Live data from Hacker News

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

danluu.com

71–80 of 191 posts

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

#71
post #66

I don't think I agree with this at all. I can see how you would have this impression with something like a video game or I guess public speaking where maybe a lot of people do it un-seriously and you can be better than most with some intentional effort. It's weird to choose a super mass market game like Overwatch to make that point though, if you were in the 95th percentile of Starcraft players, that would truly be s…

It probably depends on your objective. 352 lb is about the cutoff before you become a competitive weight lifter. Most people probably don't want to do that. Same thing happens in gaming, 95th percentile is probably about where you get to the point of "learned how to play the game up to the competitive level".

The majority of my hobbies are non-competitive. If I get to one day be a 95th percentile hiker, I will be ecstatically happy. However I'd be willing to bet that for my (narrow) niche of software - I am well into the three 9s by virtue that while most backend engineers will have touched what I work on, they haven't done it for 10 years in a demanding environment.

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

#72
I think the way this is phrased is weird. It should probably be rephrased as "becoming 95% percentile at a thing is not that hard". Which I might agree with. Becoming good at a lot of tasks and skills really just comes down to finding the right approach to training and then doing that training. The last 5% is much harder because that's generally filled with truly talented people that also know how to train and are diligent with it.

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

#73
post #58

Earlier quoted context omitted.

I think what the commenter did was reasonable, since it also lets other HNers know that this is a behavior to be on the lookout for. I knew people spammed HN with links to their own websites, but I didn't realize people tried did the same with previously-successful submissions. I agree with the commenter that this isn't appropriate, even if it sometimes leads to good discussions. I appreciate that he posted it here,…

The guidelines are pretty explicit about asking people not to write comments like this. Either way: "@dang" is not a thing. https://news.ycombinator.com/newsguidelines.html

I'm curious to know which of those guidelines covers this somewhat-novel situation. @-ing the mods may not officially be a thing, but I have to say whenever there is a comment starting with that handle that rises up in the comments, it gets attention right quick.

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

#74

Earlier quoted context omitted.

It's easy to be in the top 99% in a competition nobody cares about. Try being in the top 1% of basketball players if you're under 5'9". Easier said than done.

Absolutely. The article hints at this problem by saying 95th percentile is easy but the grind from 95th to 99th is more difficult, requiring explicit effort and reflection on your process. To use a gaming metaphor that hopefully lands, the XP grind from level 1 to 92 in Runscape is the same XP amount as the XP needed to go from 92 to 99 [0]. It is also illustrated with the Pareto Principle and the 90/10 rule where th…

> To use a gaming metaphor that hopefully lands, the XP grind from level 1 to 92 in Runscape is the same XP amount as the XP needed to go from 92 to 99 [0].

Do you get more XP on average when you're a higher level?

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

#75
I'm 70%-ile online chess player. That is average relative to players active on lichess in a given month.

I'm better than most people (maybe top 10% even) that "play a little." I'm (I assume) well below average for adult tournament player. The pool matters. I think if I took lessons and dedicated a serious 100-200 hrs over some period, I might reach 85th in my online pool.

95%-ile is currently ELO 2075... just under where junior masters (CMs) seem to be on lichess. I suspect that extra 150 elo would be a difficult jump... but maybe doable for me and other average players.

Chess is probably structured similar to athletics and other such pursuits. You can probably expect a similar distribution for weight lifting, for example. The average person can probably reach the level where everyone (including most over average people) is "actually trying" simply by being systematic. Where that point is varies, but weight lifting and chess have similar ranges of participation levels... like many hobbies.

In any case... I think unlike chess, it's worth focusing on either strategy or technique... not tactics.

Strategically defining the thing (chess is defined, but "software engineering" is malleable). Meaningfully selecting the pool. Actually deciding the value. Getting to top 10% rec chess player sounds fun. Getting to CM sounds arduous and pointless. There's not much value in being an entry level chess master... or power lifter.

Techniques, like the habit of planning a training plan, creating a strategic plan with goals, timelines and such... that's a valuable asset... very transferable. The tactics... not so much. Meanwhile, you don't need great tactics to get results, and tactical advice is the easiest advice to get.

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

#76
Are there any good platforms out there that people would recommend where you can have your code reviewed? Either paid for or for free.

As this article points out, feedback is an essential part of learning, and open source projects would be a great opportunity to get feedback on something you have sole agency over as opposed to many workplace projects where it's harder to receive feedback for your individual decisions.

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

#77

95%-ile isn't that good.* * The caveat of course being that you dedicate all your available waking hours (not occupied with life-sustaining activities) to training and deliberately practicing a single particular skill/game and also trying to do your best every single time, which is incredibly mentally taxing. It's even more mentally taxing as you get older, since your fluid intelligence diminishes. Also, another auth…

I think the author's point is really that you can outperform a huge slice of people with just a little bit of careful self-reflection and asking for help, instead of putting even more effort into grinding and hoping you'll improve.

Or, even: by throwing some deliberateness in to things we're doing, we can avoid a lot of frustration.

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

#78

Are there any good platforms out there that people would recommend where you can have your code reviewed? Either paid for or for free. As this article points out, feedback is an essential part of learning, and open source projects would be a great opportunity to get feedback on something you have sole agency over as opposed to many workplace projects where it's harder to receive feedback for your individual decisions…

Send PRs to open source projects you’re interested in?

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

#79

I would have to agree about 95th percentile being not that difficult. A streamer that I occasionally tune into (Asmongold) says that if you are seeking information outside of the game such as watching streams of the game, that makes you a top 10% player automatically. Most people are passive consumers and participants in their activities and that isn't exclusive to the individual. I am a passive consumer & participan…

> Most software engineers aren't listening to podcasts about software engineering. If you are religiously listening to and learning from podcasts, that makes you more suspectible to being a higher percentile performer in software engineering.

It's generally true that seeking more information can lead to increased knowledge, yes.

But I've also met a lot of people in my career who can confidently recite factoids from hundreds of podcasts, books, and blog posts they've read but then struggle when they have to tackle real-world problems. Podcasts don't substitute for experience and someone with real-world experience is going to run circles around someone who merely listened to a lot of podcasts or read a lot of books.

Podcasts have also become a trap for inexperienced engineers in recent years. It's easy to listen to a lot of podcasts with guests who are trying to flex their knowledge or look extra smart by claiming their subject is the only way to do something. Then you end up with juniors trying to make complex microservice architectures with over-engineered single page app frontends for trivial tasks because that's what they've been learning about from podcasts. Podcasts try to capture the most exciting topics, whereas many engineering problems are better solved by boring old technology. Without experience to know the difference, podcasts can really lead people astray.

> A streamer that I occasionally tune into (Asmongold) says that if you are seeking information outside of the game such as watching streams of the game, that makes you a top 10% player automatically.

Influencers telling their followers that following them will provide a leg up on the competition is the oldest marketing trick in the book. :) I'm sure some amount of stream watching can provide some learning, but for fast-twitch games I'll put my money on the player who has more hours of practice over the player who watches a lot of streams of other people playing.

Post reply on HN