The Partial Control Fallacy (opens in new tab)

I mostly agree with this, but don't discount the idea of only improving by a couple percent. Most things (like grants) are stack ranked, and if everybody has between an 85% and 95%, a 10% improvement could put you at the top of the list.

There's a similar argument about athletics. Sure, the people who win gold medals and get MVP at the Super Bowl have amazing genetics, but they also worked crazy hard to get there. Your hard work isn't going to get you a gold medal, but if they didn't work for it, they wouldn't get it either.

The core message is good, though: if you don't have whole package, don't apply the polish. But if you think you have a shot, don't think you can skate by on what you already have.

> Applying that to the above, and you can see that, if I worked infinitely hard on my essays, I could only make my apps 11% better versus not doing anything at all. (At least to the extent that it really does follow that rubric [the essay being worth ten percentage points of 100], even if I submit a blank essay.)

This is wildly incorrect. Assuming you have a perfect score on every other part of the rubric, the maximum improvement you can get through the essay is 11% of your non-essay score. Without that assumption, the maximum improvement is positive infinity percent.

This is such a useful concept! Surely someone gave it a name much earlier, right?

Edit: The closest thing that came to mind is the idiom "penny-wise and pound-foolish".

Although they may give out a "rubric"; this is just a smokescreen. In most places, Grad School Fellowships are basically awarded based on 1) whose advisors are hotshots and 2) whose advisors are willing to play politics for them. Very little else matters, unless your essay is a steaming pile.

If you strongly desire a fellowship, I recommend researching which professors' students have gotten them in the past. Then, make sure you are on good terms with your PI, that the PI thinks you are a hard worker (whether or not you actually are), and write milquetoast essays that hit all the science buzzwords and PC stuff ("I'm interested in mentoring minorities"[0] - whether or not you actually do any of that).

[0] I actually did mentor minorities and didn't write about that in my essays, taking a more frank approach about what I thought was wrong with minority recruitment, and I lost out on a fellowship to a guy who now is a professor and has exactly zero nonwhite nonasians in his lab. Let's just say they don't really care about honesty and followthrough.

Incidentally, this student also had a research project where all of the data presented in the primary paper were artefacts of the preparation method (It didn't affect the overall conclusion). I confronted the student about this and even went through the process of repeating the experiment with a better prep, resulting in data that actually made sense. Some of the figures were completely invalid, and one of the subsidiary conclusions was wrong. I suggested that he issue a correction and at least stop talking about the subsidiary conclusion, but he continued to present it at several conferences afterwards, and as far as I can tell, the data have not been corrected in the literature. But he did get that fellowship. And now, is getting NIH grants. Your tax dollars at work.

It's a great point, too bad the author is focused on loserisms like winning competitive awards. Most ironic of all is applying for a Thiel fellowship because it's prestigious.

From "Competition is for losers" (http://www.wsj.com/articles/peter-thiel-competition-is-for-l...):

"Always prioritize the substance of what you're doing. Don't get caught up in the status, the prestige games. They're endlessly dazzling, and they're always endlessly disappointing.” -Peter Thiel

I think this is called bikeshedding.

Amdahl's Law: if you toil and sweat to thoroughly optimize some step in a process that takes 10% of the time, so that it takes almost no time at all, you get only a 10% improvement out of it overall.