Tuesday, January 8, 2013

Crapcanalysis, Part 1: An Introduction

Hong Norrth's Mazda MX-3 is one of a handful of teams to have won six crapcan races, nabbing six 24 Hours of LeMons races on laps. (Murilee Martin photo)

The Rusty Hub loves numbers.

Not just those numbers cleverly painted on the door of your heaps, either. We love statistics, though we would be misleading you if we suggested that we'd ever studied statistics deeper than a college class or two.

But nevermind that. This is the Internet, after all.

A few months ago (Yes...months...), we set out to find which makes and models do the best in crapcan racing, both in the 24 Hours of LeMons and in the ChumpCar World Series. After dozens of hours exercising our mental and search-engine skills, we came up with tables to describe the most common top finishers in a number of categories, but prime among them are the Wins and Podiums column. [And if you're one of the many who care little for overall wins, fret not; we also compiled numbers for LeMons' Class B, Class C and Index of Effluency.]

But before we get to the crapcanalysis (crapcan + analysis...get it?), let me give some introduction to the statistics by way of a few big questions:

Where did you get all of the winners' information?

The majority of this information is easily accessible on the web, if a pain in the butt to sift through. For the 24 Hours of LeMons' winners, the Events and Results page is a great starting place. Using the LeMons' MyLaps results will also give the podium finishers' team names, though you might have to do some fishing for car type. In many cases, Murilee Martin's complete standings posts on Jalopnik fill in those blanks nicely. Since those posts only existed from about 2008 to late 2010, we occasionally found it necessary to compare car numbers from the MyLaps pages with Murilee Martin's uber galleries or other random sources.

The POS Racing/F'ed Up Express/Spin 'N' Out E30 is a common sight in most West Coast crapcan teams' rearview mirrors, having enjoyed success in both the 24 Hours of LeMons and ChumpCar World Series. (Murilee Martin photo)

Unfortunately, ChumpCar's results frequently were more difficult to come by. Some teams like Tubby Butterman Racing and rbankracing.com are just part of the crapcan lexicon and people just know they are a BMW E36 and a bevy of Saabs, respectively. But for many others, the process involved collecting MyLaps data and then using a combination of ChumpCar blog posts, forum posts (from both ChumpCar and elsewhere), assorted photo galleries and Rob Krider's periodic coverage of ChumpCar. Luckily, ChumpCar director John Condren sent us a partial spreadsheet of results by type as compiled by the assorted regional directors.

We're reasonably sure that our ChumpCar list is 90 percent accurate, although there were a handful of teams whose information we just absolutely could not find. These mystery teams were omitted from the totals.

Hasn't this been done before?

As best as we can tell, there have been some attempts at this previously. In 2010, Murilee Martin famously noted in which third of the field many common LeMons types was likely to finish. This was a pretty good (and surprising) list at the time, but some commonly accepted notions have changed. In reality, we would have preferred to simply update these results, but we didn't have a complete list of entries by type and where they have finished. Even if we did, I'm not sure we'd have had the time to compile all that data in a reasonable way. There is also a list floating around somewhere of LeMons wins and IOEs by make and model, but this list does not include podiums or class winners.

As far as we can tell, there is no master list for ChumpCar winners. The list we got from Condren had Top 10 finishers, but it also had results for less than half of the ChumpCar races since the series began in 2009. Condren, however, recently released the Top 10 results for all 2012 races, which can be viewed here.

A new crapcan team purchased one of Robin Bank's successful Saabs, post-rollover. The battered old Swede ran a few races before meeting its end at Road America. (The Rusty Hub photo)

What do these numbers exactly tell us?

Well, that's the real question, isn't it? To be honest, the make/model totals are frequently upset by outliers. Outliers, in this case, refer to teams that have exercised great success with their car type that others have not found. This essentially shows that a given type has potential to win, but it also shows that not just any old boob can pick up that car and win with it.

To illustrate that, I've also added a column for number of teams to win with the given type. Some of the major outliers are household names in the crapcan world: Eyesore Racing (Mazda Miata), Hong Norrth Racing (Mazda MX-3), BoomPowSurprise (Ford Probe), Z-Wrecks (Datsun Z-car), Red Rocket Ratnest Revival/SHOtime (Ford Taurus SHO) and Geo Metro Gnome Racing (Geo Metro...kind of).

As noted above, the list doesn't include total number of entries of a type in every race. If it did, one could calculate the success rate by percentage of Wins/Podiums/Top 10 finishes. But with the means available to us, we felt that Podium finishes was the best way to show which cars are most capable of winning.

The numbers for Class B, Class C and IOE in LeMons can't be considered entirely statistically significant because there is an element of subjectivity involved in the selection of class and IOE. But for all the subjectivity involved, we saw some undeniable patterns that we'll get to with the LeMons numbers.

Sometimes racing a rustbucket yields the exact results you think it will. (Murilee Martin photo)

Is any of this information meaningful?


Not really, anyway (but we enjoyed compiling them). These numbers show some trends, but they are not exhaustive.

Some crapcan racing officials will tell you that winning is the least important part of this sort of racing; the point of all this nonsense is to have fun. (If you're dead-set on competing, you probably already know to look for an increasingly-hard-to-find BMW E30 because they dominate the series.) What interests us more are the unassuming car types that appear farther down the standings, but our opinion doesn't matter. What we've done is given you the data so you can find things that interest you.

So enjoy the data as we post it over the next few days and follow it up with some additional work.

Read Part 2 HERE.
Read Part 3 HERE.
Read Part 4 HERE.

1 comment: