Mitch's Wood Coaster Poll '06

Up and ready to roll for this year. Go vote. :)

(driect link to the ballot for the impatient)

Vote'D

Everyone get the word out. The more votes, the better the poll. :)

I would like to be the first person to say, "This poll makes it hard to vote."

I would also like to be the first person to say that "I just modify my ballot from last year. It only takes five minutes!"

Now we can skip over all of that. ;)

Voted for the first time ever. Hate ranking coasters like this but hate not participating even more.

I'm apathetic.

http://cse.umich.edu/

I've always been super high on this poll and still believe in it.

Problem is, For the first time every since I've been taking it for 7 years. I feel one coaster surpasses all others by a longshot.

The thought that it might not will would be a huge dissapointment to me.

Chuck

Brian Noble said:I'm apathetic.

Oddly enough, I'm getting less apathetic about Mitch's poll, and more apathetic about the US Government voting...

Maybe if we had better choices about who to vote for when it came to politicians...and for SOME reason I can't help but thinking about the South Park episode where it came down to "Giant Douche v. Turd Sandwich"....

Charles Nungester said:
The thought that it might not will would be a huge dissapointment to me.

My guess is that Balder is going to crush Voyage pretty handily. Is that what you're referring to?

funny, everyone I know who's ridden both has voyage ahead by a long shot

Being the dork I am, I can settle the early debate.

As of the moment I post this, Mitch is showing 155 processed ballots. Of those, 7 people have ridden both Balder and Voyage.

The head to head?

Voyage 4, Balder 3

(and interestingly on a couple of those polls from people who've ridden both, neither Voyage or Balder is listed as #1)

coasterimage - lordgonchar

Wasn't there a site that dumbed down the voting process for Mitch's coaster poll. The name escapes me right now!

Fate is the path of least resistance.

^coasterfanatics is the site you're thinking of. And yes, it helps a lot.

I feel so depressed, I have nothing new to add to my ballot from last year. Just changing a bunch of y's to n's in the "rode the coaster this year" column. How boring, just rode Phoenix and Twister and Lightning Racer over and over and over...

Lord Gonchar said:
Voyage 4, Balder 3

Pretty much a dead heat, so far, but if around 600 people are going to vote (581 ballots last year) we still have a ways to go. I'm just basing this on hunches and perception but I'd still put money on Balder coming out on top. *** Edited 11/4/2006 1:41:37 PM UTC by matt.***

Yep, it's that time of year again! I really enjoy filling out the poll, dumping it into Excel and sorting and re-sorting until I have it as right as I can get it.

My only gripe is allowing mass groupings, there should be a limit to this. I can see having a couple of coasters tied as a favorite in your top ten, or grouping least favorites at the bottom. But to have ten number ones, ten number twos and so on just defeats the purpose of naming favorites and ranking them, doesn't it? Some ballots do not even declare a number one.

I realize groupings are allowed per the instructions, but you would think there would be rules on how it can be done. In theory someone could vote on 100 coasters and rank them all as number one and it would be included in the calculation.

^The calculations are based on head to head comparisons, so in your example, wouldn't that ballot basically cancel out itself and not affect the results?

I rely on the grouping to get the job done. I have ridden around 100 woodies and other than my top three: Voyage, Tsunami/J2 and Avalache, I have great difficulty picking between my other favorites. There are about 50 I love or like a lot and another 50 I'm kind of indifferent to. ABout the best I can do is group them in five levels of preference. And even that is a stretch for me. I try to do a little additional separation within the levels using decimals, but there are still alot of ties.

Chances are the coasters I have in each level are very different from many other peoples' so there are still plenty of head to head comparisons. For example (and I'll use this one just 'cause I know it will spark controversy), I put El Toro in level 2. So everything in level 1, as well as anything in level 2 with a lower decimal increment will be tallied as preferred over El Toro. It still works. :) *** Edited 11/4/2006 9:14:06 PM UTC by Acoustic Viscosity***

AV Matt
Long live the Big Bad Wolf

Marko - The poll is based on head to head comparisons, but on your ballot only. That's the key.

On top of that the numbers are arbitrary to just your poll. I ranked 74 different woodies from 1-74. I could have also numbered them from 26-100 instead. Or done it only using even numbers between 126 and 274. It would all make my vote exactly the same.

All you'd achieve by ranking every coaster on your list with a '1' is to say that all those coasters are equal.

All the poll does is pull every possible head-to-head combination of coasters and then assign each a 'winning' percentage based on the preference of all the voters who've ridden that combination.

It's deceptively simple and brutally accurate at defining the preference trends of voters.

*** Edited 11/4/2006 9:43:12 PM UTC by Lord Gonchar***

coasterimage - lordgonchar

I trust the poll, love voting in it, but will probably never understand how it really works! From the official web site:

"Each coaster is compared one at a time to every other coaster to see whether more people who have ridden both of them preferred one or the other. A coaster is given a "Win" for each coaster that more mutual riders ranked behind it, given a "Loss" for each coaster that more mutual riders ranked ahead of it, and given a "Tie" for each coaster that the same number of mutual riders ranked above it and below it.

Coasters are ranked by their overall winning percentage (where ties count as half of a win and half of a loss). In the event that two coasters end up with identical winning percentages, the tie is broken (if possible) by determining which of the two won the mutual rider comparison between those two coasters."

Honestly I'm sure this is absolutely brilliant but I'm lost after the first sentence, especially when it involves hundreds of voters determining the rankings of 186 wood coasters.

Let's say only three people voted and I'm one of them, listing The Phoenix as my #1. The other two voters also list The Phoenix as their #1 along with 9 other number ones. In this scenario isn't the determination of a #1 less meaningful than if all three voters had assigned a unique ranking to each coaster, at least in the top 10?

Apologies if this is irritating anyone! :) There should be a book, Mitch Hawker's Poll for Dummies. What other rankings/polls are conducted using this method?

Thanks!

Marko - in your scenario, Phoenix would still be #1 because of your tiebreaking vote. Ranking in groups does dilute the rankings, but the more variety there is in voting style, the less diluted the results will be. If everyone who took the poll had 30 coasters tied at #1, then the poll wouldn't be as informative.

I don't know if there are any other polls like this, but it definitely works for ranking coasters. It's like a round-robin tournament... every coaster must "play" every other coaster. The winner of each matchup is determined by the votes of mutual riders. Some matchups only have a handful of mutual riders (Balder vs. anything), while some have a bunch (The Beast vs. anything). The coaster who "wins" the most matchups is #1.

Thanks Olsor. I would still like to see the rules for the poll evolve more into a 'best in show' approach where each voter or judge is required at a minimum to name a unique #1, 2 and 3, and then group all you want after that.