Matt N
TS Member
- Favourite Ride
- Mako (SeaWorld Orlando)
Disclaimer: This is a very long post with a fair amount of statistical talk. I'll try my best to simplify some of it a bit and explain a few things, but if you don't like statistics, this post may not be for you!
Hi guys. Over the course of my many years being a Brit following the UK enthusiast scene, I've noticed that that there is a common feeling among British enthusiasts of the country being a poor relation to other countries in Europe in terms of roller coasters and theme parks. Many British enthusiasts make out that parks in Britain are being left behind by parks on the continent in the roller coaster stakes, and overall, the topic of the British coaster scene lagging behind that of other countries in Europe has been well-discussed at this point, along with many other variations on the wider point of "British parks and coasters are rubbish compared to those in mainland Europe". With this in mind, I thought that I'd put this common theory to the test and perform a statistical analysis of European countries' roller coaster selections, and attempt to answer the question of "Which European country is best for roller coasters?". I thought that it would be interesting to see which European country comes out on top in the coaster stakes when applying various statistical measures and asking various questions. As inferred above, I also thought it would be interesting to see whether that common British feeling that our coasters are lagging behind those of other mainland European countries is 100% justified, or whether it's simply a case of the grass seeming greener where we're not.
I hear you asking "Hold on a second, Matt. Haven't you analysed European coaster selections before?". Well, my answer to that would be that I have indeed analysed European coaster selections before, but my previous analysis focused on parks rather than countries, so with me focusing on whole countries this time as opposed to individual parks, the analysis will be different and will produce very different results.
So without further ado, let's get started on the analysis! Firstly, let me explain some of my methods...
Methodology
Now I've gotten that out of the way, let's head into the main analysis...
Main Analysis
Just as a recap, our question here is "Which European country is best for roller coasters?"!
There are a number of ways in which you could answer this question, as "best" is an open-ended metric that can be assessed from many different viewpoints, but I chose two different angles to approach the question from; the overall quality of a country's top 20 (or entire scoreable selection where this was not applicable), and the quality of a country's top end (i.e. its main headlining draws). Let's start with the overall top 20 quality...
Overall Top 20 Quality (Which country's coaster lineup is the most consistently strong?)
My first question asks which country's lineup is the most consistently strong. To determine this, I looked at the overall top 20 (or entire selection where this is not applicable) to gauge an idea of which countries have the most consistently strong coaster selections.
The first statistical measure I used to determine this is the mean coaster score per country. The mean is a calculated average where all of a country's scores are added together and divided by the number of scores; the formula is as follows:
Mean Score per Country = Overall Sum of all Scores/Number of Scores
When this formula was applied to all the countries used, the ranking was as follows:
Using the mean, our top 3 most consistently strong countries are Germany, France and the United Kingdom.
However, one key flaw of the mean is that it can be heavily influenced by extreme values on the high or low end of the spectrum, and this becomes more apparent as the dataset becomes smaller.
As such, I used another statistical measure to determine the answer; the median coaster score per country. The median is the middle value within an ordered dataset, so in a dataset with 20 numbers as many countries in this dataset are, the median value will be the midpoint between the 10th and 11th values.
When the median was applied to all the countries used, the ranking was as follows:
Using the median, our top 3 most consistently strong countries are Germany, the United Kingdom and France, with the same countries coming out on top but the UK and France having swapped places.
While the median is not overly influenced by extreme values on the top and bottom ends of a dataset as the mean can sometimes be, it arguably goes too far in the opposite direction in that it doesn't consider what's happening on the top or bottom at all; it is purely the middle value, and it will stay the same whether the top or bottom value is 5 above/below it or 50 above/below it.
Seeing as both the mean and the median are arguably imperfect measures in their own different ways, I used a final statistical measure that I invented myself... The Matt N Consistent Strength Coefficient. As the mean can be overly influenced by extremely high and low values and the median completely ignores what's going on outside of the midpoint of the dataset, I decided to try and create a coefficient that combined the best of the mean and median and cancelled out the flaws of both methods. The formula for this coefficient is as follows:
Matt N Consistent Strength Coefficient = ((Highest Value + Upper Quartile)+(Lowest Value + Lower Quartile))/2
It considers the top and bottom values, similarly to the mean, but also considers the upper and lower quartiles (the values that are higher than 75% and 25% of the dataset, respectively), so considers the more centralised focus of the median too.
When I applied the Matt N Consistent Strength Coefficient to all the countries used, the ranking was as follows:
Using the Matt N Consistent Strength Coefficient, our top 3 countries were Germany, France and the United Kingdom.
Finally, here is a boxplot showing the distribution of all 14 countries' top 20(ish) lists, created using Python's MatPlotLib library:
That should be clear enough to see which country is which, but if you have trouble, the countries are listed in alphabetical order from A to Z.
So if we're working based on consistent strength across a wider cross-section of a country's coaster selection, I think it's fair to say that Germany is our comfortable winner, with some combination of France and the United Kingdom taking 2nd and 3rd place.
However, this is not the only metric by which you can assess which European country is "best" for roller coasters. And indeed, one flaw of this metric is that to some extent, it is a bit biased towards countries with more roller coasters; the top 20 of a country with over 100 scoreable coasters will likely be very different to the top 20 of a country with 25 scoreable coasters. I would personally argue that although I accept and agree with the "quantity isn't everything" point, I do think it should still be a factor when deciding which country wins out over another, so to some extent, I would argue that having a lower quantity of scoreable coasters could make the country "worse" for roller coasters overall. Nonetheless, the flaws of this metric should still be taken into account, and I will approach the question from another angle for some balance.
Let's now move onto the other way in which I'm going to try and answer the question... I'm separating this into a new post below, as I'll breach the character limit otherwise.
Hi guys. Over the course of my many years being a Brit following the UK enthusiast scene, I've noticed that that there is a common feeling among British enthusiasts of the country being a poor relation to other countries in Europe in terms of roller coasters and theme parks. Many British enthusiasts make out that parks in Britain are being left behind by parks on the continent in the roller coaster stakes, and overall, the topic of the British coaster scene lagging behind that of other countries in Europe has been well-discussed at this point, along with many other variations on the wider point of "British parks and coasters are rubbish compared to those in mainland Europe". With this in mind, I thought that I'd put this common theory to the test and perform a statistical analysis of European countries' roller coaster selections, and attempt to answer the question of "Which European country is best for roller coasters?". I thought that it would be interesting to see which European country comes out on top in the coaster stakes when applying various statistical measures and asking various questions. As inferred above, I also thought it would be interesting to see whether that common British feeling that our coasters are lagging behind those of other mainland European countries is 100% justified, or whether it's simply a case of the grass seeming greener where we're not.
I hear you asking "Hold on a second, Matt. Haven't you analysed European coaster selections before?". Well, my answer to that would be that I have indeed analysed European coaster selections before, but my previous analysis focused on parks rather than countries, so with me focusing on whole countries this time as opposed to individual parks, the analysis will be different and will produce very different results.
So without further ado, let's get started on the analysis! Firstly, let me explain some of my methods...
Methodology
- To perform this analysis, the dataset I used was Captain Coaster's World Roller Coaster Ranking for August 2023 (https://captaincoaster.com/en/ranki...s[manufacturer]=&filters[openingDate]=&page=1), with the Country filter being used to discern the top-ranking coasters in each country. Captain Coaster is a site run by members of the French enthusiast group CoastersWorld.fr, but it has a very wide user base encompassing plenty of users of many different nationalities, so as far as sources on roller coaster opinions go, I figured that Captain Coaster was a pretty good, unbiased one to use with a large amount of available data.
- For each country, I used the top 20 extant roller coasters from the list (so the first page of the country's Captain Coaster ranking if no defunct coasters made the all-time top 20), as I figured that in most cases, 20 would be enough to get a broader understanding of a country's roller coaster selection beyond its headliners without necessarily delving deeply into the entire coaster selection to the point of reaching rides that people don't really care about. By an extant roller coaster, I'm referring to anything that has not been confirmed as permanently defunct or has some kind of potential operational future ahead of it. So for instance, I included Nemesis in the UK's top 20, as it is returning next year, but I did not include Wild Mouse, The Ultimate or Raptor Attack, as those coasters have either been demolished or left the UK permanently. There were also a few cases where travelling coasters were present, and I did not count those; for instance, Olympia Looping is listed in the lists of both Germany and Austria, but I did not include it in either country's ranking as it is a travelling ride with no permanent base.
- Where there were not 20 scoreable roller coasters available, but I felt that there was a sufficient amount to play with for one or both of the questions I asked, I simply included the whole selection. Two countries with scoreable roller coasters were excluded on the basis of them not even having enough to form a top 3, and those are Ireland with 1 scoreable coaster and Switzerland with 2.
- For each coaster, I used the score given by Captain Coaster to determine its ranking. The score is out of 100 and is given to one decimal place. While Captain Coaster has never officially revealed exactly what determines a coaster's score, I ascertain that it is determined by some combination of its ratings and rankings versus those of other rides. It should also be noted that a coaster has to have at least a certain degree of ridership to get a score in the first place, so none of the scores here are influenced by the opinions of a vast minority of riders. Coasters that the site classifies as "kiddie" are also excluded from receiving a score.
- Gesengte Sau - Wiener Prater (71.3/100)
- Wild Train - Fantasiana (69.7/100)
- Rattenmuhle - Familypark (66.3/100)
- Megablitz - Wiener Prater (65.1/100)
- Fridolin's verruckter Zauberexpress - Fantasiana (59.3/100)
- Big Bang - Freizeitpark Familienland (55.1/100)
- Gotterblitz - Familypark (53.8/100)
- Insider - Wiener Prater (49.1/100)
- Boomerang - Wiener Prater (46.4/100)
- Hochshaubahn - Wiener Prater (40.9/100)
- Super 8er Bahn - Wiener Prater (36.7/100)
- Maskerade - Wiener Prater (29.6/100)
- Dizzy Mouse - Wiener Prater (18.4/100)
- Roller Ball - Wiener Prater (16.0/100)
- Wilde Maus - Wiener Prater (8.9/100)
- Volare - Wiener Prater (3.2/100)
- Ride to Happiness - Plopsaland de Panne (99.4/100)
- Kondaa - Walibi Belgium (98.5/100)
- Fury - Bobbejaanland (86.6/100)
- Anubis The Ride - Plopsaland de Panne (80.3/100)
- Heidi The Ride - Plopsaland de Panne (78.6/100)
- Pulsar - Walibi Belgium (76.1/100)
- Wakala - Bellewaerde (69.7/100)
- Psyke Underground - Walibi Belgium (68.6/100)
- Tiki-Waka - Walibi Belgium (66.4/100)
- Vicky The Ride - Plopsa Coo (61.0/100)
- Huracan - Bellewaerde (61.0/100)
- Typhoon - Bobbejaanland (54.0/100)
- Revolution - Bobbejaanland (53.6/100)
- Calamity Mine - Bobbejaanland (51.1/100)
- Loup Garou - Walibi Belgium (49.8/100)
- Schtroumpfeur - Plopsa Coo (47.0/100)
- Naga Bay - Bobbejaanland (43.1/100)
- SuperSplash - Plopsaland de Panne (37.7/100)
- Bob Express - Bobbejaanland (36.8/100)
- Oki Doki - Bobbejaanland (33.8/100)
- Fonix - Farup Sommerland (96.0/100)
- Piraten - Djurs Sommerland (94.2/100)
- Juvelen - Djurs Sommerland (85.7/100)
- Rutschebanen - Tivoli Gardens (84.3/100)
- Daemonen - Tivoli Gardens (78.5/100)
- Lynet - Farup Sommerland (78.2/100)
- Mine Train Ulven - Bakken (73.1/100)
- Orkanen - Farup Sommerland (71.2/100)
- DrageKongen - Djurs Sommerland (69.4/100)
- Polar X-plorer - Legoland Billund (66.6/100)
- Falken - Farup Sommerland (63.9/100)
- Tornado - Bakken (63.1/100)
- Thor's Hammer - Djurs Sommerland (54.9/100)
- Saven - Farup Sommerland (53.6/100)
- Skatteoen - Djurs Sommerland (53.5/100)
- Han-Katten - BonBon Land (51.6/100)
- Rutschebanen - Bakken (51.4/100)
- Vild-Svinet - BonBon Land (51.1/100)
- Flying Eagle - Legoland Billund (33.9/100)
- Maelkevejen - Tivoli Gardens (33.1/100)
- Taiga - Linnanmaki (99.5/100)
- Junker - PowerPark (93.4/100)
- Thunderbird - PowerPark (79.9/100)
- Tornado - Sarkanniemi (78.6/100)
- Vuoristorata - Linnanmaki (76.4/100)
- Pitts Special - PowerPark (70.0/100)
- Hype - Sarkanniemi (66.4/100)
- Salama - Linnanmaki (59.5/100)
- Kirnu - Linnanmaki (45.8/100)
- Ukko - Linnanmaki (43.3/100)
- MotoGee - Sarkanniemi (42.6/100)
- Linnunrata eXtra - Linnanmaki (35.0/100)
- Joyride - PowerPark (27.3/100)
- Cobra - PowerPark (26.0/100)
- Pikajuna - Linnanmaki (12.9/100)
- Neo's Twister - PowerPark (12.2/100)
- Tulireki - Linnanmaki (8.5/100)
- Trombi - Sarkanniemi (3.0/100)
- Toutatis - Parc Asterix (98.5/100)
- OzIris - Parc Asterix (93.2/100)
- Monster - Walygator Grand Est (91.3/100)
- Alpina Blitz - Nigloland (90.2/100)
- Mystic - Walibi Rhone-Alpes (88.4/100)
- Yukon Quad - Le Pal (85.4/100)
- Wood Express - Parc Saint Paul (83.7/100)
- Big Thunder Mountain - Disneyland Park Paris (81.4/100)
- Tonnerre 2 Zeus - Parc Asterix (80.2/100)
- Timber - Walibi Rhone-Alpes (79.8/100)
- Pegase Express - Parc Asterix (78.1/100)
- Namazu - Vulcania (75.4/100)
- Vertika - La Recre des 3 Cures (74.2/100)
- Orochi - Parc du Bocasse (74.1/100)
- Objectif Mars - Futuroscope (71.7/100)
- Twist - Le Pal (71.6/100)
- Crush's Coaster - Walt Disney Studios Park (69.1/100)
- Star Wars Hyperspace Mountain - Disneyland Park Paris (68.8/100)
- Timber Drop - Fraispertuis City (68.6/100)
- Avengers Assemble: Flight Force - Walt Disney Studios Park (67.5/100)
- Taron - Phantasialand (98.2/100)
- Expedition GeForce - Holiday Park (98.0/100)
- Schwur des Karnan - Hansa Park (96.9/100)
- FLY - Phantasialand (96.8/100)
- Black Mamba - Phantasialand (91.9/100)
- Wodan Timbur Coaster - Europa Park (91.6/100)
- Colossos - Heide Park (91.2/100)
- Dynamite - Freizeitpark Plohn (89.3/100)
- Silver Star - Europa Park (89.3/100)
- Karacho - Tripsdrill (88.3/100)
- Blue Fire - Europa Park (88.3/100)
- Hals-uber-Kopf - Tripsdrill (88.2/100)
- Flucht von Novgorod - Hansa Park (88.1/100)
- El Toro - Freizeitpark Plohn (87.8/100)
- Flug der Damonen - Heide Park (87.1/100)
- Star Trek Operation Enterprise - Movie Park Germany (85.4/100)
- Krake - Heide Park (79.1/100)
- Desert Race - Heide Park (78.8/100)
- Winjas Fear - Phantasialand (78.2/100)
- Sky Scream - Phantasialand (77.6/100)
- iSpeed - Mirabilandia (95.7/100)
- Katun - Mirabilandia (95.4/100)
- Oblivion The Black Hole - Gardaland (86.7/100)
- Storm - Etnaland (86.6/100)
- Raptor - Gardaland (85.4/100)
- Shock - MagicLand (84.6/100)
- Inferno - Cinecitta World (80.1/100)
- Freestyle - Cavallino Matto (78.5/100)
- Altair - Cinecitta World (75.6/100)
- Vertigo - Zoomarine (69.1/100)
- Diabolik - Movieland Park (58.1/100)
- Divertical - Mirabilandia (55.9/100)
- Eurofighter - Zoosafari Fasanolandia (54.2/100)
- Fuga de Atlantide - Gardaland (52.4/100)
- Mammut - Gardaland (51.7/100)
- Fun Bob - Haunold Baranci (51.5/100)
- Eldorado - Etnaland (50.2/100)
- Gioco Delle Onde - Osteria Ai Pioppi (46.9/100)
- Desmo Race (Right) - Mirabilandia (46.7/100)
- Desmo Race (Left) - Mirabilandia (46.3/100)
- Untamed - Walibi Holland (98.9/100)
- Goliath - Walibi Holland (93.3/100)
- Fenix - Toverland (92.0/100)
- Troy - Toverland (91.0/100)
- Lost Gravity - Walibi Holland (86.8/100)
- Gold Rush - Attractiepark Slagharen (86.3/100)
- Baron 1898 - Efteling (85.4/100)
- Joris en de Draak (Vuur) - Efteling (83.2/100)
- Joris en de Draak (Water) - Efteling (82.5/100)
- Dwervelwind - Toverland (74.9/100)
- Vliegende Hollander - Efteling (74.9/100)
- Formule X - Drievliet Family Park (72.0/100)
- Xpress Platform 13 - Walibi Holland (61.4/100)
- Dragon Fly - Duinrell (59.4/100)
- Vogel Rok - Efteling (58.9/100)
- Tweestryd (Duurzaam Landbouw) - Wildlands Adventure Zoo Emmen (58.7/100)
- Python (2018) - Efteling (57.5/100)
- Falcon - Duinrell (56.5/100)
- Tweestryd (Eerlijk Mijnbouw) - Wildlands Adventure Zoo Emmen (55.4/100)
- Booster Bike - Toverland (52.5/100)
- Speed Monster - TusenFryd (80.9/100)
- Storm The Dragon Legend - TusenFryd (79.1/100)
- Thundercoaster - TusenFryd (62.4/100)
- SuperSplash - TusenFryd (35.8/100)
- Western-Expressen - TusenFryd (20.1/100)
- Loopen - TusenFryd (12.2/100)
- Zadra - Energylandia (99.7/100)
- Hyperion - Energylandia (98.1/100)
- Lech Coaster - Legendia (96.7/100)
- Abyssus - Energylandia (90.4/100)
- Werewolf - Majaland Kownaty (84.8/100)
- Formula - Energylandia (82.7/100)
- Dragon Roller Coaster - Energylandia (70.1/100)
- Speed - Energylandia (61.8/100)
- Light Explorers - Energylandia (46.9/100)
- Boomerang - Energylandia (46.6/100)
- Roller Coaster Mayan - Energylandia (36.6/100)
- Frida - Energylandia (31.2/100)
- Rollercoaster Wikingow - Majaland Kownaty (30.8/100)
- Energus Roller Coaster - Energylandia (30.1/100)
- Devil's Loop - Legendia (29.7/100)
- Toboggan Run - Gorski Park Rownica (20.4/100)
- Dream Hunters Society - Legendia (13.7/100)
- Scary Toys Factory - Legendia (5.2/100)
- Toboggan Run - Gora Zar (3.8/100)
- Viking Roller Coaster - Energylandia (0.4/100)
- Batman Gotham City Escape - Parque Warner Madrid (97.7/100)
- Shambhala - PortAventura Park (97.2/100)
- Superman la Atraccion de Acero - Parque Warner Madrid (92.2/100)
- Red Force - Ferrari Land (91.0/100)
- Stunt Fall - Parque Warner Madrid (86.0/100)
- Shadows of Arkham - Parque Warner Madrid (84.7/100)
- Dragon Khan - PortAventura Park (83.8/100)
- Muntanya Russa - Tibidabo (75.1/100)
- Tarantula - Parque de Atracciones de Madrid (73.6/100)
- Abismo - Parque de Atracciones de Madrid (73.6/100)
- Furius Baco - PortAventura Park (72.8/100)
- TNT Tren de la Mina - Parque de Atracciones de Madrid (59.1/100)
- Correcaminos Bip Bip - Parque Warner Madrid (54.7/100)
- Tornado - Parque de Atracciones de Madrid (54.6/100)
- Stampida (Red) - PortAventura Park (51.1/100)
- Stampida (Blue) - PortAventura Park (49.8/100)
- Montana Suiza - Parque de Atracciones Monte Igueldo (48.1/100)
- Inferno - Terra Mitica (47.0/100)
- Tomahawk - PortAventura Park (39.9/100)
- Diablo Tren de la Mina - PortAventura Park (37.2/100)
- Wildfire - Kolmarden (98.5/100)
- Helix - Liseberg (97.8/100)
- Balder - Liseberg (93.9/100)
- Valkyria - Liseberg (88.2/100)
- Monster - Grona Lund (87.6/100)
- Lisebergbanan - Liseberg (82.0/100)
- Twister - Grona Lund (79.4/100)
- Jetline - Grona Lund (75.7/100)
- Vilda Musen - Grona Lund (63.8/100)
- Insane - Grona Lund (50.1/100)
- Kvasten - Grona Lund (48.8/100)
- Tranan - Skara Sommarland (48.0/100)
- Luna - Liseberg (41.3/100)
- Delfinexpressen - Kolmarden (22.9/100)
- Gruvbanan - Skara Sommarland (22.8/100)
- Godistaget - Kolmarden (21.3/100)
- Spinner - Skara Sommarland (19.1/100)
- Rabalder - Liseberg (15.8/100)
- Hyper Coaster - Land of Legends Theme Park (94.0/100)
- Red Fire - Korsan Adasi (93.9/100)
- Nefeskesen - Isfanbul (76.0/100)
- Typhoon Coaster - Land of Legends Theme Park (60.7/100)
- Family Coaster - Korsan Adasi (53.2/100)
- Maceraperest - Isfanbul (52.4/100)
- Family Coaster - Land of Legends Theme Park (46.1/100)
- Nemesis - Alton Towers (95.4/100)
- Icon - Blackpool Pleasure Beach (91.1/100)
- Stealth - Thorpe Park (89.8/100)
- The Smiler - Alton Towers (87.7/100)
- The Swarm - Thorpe Park (87.2/100)
- Galactica - Alton Towers (83.7/100)
- Wicker Man - Alton Towers (82.7/100)
- Nemesis Inferno - Thorpe Park (80.9/100)
- Sik - Flamingo Land (80.0/100)
- Oblivion - Alton Towers (79.8/100)
- Megafobia - Oakwood Theme Park (79.6/100)
- Dragon's Fury - Chessington World of Adventures (74.1/100)
- Saw The Ride - Thorpe Park (74.0/100)
- Thirteen - Alton Towers (72.8/100)
- Rita - Alton Towers (72.2/100)
- Storm Chaser - Paultons Park (68.8/100)
- Rage - Adventure Island (65.4/100)
- Speed No Limits - Oakwood Theme Park (63.8/100)
- Mandrill Mayhem - Chessington World of Adventures (62.7/100)
- Big One - Blackpool Pleasure Beach (62.6/100)
Main Analysis
Just as a recap, our question here is "Which European country is best for roller coasters?"!
There are a number of ways in which you could answer this question, as "best" is an open-ended metric that can be assessed from many different viewpoints, but I chose two different angles to approach the question from; the overall quality of a country's top 20 (or entire scoreable selection where this was not applicable), and the quality of a country's top end (i.e. its main headlining draws). Let's start with the overall top 20 quality...
Overall Top 20 Quality (Which country's coaster lineup is the most consistently strong?)
My first question asks which country's lineup is the most consistently strong. To determine this, I looked at the overall top 20 (or entire selection where this is not applicable) to gauge an idea of which countries have the most consistently strong coaster selections.
The first statistical measure I used to determine this is the mean coaster score per country. The mean is a calculated average where all of a country's scores are added together and divided by the number of scores; the formula is as follows:
Mean Score per Country = Overall Sum of all Scores/Number of Scores
When this formula was applied to all the countries used, the ranking was as follows:
| Ranking | Country | Mean Coaster Score out of 100 (to 1dp) | Number of Coasters Used |
| 1 | Germany | 88.5 | 20 |
| 2 | France | 79.6 | 20 |
| 3 | United Kingdom | 77.7 | 20 |
| 4 | Netherlands | 74.1 | 20 |
| 5 | Spain | 68.5 | 20 |
| 6 | Turkey | 68.0 | 7 |
| 7 | Italy | 67.6 | 20 |
| 8 | Denmark | 65.4 | 20 |
| 9 | Belgium | 62.7 | 20 |
| 10 | Sweden | 58.7 | 18 |
| 11 | Poland | 50.0 | 20 |
| 12 | Finland | 48.9 | 18 |
| 13 | Norway | 48.4 | 6 |
| 14 | Austria | 43.1 | 16 |
However, one key flaw of the mean is that it can be heavily influenced by extreme values on the high or low end of the spectrum, and this becomes more apparent as the dataset becomes smaller.
As such, I used another statistical measure to determine the answer; the median coaster score per country. The median is the middle value within an ordered dataset, so in a dataset with 20 numbers as many countries in this dataset are, the median value will be the midpoint between the 10th and 11th values.
When the median was applied to all the countries used, the ranking was as follows:
| Ranking | Country | Median Coaster Score out of 100 (to 1dp) | Number of Coasters Used |
| 1 | Germany | 88.3 | 20 |
| 2 | United Kingdom | 79.7 | 20 |
| 3 | France | 79.0 | 20 |
| 4 | Netherlands | 74.9 | 20 |
| 5 | Spain | 73.2 | 20 |
| 6 | Denmark | 65.3 | 20 |
| 7 | Italy | 63.6 | 20 |
| 8 | Belgium | 61.0 | 20 |
| 9 | Turkey | 60.7 | 7 |
| 10 | Sweden | 57.0 | 18 |
| 11 | Norway | 49.1 | 6 |
| 12 | Austria | 47.8 | 16 |
| 13 | Finland | 44.6 | 18 |
| 14 | Poland | 41.6 | 20 |
While the median is not overly influenced by extreme values on the top and bottom ends of a dataset as the mean can sometimes be, it arguably goes too far in the opposite direction in that it doesn't consider what's happening on the top or bottom at all; it is purely the middle value, and it will stay the same whether the top or bottom value is 5 above/below it or 50 above/below it.
Seeing as both the mean and the median are arguably imperfect measures in their own different ways, I used a final statistical measure that I invented myself... The Matt N Consistent Strength Coefficient. As the mean can be overly influenced by extremely high and low values and the median completely ignores what's going on outside of the midpoint of the dataset, I decided to try and create a coefficient that combined the best of the mean and median and cancelled out the flaws of both methods. The formula for this coefficient is as follows:
Matt N Consistent Strength Coefficient = ((Highest Value + Upper Quartile)+(Lowest Value + Lower Quartile))/2
It considers the top and bottom values, similarly to the mean, but also considers the upper and lower quartiles (the values that are higher than 75% and 25% of the dataset, respectively), so considers the more centralised focus of the median too.
When I applied the Matt N Consistent Strength Coefficient to all the countries used, the ranking was as follows:
| Ranking | Country | Matt N Consistent Strength Coefficient (to 1dp) | Number of Coasters Used |
| 1 | Germany | 177.1 | 20 |
| 2 | France | 161.9 | 20 |
| 3 | United Kingdom | 156.8 | 20 |
| 4 | Netherlands | 148.3 | 20 |
| 5 | Italy | 139.2 | 20 |
| 6 | Turkey | 138.9 | 7 |
| 7 | Spain | 135.4 | 20 |
| 8 | Denmark | 130.2 | 20 |
| 9 | Belgium | 129.5 | 20 |
| 10 | Sweden | 114 | 18 |
| 11 | Poland | 105.4 | 20 |
| 12 | Finland | 101.8 | 18 |
| 13 | Norway | 96.0 | 6 |
| 14 | Austria | 81.0 | 16 |
Finally, here is a boxplot showing the distribution of all 14 countries' top 20(ish) lists, created using Python's MatPlotLib library:
That should be clear enough to see which country is which, but if you have trouble, the countries are listed in alphabetical order from A to Z.
So if we're working based on consistent strength across a wider cross-section of a country's coaster selection, I think it's fair to say that Germany is our comfortable winner, with some combination of France and the United Kingdom taking 2nd and 3rd place.
However, this is not the only metric by which you can assess which European country is "best" for roller coasters. And indeed, one flaw of this metric is that to some extent, it is a bit biased towards countries with more roller coasters; the top 20 of a country with over 100 scoreable coasters will likely be very different to the top 20 of a country with 25 scoreable coasters. I would personally argue that although I accept and agree with the "quantity isn't everything" point, I do think it should still be a factor when deciding which country wins out over another, so to some extent, I would argue that having a lower quantity of scoreable coasters could make the country "worse" for roller coasters overall. Nonetheless, the flaws of this metric should still be taken into account, and I will approach the question from another angle for some balance.
Let's now move onto the other way in which I'm going to try and answer the question... I'm separating this into a new post below, as I'll breach the character limit otherwise.
