Monday, May 16, 2011

Some Statistics

Unfortunately, Boston has been cloudy since I've been home, so today's post may be a bit more boring than it might have been in the weather had felt cheerier. Of course, I'm talking about statistics today, so it's possible that no amount of pictures would make that better. On the other hand, I don't think I would have decided to talk about some statistics if I had pictures...

Anyway, you've been duly warned.

I've been using Excel to keep a record of my LEGO collection for some years now. I had a paper record ca. 2000 that listed when I acquired all of my older sets, which remained current until about 2003-4ish (just before the Great Take Apart), and the current Excel document was started in 2007, so it's a mostly accurate record of when I acquired my sets, a completely accurate record of what sets I have, and has also developed into a bit of a statistical chart, as I keep track of the "ratings" my sets get.

Each set is evaluated on five criteria (developed on Classic-Castle in the course of reviewing Castle sets on the official review threads): Parts, Playability, Figs and their Accessories, Design, and "Nostalgia." Obviously, these are all open to subjectivity, and since the point of the whole exercise is more or less to figure which sets are my favourites, that's not really problematic. It's been interesting to compare where sets line up with Lugnet's rating system, and it would seem that (for the most part) once you average out my five criteria, you end up with fairly standard assessments. "Nostalgia" is the trickiest of the five, because while the other four aim for objectivity (while admitting it's a biased process), Nostalgia is there specifically to provide a category to account for sets that I really like (or dislike)... and can't give any good reason for.

On the basis of these statistics, I've also calculated the averages for each subtheme. The numbers are less fair here, since I've only evaluated sets that I own, so a theme with two small (and often mediocre) sets in my collection will probably not rate as well as a theme with 15 sets ranging from small to large, which in turn will not rate as well as a set with only one or two of the largest sets (which often get the best ratings from me). However, as a measure of "favourite," I think this list is fairly accurate, and it's as good a way as any of listing them.

Some other day I might go into a discussion of my favourite or least favourite sets, but for today we'll just look at the favourite and least favourite themes in my collection. Starting at the top, my favourite themes according to this criteria are:

1. Adventurers-the Egyptian Subtheme from 1998, with a 9.85 rating
2. Black Knights, ca. 1992, with a 9.3 rating
3. Designer (the large house sets from the 2000s-2010s), with a 9.25 rating
4. Imperial Guards, 1992-5, with a 9.0 rating
5. Legoland Castle, ca. late 1980s, with a 8.94 rating

The interesting thing about this list, is that the only theme from the 3rd Millennium is the Designer line, probably reflecting the fact that the earlier sets have a higher nostalgic quotient. Indeed, both the Black Knight and the Imperial Guards have been looked at as "Ones That Got Away," and Legoland Castle and the Egyptian Adventurer line would have counted as the same if I had not, in fact, caught them.

At the bottom of the list (of 86 distinctly considered themes--not including Bionicle, Technic, or Duplo) are the following:

82. Alpha Team-Mission Deep Sea, from 2002, with a 7.35 rating
83. Soccer, from the early 2000s, with a 7.25 rating
84. Aquasharks, from 1995-8, with a 7.25 rating (technically a tie with Soccer)
85. Xtreme Island Stunts, from the early 2000s, with a 7.15 rating
86. Junior Pirates, also from the early 2000s, with a 6.4 rating

I do not think it is coincidental that the bottom bunch of sets tends to hail from the early part of the 2000s, nor that Junior Pirates is so abysmally below the rest--Jack Stone would probably be there too, if I'd ever got a set from that line. The Aquasharks are perhaps unfairly in the bottom five, mostly due to the fact that I have only got two sets, and one of them is the abysmal Super Sub, also know as The Aquashark Dart, which has a 6.6 rating.

All of this only goes to show, perhaps, that I am either too much of a nerd or have way too much time on my hands... or both. But I doubt I'm the only one out there who has done anything like this.


  1. How did Jr. Pirates get that high a rating? :-P
    God bless,
    Joel ><>.

  2. I think the thing with my rating system is that it more or less correlates to percentile grades--not surprising, perhaps, as I aspire to be an English professor someday. A 6.4, therefore, works out to 64%, or a "C" grade.

    Perhaps the moral of the story is that... bad as they are, they still have a "passing grade" insofar as they are LEGO--but Jr. Pirates will never get into Grad School with those grades.

    I suppose it's also worth looking at from a relational point of view: relative to a set average of low 8s, it's a bad thing when you're one of a handful of sets that clocks in at a low 6.