1 (edited by Max 2021-12-15 09:52 AM)

Topic: 2021 Lemons Results

I posted this link on the unofficial FB group but figure I should post here too since the traditional forum has more permanence. 

Starting mid-summer I began matching the race entries posted by Eric Rood with the final results from speedhive to see how the various cars and models do.  I ended up updating the file with all of the prior races and added subsequent races as they have occurred. 

Note - the data on this file is representative of the results and does NOT include any entries who signed up but didn't make it to race day.  I designated team classes as "DNS" (did not start) if teams show up on the results with zero laps. 

Note 2 - I did spend some time cleaning up the data but, as I put this together for fun, I didn't go too crazy.  I'm sure some stuff is off here and there and I had to make a few calls on naming conventions because of how some teams spelled this and that (like "Camaro" and "Camero"). 


General description of the various sheets: the green sheets are where the race data is combined and pivots/calculations created/done.  The grey sheets show each race's results joined with their entries.  The source of ALL of the pivot tables is the "Stacked" sheet. 


Some Sheet-specific explanations:

Model % and Manufacture % sheets- pivot table by model/manufacturer.  Includes manufacturer, how many times it was raced ("Model Count"), the average year of manufacture (Average of Year), average finishing position, total overall wins, etc.  I added some formulas to the right of the pivot tables showing how each model/manufacturer is weighted and represented in each results category. 

Examples of what the formulas show:

From the Model % Sheet: Miatas made up 8.1% of the raced models, took 15.8% of the overall wins, made up 15.3% of the top ten, 11.6% of the top third, 8.6% of the middle third, and 4.2% of the bottom third.  Comparing the percentage of Miatas in the results to the percentages in each category shows they were: overrepresented in the overall wins, top ten, and top third, slightly overrepresented in the middle third and underrepresented in the bottom third. 

From the Manufacturer % Sheet: Fords made up 13.1% of the raced manufacturers, took 0% of the overall wins, made up 9.5% of the top ten, 13.2% of the top third, 13.6% of the middle third, and 12.4% of the bottom third.  Comparing the percentage of Fords in the results to the percentages in each category shows they were: roughly equally represented in the top third and middle third, slightly underrepresented in the bottom third and in the top 10, and totally unrepresented in the overall wins. 

Model by Class & Manufacturer by Class Sheets: similar data to the % sheets but includes how the cars were classed

Manufacture Details: similar data to the Manufacturer by Class Sheet but includes the models.

Teams Sheet: this is my favorite sheet by far as it lists all of the teams, which races they attended, how they were classed, what cars they ran, and how they did.  The sheer volume of puns employed in the team names is simply staggering. 

Distance Sheet: this is where I calculate how many racing miles were accumulated over 2021.  Almost a million!

Team Numbers: I added this one as I was curious which numbers were most commonly used.  The answer for 2021 was 11.  It was used 17 times in 19 races. 

Stacked Sheet: this is the source of ALL of the pivot tables.  Starting in column R you can see the formulas I used to calculate how each competitor fared against the rest of the field in each race. 

https://1drv.ms/x/s!AsQEf_u4ajkkhALZFdBvoayTiDcz

Re: 2021 Lemons Results

Data is King. I just need to turn this into some pie charts and graphs and I'm in heaven . Thanks for compiling it!

Cordoba

3 (edited by chaase 2021-12-15 12:51 PM)

Re: 2021 Lemons Results

For what it's worth, Elmo's Revenge, Elmo's Revenge 2 and Oscar's Junk Heap are all one team..Oscar is the Javelin but the others are all Saturn. I am trying to get Manny to put things all under one team name in the future.

1992 Saturn SL2 (retired) - Elmo's Revenge -  Class B winner, Heroic Fix winner x2
1969 Rover P6B 3500S(sold) - Super G-Rover - I.O.E Winner, Class C Winner
1996 Saturn SW2 - Elmo's Revenge (reborn!), Saturn SL1  Dazzleshipm Class C x2 and IOE winner
1974 AMC Javelin - Oscar's Trash heap - IOE,”Organizer's Choice" and "I got Screwed" award winner

Re: 2021 Lemons Results

This is really cool, thanks putting it all together!

2021 Gingerman-I:  IOE (Strickland Propane)
2021 Gingerman-II:  20 seconds of footage on the wrap-up!
95 Chevy S-10

Re: 2021 Lemons Results

This is pretty good stuff. I have a spreadsheet with all the results and I appreciate the Figure-By-Thirds method of the torture test, which is how Phil did the first such thing in the Early Days of Lemons for Jalopnik:

https://jalopnik.com/results-of-the-lem … n-30932294

I later updated that for Hooniverse by introducting a pretty simple calculation:

DOMINATION* Factor (DF) = Actual Distance Driven (in Laps) / Average Lap Total

So if the average lap total was 200 and your car ran 220 laps, your individual car has a DOMINATION Factor of 1.100. The general idea wasn't to show which cars would win but which cars tend to be reliable and run more than an average lap total. For the most part, the race winner is between 1.500 and 1.700. Any given model is probably never going to average more than 1.200, but the idea is that with hundreds of data points over the years, you can get a decent overview of Which Cars Suck Least independent of the teams that choose them.

Unfortunately, Hooniverse redesigned their site and it munched my formatting and images as well as the final two posts, but here is the intro

https://hooniverse.com/lemons-torture-t … ate-74-71/

In Phil's 2009 analysis and in my 2013 update, the Volvo 240 was the winner. I've recalculated a few times over the last couple years and both the Camry, Mercedes W201, Mercedes W124, and the Integra have ended up higher at various times. \

YMMV, buyer beware, YOLO, not a scientific study, ICYMI, et. al.

Eric Rood
Everything Bagel, 24 Hours of Lemons
eric@24hoursoflemons.com

Re: 2021 Lemons Results

If anyone really wants to go granular, I'd like to see this done by class winners.  Class C winning cars are going to be buried in these results.

1990 RX7 "Mazdarita"  1964 Sunbeam Imp (IOE 2013 Sears Pointless) 2002 Jaguar x-type (Winner C-Class 2021 Sears Pointless)
Gone bye-bye
1994 Jaguar XJ12 (Winner C-Class 2013 Sears Pointless)  1980 Rover SD1 (I Got Screwed 2014 Return of Lemonites)

Re: 2021 Lemons Results

cheseroo wrote:

If anyone really wants to go granular, I'd like to see this done by class winners.  Class C winning cars are going to be buried in these results.

If I can get through the normal pile of weekly bullshit that comes with this job, I can pull all the Class C and IOE winners forever. Class B, too, but whatever. Everyone knows that Class C is the real-awesome class.

Eric Rood
Everything Bagel, 24 Hours of Lemons
eric@24hoursoflemons.com

Re: 2021 Lemons Results

therood wrote:
cheseroo wrote:

If anyone really wants to go granular, I'd like to see this done by class winners.  Class C winning cars are going to be buried in these results.

If I can get through the normal pile of weekly bullshit that comes with this job, I can pull all the Class C and IOE winners forever. Class B, too, but whatever. Everyone knows that Class C is the real-awesome class.

Class C is where it's at!