Academics

Here are some objective analytics of my undergraduate performance over the years.

The time series is set as number of months past since my first semester of college (August 2010, 3rd year high school, set as 0) so Spring Semester of 2011 which started in January would be set as 5. Multiple courses in a given semester would stack the datapoints vertically. The lines of fit extend to future months for forecasting an August 2024 cohort, 1 year masters program.

The purpose of these charts are to prove that my academic performance got exponentially better over time and thus you can expect my performance at your school to be much closer to my most impressive and recent semesters. The proof improves with each type shown.

Distribution

Two things stand out in this histogram. One is a point of pride; the other, embarrassment. Some key notes and takeaways:

A is the mode by a large margin
Nearly all of my A+ were earned at 400 level; the remaining at 300 level
Withdrawals (W) and E grades were merged with F
- F may be misleading as most are attributable to withdrawals which often involved health-related reasons

Descriptive Statistics

From here onward, letter grades are shown as their equivalent 4.0 grading scale. Only ASU gave 4.33 for A+ although I received A+ at other institutions. Calculations were done considering grades with 0.33 decimals as 0.3 repeated per actual practice of this scale. Some key notes and takeaways:

- Mean, median, mode: B-, B+, A
- Standard deviation: one letter grade plus two +/- symbol grades
- One sigma from mean is between A+ and D-
  - F are outliers

Absolute

Method

Scatterplot of grades over time
Includes both linear and polynomial lines of fit

Interpretation

R squares for both lines of fit are low but the trend is clearly positive
Improvement over time is better described as exponential than linear
Dispersion gets tighter over time

Weighted

Method

Quality points multiply the grade earned by the number of credit hours.
- Weighs grade by the effort required for each course ("difficulty")
However, class levels of courses (100-400) better represent actual difficulty of courses (not factored here)

Interpretation

R2 for both lines higher than Absolute grade
- improvement to my grades even more certain when difficulty is weighted in
Again, closer to exponential than linear
Dispersion gets tighter over time
When factoring in upper level courses further down timeline - beginning around month 100 - the difficulty-weighted grade improvement would be significantly more noticeable
Nearly all of my A+ were earned at 400 level; the remaining at 300 level
- Combined with law of diminishing marginal returns, getting A+ at 400 courses was exponentially more difficult than As since As have a 6.5% margin for mistakes whereas A+'s have less than half at 3%

Adaptive (Cumulating)

Method

Divides cumulative quality points by cumulative credit hours cumulated
Emulates a single institution (transfers do not "reset" cumulative GPA) but unlike a transcript:
- failed/withdraw courses later passed do not keep highest grade of all attempts (every grade received is factored in)
Each course has lower impact on GPA over time
Polynomial line was omitted because virtually same (both visually and R2 only 0.0002 better)

Interpretation

R2 nearly tripled (expected: autocorrelation)
Very clear improvement trend
Not representative of actual GPA at graduate program

Forecasted Grades (Masters Program)

The polynomial formula of the first chart was used to create this forecast for a 1 year masters program with approximately 10 courses, 3 credit hours each.

=0.0001*([@[Months]]^2)-0.0065*([@[Months]])+1.9704

The data in red are derived from this formula:

So what are the most statistically reasonable grades and GPA you should expect from my time at your graduate program?

A grades, Final GPA 3.81, Summa/Magna Cum Laude

All this is to say:

I know I've messed up in college a lot. I detailed in my About Me many different reasons I got Fs - including how long it took to rebuild my health, but lack of ambition wasn't ever one: I excel (no pun intended) at what I focus on, and for a long time, I chased becoming a world-renowned fighter because I believed I could achieve such success with a high probability. Then, one day, the final time I failed calculus, I knew I had to make a decision. Commit to one or the other 100%, not 50/50.

You can see for yourself which one I chose, when, in the graphs above. I matured. I committed. Taking so long to earn a bachelors was not in vain because time was evidently one of the factors that made me develop exponential potential.

View this incredible TEDx lecture on the myth of the 10,000 hour rule and linear development, pedagogical superiority of non-linear education, and the phenomenon of how early specialists (like so many of your youngest applicants with spotless CVs) do much better short-term, but those who, like me, explored various interests in life before committing to one later than their counterparts did did so astronomically better in the long-term. In fact, by the speaker's empirical observation of every industry's greatest figures, the frequency of non-linear, erratic, volatile, and obstacle-ridden journeys to success dominated the linear and consistent journeys. These greats include Roger Federer, Gunpei Yokoi (GameBoy), Maryam Mirzakhani, Duke Ellington, and Vincent Van Gogh.

Modern financial great Ray Dalio corroborates this with his personal journey to building the largest AUM hedge fund in the world. Watch this inspiring illustrated short below:

Believe in me for your program. The reason I don't resemble the perfect finance employee is because the path I chose better resembles starting my own hedge fund and building it to be the greatest of its kind. Short-term, many other applicants would be an asset to your university as recruiter-favorites. Long-term, I'm the value asset you acquired today that - although you're already great - associated your university's name with "greatest".

Page updated

Google Sites

Report abuse