of the day, since my project begun on the 16th of July. This is a pretty good record, in comparison to for instance Thorpe Park, where major attractions frequently become unavailable for days at a time, for unknown and undisclosed reasons. Other than the 3 listed above, all other downtime at Alton Towers has been confined to a 24 hour period, or less.

Longest Downtime Incidents

Here are some of the more major outtage incidents which I recorded, ranked by order of the downtime length:

-In The Night Garden Magical Boat Ride
8 hour downtime, 12&13 August
Closed at 10am, and did not reopen. It reopened 1 hour late the next day.

-Octonauts Adventure
6.5 hour downtime, 27 July
Closed at 11:30am and did not reopen.

-Th13teen
6 hours downtime, 16 August
Closed at 10:30am, barely makes it reopen before the end of the day.

-Hex
5 Hours downtime, 25 October
Closed at 4pm during Scarefest, and did not reopen.

Rita and Th13teen additionally both clocked several outages throughout the season lasting 2 hours or longer, and were by far the most unreliable rides.

Longest Cumulative Downtime

Rides with longest amount of downtime since the 16th of July. Opening up to 1 hour late is forgiven for the sake of these statistics, as a slightly late opening is very different in nature, and much less inconvenient for guests than actual operational outages.

-Thirteen: 35.5 hours
-Rita: 30 hours
-Sonic Spinball: 16.5 hours
-Night Garden Boat Ride: 13.5 hours
-Octonauts Adventure: 10.5 hours
-Air: 7 hours
-Hex: 6.5 hours (Only added to ridetimes since 27th Sept)
-Postman Pat: 5 hours
-River Rapids: 4.5 hours
-Chocolate Factory: 4.5 hours (Only added to ridetimes since 29th August)
-Oblivion: 4 hours

All other rides were 3 hours or less overall, but notably, many rides and attractions were not added until later in the season. The only major ride not mentioned above is good old Nemesis, with just 2 and half hours in total. Good girl.

Most Number of Days with Downtime

Finally, which rides had the most days with at least one incident of downtime since the 16th of July. Once again, opening up to 1 hour late is forgiven, opening more than 1 hour late in included. The top 3 rides in this category are considerably more unreliable than pretty much anything else.

-Th13teen: 20 days
-Rita: 19 days
-Sonic Spinball: 15 days
-Air: 9 days
-Oblivion: 6 days
-Octonauts Adventure: 6 days
-Night Garden Boat Ride: 6 days
-River Rapids: 4 days
-Chocolate Factory: 4 days (Only added to ridetimes since 29th August)
-Mine Train: 3 days
-The Flume: 3 days

All other rides had 2 days or less with downtime. Once again, the only major ride not mentioned is Nemesis, clocking in downtime on only 2 separate days. Who would of thought a ride of her age could so decisively beat out a list of rides which are so much newer than her for up time.

TL DR
And with that we reach the end. Thanks for staying with me if you did, and if not, thanks for checking out the thread, and hopefully finding some of the graphs I produced interesting. There are many more ways to use, and interpret the data, but this was my idea of keeping it brief. That didn't really work out in the end, so I'm thankful I didn't select any more graphs or information to show you or I might have spent weeks writing this up!

If anyone has any requests, e.g what was the average Nemesis queuetime at 11am on Thursday mornings, or any other obscure questions I'll be happy to try and answer them. I have the data for every ride, every day since the 16th of July, including those added to ridetimes later on in the season. In total, I collected over 60,000 individual wait times, across 35 different attractions, on 116 days.

I hope to be doing the same again next season, this time crucially with an entire seasons worth of data, rather than only a partial season.

In the closed season, I continue to monitor wait times at Disneyland Paris, and may branch out to include some other parks next season.

Thanks again.
 
Wow. I'll be honest, I've not read it all yet (will save that for a long dull shift at some point), but you get a "like" just for the ammount of effort this must have taken. ;)
 
  • Like
Reactions: Sku
Read it all :). I know you used ridetimes but how did you collect the information of there? Great effort though and quite accurate (as much as possible with averages and ridetime.com)
 
  • Like
Reactions: Sku
Wow. I'll be honest, I've not read it all yet (will save that for a long dull shift at some point), but you get a "like" just for the ammount of effort this must have taken. ;)
It's definitely going to be a focus to try and be more concise with this type of thing in future, and not let it grow into such a monster.

Read it all :). I know you used ridetimes but how did you collect the information of there? Great effort though and quite accurate (as much as possible with averages and ridetime.com)
With a trusty old notepad and pen. But seriously it was just a simple script that visits the page every 30m during opening hours and records what is shown. It is a similar technique used for instance by price comparison sites which automatically check for the most up to date prices on other websites.
 
Not read all of it fully yet, but did a quick scan of some parts of it and must say thank you for taking the time and effort to do this. It makes for a very interesting read. Would love to see how next years results vary to this half seasons results.

Look how Nemesis still shines through all the new rides when it comes to reliability - well done B&M, a masterpiece.

Once again, thanks you Sku.
 
  • Like
Reactions: Sku
Wow good effort there mate.
Would be interesting to know what problems Rita and Th13teen have to make them the most unreliable
 
  • Like
Reactions: Sku
Top