Press Releases - Researching Sponsors: Part 2: Further Research

Researching Sponsors: Part 2: Further Research

Author’s note: This is part 2 of a 4-part series on deriving the sponsor formula. Part 1 on Tuesday discussed the genesis of the project and the early stages of research. Part 3 on Friday will cover the data analysis. Part 4 on Saturday will reveal and discuss the formula itself.

Previously, I discussed how I ruled out a variety of non-factors in the formula and derived the effects of Charisma and PR managers independently and then determined their interaction. The next step was to examine the interaction between PR managers and Man Management.

This is where I had my first major surprise that made me question what I thought I knew about MRC. Previously, it has taken three to four data points for me to begin to identify patterns and form hypotheses as to relationships. With this interaction, the hypothesis jumped straight off the spreadsheet and stared me in the face after my very first data point.

I knew that Man Management had no effect without employees, and I knew how to calculate a sponsor offer with a PR manager (and both Charisma and Man Management at 50). My first test was Man Management 75, PR Manager 25. And when the number came in, it was exactly the same as if Man Management were 50!

I sat and stared at this result for a few minutes, then proceeded to run a few additional tests with Man Management maxed out at 98 (the maximum starting skill for a driver at that time). All produced the same result: the sponsor offer was the same as if Man Management were 50, as I had previously held in my controlled tests.

It quickly became clear: Man Management had no effect on the sponsor offer.

At this point, I was severely struggling with my college classes that had suddenly been shifted to 100% online, and I was forced to put this project on the back burner for a couple of weeks while I got things under control. I ran my last test driver as if he were a normal driver during that time, and collected some data involving ratings that was not of immediate use but would become useful later.

In early April, I started a new driver. The intention of this driver was to test with Charisma and Man Management both at 100 to ensure that the previously derived formulas still held true at the top of the scale. Doing so required training since 98 was the maximum starting skill. Two weeks of training enabled me to get both skills to 100 right as the season rolled over and confirm that the effect of Charisma was fully linear and Man Management still had no effect even at 100.

The driver was not assigned a Rookie Series race for Week 1 of this season, enabling me to take the time to collect some data on multiple PR managers, which was my next topic of inquiry. A couple of additional drivers were also used to collect data on this, and I tried to figure out how to combine multiple PR managers into a single number that could replace the single PR manager in the existing formula. Unlike before, there was no apparent pattern, so I had to use trial and error to find something that fit. I formed a hypothesis that the sponsor offer received full benefit from the highest PR manager skill and diminishing benefits from lower-rated ones. After three attempts, I found a formula involving exponentially diminishing returns that produced the correct number in each case to make the formula work out.

With Charisma, PR managers, and Man Management all sorted out or eliminated, it was time for the biggest factor of all: rating. Unlike previous tests, where I could get one data point per hour, rating required me to spend several weeks with a single driver running races and slowly building the rating up, with only one data point every three days. Therefore, I assumed that this would take the bulk of the season, if not more.

I began in early to mid-May with a driver that would be used to test rating alone. His Charisma and Man Management were both 50, and I was careful to ensure that any employees he hired (he picked up a couple of race engineers) had PR manager scores of zero and stayed that way (since I had proven that a PR manager score of zero had no effect).

Three weeks of racing with rookie quick races and two Rookie Series events, during which he was undefeated (as I put all of his starting skill points into Pace, Racing Line, and mental skills), brought his rating up to 95.926 as I collected seven data points. From this, I analyzed the data and determined, as I had previously done for a single PR manager alone, a constant coefficient that could be multiplied by the rating to give the exact difference between the actual sponsor offer and what the sponsor offer would have been with a rating of 0.000; in other words, the “bonus” provided by the rating.

Look for Part 3 tomorrow, discussing the data analysis that brought me to my final conclusion.