Quest for the Golden One: Honda Civic Si

TheMakersBox Wednesday, 9/15/2021

The Quest for the Golden One is more than just looking for the magical unicorn, the casting that is perfect in every way and faster than any other.  Every casting is different, just as every track is different, so this is a bit of a quixotic quest.  What we are looking for is clues about what factors make a casting fast so that we can direct our racing efforts.  My current philosophy is that to build a fast car, you have to start with fast wheels.  I am trying to promote a data-driven approach to racing, but with so many factors to account for, I’m afraid it may be a long journey.  As Don Quixote said, “When life itself seems lunatic, who knows where madness lies?”

In the last installment, we looked at the Kia Stinger GT, and while we found the casting was faster than the Mod Rod, we found some outliers that were related to the date of manufacture.  We will return to that, but for now, I want to look at the Honda Civic Si, as it held the podium at Xtreme Diecast’s Case Races for several months.  Unfortunately, unlike the Stinger, it is a bit harder to find since there was only one included in each of the H and J cases.  It took me a while to put together enough examples to test.

As before, each car gets three runs on a sixteen foot straight track with the times averaged.  

For the first time, every result is well below 1.7 seconds, and the average time of 1.673 seconds is also well below that of the Stinger’s 1.694.  I believe it is time to start compiling the data for easy reference:

You will note that I tested both the Kia and the Honda on the same day, so that removes any track setup differences.  You will also notice the high standard deviation of the Kia indicating the variations we saw related to the date of manufacture.  The Kia does have a quicker example (1.653 sec) than the Honda, so it may have a chance to redeem itself in the future, but for now, it is looking like Team Honda!

Let us look at data added to the box plot we used before:

It shows a nice, tight distribution for the Honda which makes it a great candidate if we are interested in the 10 spoke design.  Unfortunately, as I mentioned, it is rather hard to find on the pegs.  

In the prior Stinger data, it was observed that adding graphite had the greatest effect on the slower cars.  I repeated the process of selecting the fastest, the slowest, and a middle-pack Honda Civic and applied graphite, this time Zero Friction Graphite Lube from DerbyDad4Hire

Again, a larger improvement was seen in the slowest car.  I’m not sure this matters to us if we are selecting only the fastest car to modify.

As a practical application of the data we are acquiring, here is the Civic I sent in for the Clutch Mountain Contender Series.  They are on road courses, and all our investigation so far has been on a straight drag strip, so the dynamics are considerably different.  I would point out that the beginning of a road course is usually a drag race to the first corner.  After that, chaos and chance rule the day.

Next we revisit the Kia Stinger to see if we can address the date of manufacture issue.


I haven't seen these posts and need to go back through to read it all but I love the approach to finding the best casting. When building I typically always start with a fast stock car first and then build from there. 

Look forward to reading more!

to join the conversation or sign-up now