95%-ile isn't that good (2020)

Distributional assumption - The assumption is that the mean and std dev of the participating class is in line with the general population.

Overwatch and public speaking sneakily satisfy this assumption. Everyone is forced to speak in public at some point, and overwatch is the archetypal 'normie' game of the of the past few years.

Dan then moves from these 2 scenarios to programming, where this assumption doesn't work. There are many formal and informal filters involved in making it an employable programmer. (Say complete a BSCS or equivalent experience). There are also supply-demand factors, where the top %ilers are presorting themselves into CS like no other time in history.

Landing a FANGesque job straight out of college is already putting you in the 95th percentile, because of sheer number of filters that one person passes through. Other situations with strong filtering include inaccessible games like Dota2 or high barrier of entry sports like climbing.

Now, I agree with Dan's title and the contents, but for a different reason. There is perpetual headroom in the pyramid of excellence. Eventually, those better than you at a particular thing go from being a number to a tangible list of names. But, whether that should mean anything is a purely personal decision. The argument that anyone can go beyond x-percetile through deliberate practice is always true until it isn't. That's because we tend to take our personal IQ, resilience, discipline and acumen for granted.

So, I would rephrase that as:

'When the distributional assumption holds, 95th percentile would not be considered noteworthy by any institutions or people that the average FANG/HN reader would find worthy of respect.'

Doesn't sound as useful with the qualifications.

P.S: something something nuance for large scale communication is impossible. Criticizing danluu for one post lends credence to another post of his. Such is life.

With the way the author described Overwatch, it sounds like the equivalent to 95th percentile for chess would be more 50th percentile for chess, because in order to reach around that level, one would have to put in that type of work as described to purportedly reach the 95th percentile in Overwatch.

Based on my understanding with the Lichess's blitz rating system (which of course can't be directly mapped to any other rating system), I expect that players rated 2175 Glicko in Lichess blitz could roughly be getting close to 1800-2000 Elo with FIDE if they played FIDE classical tournaments and were rated accordingly.

For comparison, FIDE's rating floor is 1000, but newbie players would likely remain "unrated" because they "aren't (yet) good enough" to be 1000 after 5 rated games. Meanwhile, GM Magnus Carlsen currently has a "live" FIDE rating of 2863.9 Elo (live, as opposed to published monthly rating).

As a further comparison, popular Twitch chess streamers WFM Alexandra Botez and WFM Anna Cramling have FIDE ratings of 2020 and 2057 Elo respectively, and Chesscom blitz ratings of 2195 and 2084 glicko respectively. I think they would simply wipe the floor at most (but not all) chess clubs that exist in the world.

There really aren't a lot of players who go through my chess club who are at or near this level. It really takes a lot of work to get anywhere close to that standard. The amount of work it takes for most to get that good would require chess to be their main hobby, if not something more. That, or they have to start it from a young age and also be quite obsessive about it.

Exact numbers aside, this level strength in chess isn't to be sneezed at, and requires significant investments of time - often over a "lifetime" (where lifetime = 10 or more years of a high level of dedication, focus, and good guidance), which for many young players, it is most of their lifetime. Most chess players will never reach this standard of chess, despite following all the basic advice as per the post. So I just find top 5% / 95th percentile being bandied about to be applicable for any field as somewhat laughable or fabricated. Maybe the figure is real for Overwatch, but for things like chess, speedcubing, and instrumental/vocal music, it just doesn't quite work that way as much as many would wish it did. Yes, the overall message of practising and learning is good, but when extraordinary claims such as "95%-ile isn't that good" get thrown about with Overwatch being used as primary evidence, the substance isn't much more convincing than typical clickbait. Survivorship bias?

When a game is just outright complex like SC2 in its decision making(hundreds of units and structures that could be individually commanded), all the skills needed to reach 95%ile boil down to rote technique: "good builds", "meta strats", etc. This is where most video games sit, since this kind of overt complexity is something you can learn in little nibbles: A is more powerful than B in situation X.

In comparison, games like pinball have relatively little going on in their vocabulary: there are plenty of techniques and rulesets, but much of the good strategy boils down to "hit accurately" and "recover successfully", and those are entirely sources of error, reduced only through many hours of play to find methods of hitting and recovering with good expected value and low risk - sweet spots on a playfield and a deep understanding of the physics to enable combo setups. If you have accuracy, you need only have a small layer of rules and shots knowledge to smoke tournaments. With little to consciously memorize to overcome the skill gap, casual players conclude that pinball is "just random".

Sometimes I feel that people try to over-optimise grocery shopping. Basically, my process is simple and works: 1. realise I need more food, 2. grab keys, phone (and maybe wallet) and reusable bags, 3. go to the supermarket, 4. browse the usual stuff I'm interested in and look for what's on sale, 5. grab it off the shelf if I want it.

It's really not that complicated, nor at all stressful to shop without a list. Not everything in life needs a list.

In my mind, the top 20% are people who actually play. If you are stuck anywhere below that, you aren't even trying to improve. If a leaderboard says you haven't even cracked top 50%, then that leaderboard is definitely cutting players that don't play much from the bottom of that list. In most competitive games, the bottom 80% is just casuals who play a few times and stop, or people who only play the game while high.

Then the top 5% is when you start understanding the game at a competitive level. The very start.

Then the top 1% is where you can consider yourself good at the game, but, still, there are a ton of fairly bad players that play enough to get here but don't understand what they're doing. It's a combination of grinders and skilled players.

Then the 0.1% is where you have good players. Not pros, but definitely, unarguably, good.

Then you have pros.