Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

95th percentile isn't that hard to reach

danluu.com

Post reply on HN
Print view View original

61–70 of 299 posts

Re: 95th percentile isn't that hard to reach

#61

Post by mandelbrotwurst »

libria wrote: post #50
5cott0 wrote: post #45

There is something so depressingly robotic about the obsession with being the best at everything all the time. Career, income, diet, exercise, lifestyle, hobbies; can't even have fun anymore without spending days holed-up researching esoteric hyper-optimizations. Some of you need to relax and stop sucking the fun out of everything.

He leaves room for that. > And for games like Overwatch, I don't think improving is a moral imperative; there's nothing wrong with having fun at 50%-ile or 10%-ile or any rank.

Leaving room for being less competitive in games is not the same as doing so for non-games.

Re: 95th percentile isn't that hard to reach

#62

Post by SkyBelow »

tyingq wrote: post #9

95th percentile for an SAT score is something like 1450. That's fairly impressive to me. Feels like scoring higher might often have more to do with effort than ability. Is that excluded from the line of thought in the blog post because it's among "people who regularly practice?" Or is it that effort vs ability thing? Not impressive because you were born smart? In a totally "no effort" arena, 95th percentile height fo…

The 95th percentile of SAT seems quite different than the 95th percentile of people taking in the first time and doing no special prep. Sadly that second score is near impossible to measure because it would be useful to help many kids compare their first score with no dedicated prep to a more realistic percentile (and better encourage them to work to improve their score).

Re: 95th percentile isn't that hard to reach

#63

Post by Spooky23 »

bryanrasmussen wrote: post #20
tyingq wrote: post #9

95th percentile for an SAT score is something like 1450. That's fairly impressive to me. Feels like scoring higher might often have more to do with effort than ability. Is that excluded from the line of thought in the blog post because it's among "people who regularly practice?" Or is it that effort vs ability thing? Not impressive because you were born smart? In a totally "no effort" arena, 95th percentile height fo…

Perhaps the 95%-ile is actually good when it is something that almost everyone has to do. Everyone in school is practicing at least a bit on reading comprehension, and the SAT is across lots of schools, if you are the in 95%-ile of millions of people it might be more impressive than being in the 95%-ile of tens of thousands of people.

Remember that everyone has different goals. My cousin was heartbroken when she got a 1570 on the old SATs, as her dream of going to some university was crushed, as she was competing with peers in an elite high school. She had to settle for Brown or something, and was convinced that life was over.

For me, I didn't have the desire or resources to do that. I got my 1400 and was able to choose from any of a dozen schools. That meant that being a white shoe attorney or McKinsey consultant was out of my reach. Life is good.

On another note, my son is passionate about little league. He's good, has fun, has a competitive spirit, has made great friends, etc. It's a great thing. He's probably in the 75th percentile of players at his age and region. So unfortunately, major league baseball dreams are unlikely to be fulfilled.

This article is similar. Being in the 95th percentile of people playing some video game is interesting, but isn't something that I would have any desire to do, and probably is a fools errand unless you're looking to compete at a professional level. I'd like to be in the 99th of people having fun playing a game -- the time and mechanics of getting to be an elite player is probably in opposition to that goal!

Re: 95th percentile isn't that hard to reach

#64

Post by spdionis »

You can look at the rank distribution of every competitive game and this concept becomes obvious.

I've played Dota for many years. Managed to reach the Divine rank last year which at the time iirc started at the 95th percentile (maybe 98th?).

Any player who has played Dota for some time knows that while Divine players are good for average player standards, compared to any semi-pro or top 1000 player (not even pro) they are god-awful. To the point where they basically cannot compete in any aspect of the game.

Re: 95th percentile isn't that hard to reach

#65

Post by eshyong »

Traster wrote: post #51

I find it really very frustrating that the author talks about the "95%" percentile and then does very little to actually justify why he's talking about the 95% percentile and how he's measured it. He's got some anecdata about what he views in overwatch, but you actually need some statistical techniques if you want to justify a statistical number. Of the players in the top 5% how many times was it observed that they c…

I agree that the author isn't totally rigorous in his methodology, which I understand is frustrating, but I think you're missing the bigger point he's trying to make. As I interpreted it, the article is mostly about self-improvement and the fact that even at 95%, there's still a lot of room for optimization between 95 and 99. He's saying, "hey, even this competitive Overwatch streamer makes game-losing mistakes, but it's very easy to improve on these things, if you have the right mindset". In other words, instead of focusing on your rank, you should be focusing on the low-hanging fruit, no matter where you land on the spectrum of talent.

Re: 95th percentile isn't that hard to reach

#66

Post by libria »

Relevant to the article, but more on the tooling side, a question (especially for you consultants):

What tool do you use to track where your time is spent?

This can either be for the "film study" of how I work or for billing clients in 15 minute increments. I've tried apps that take screenshots every 60s and others that track accumulated time on foreground application.

The end result being, same as Dan's friend, I'd like to know where I'm wasting time and how to improve. Forget the 95/99% numbers, I just want +10% this week, +8% next week, etc.

Re: 95th percentile isn't that hard to reach

#67

Post by datashow »

LeifCarrotson wrote: post #52

One other area where this is true and might be similarly easy to analyze is competitive running. A local road race 5k might have a thousand entrants. You might feel bad about occasionally walking and finishing in 500th place, but that's just in the subset of the population that participates in those events. The biggest race in our 1,000,000-person metro area draws about 20,000 entrants (many from outside that area),…

It is true that regular practice can easily put you on very high percentile among participants. So it is easy to be on 95% in one or two things.

On the other hand, for the same reason, it is very hard to be on 95% in multiple things at the same time, because you won't have time to regularly practice on multiple things.

For the same reason, I would also argue that for regular people it is also hard to be on 95% in just one thing outside of work, because people need to take care of their family.

Re: 95th percentile isn't that hard to reach

#68

Post by tzone »

As they say: "there are levels to this shit". It is very noticeable in most individual games and sports that as you go to the top the difference between skill level is huge, even between top 5 percentile of people or between top 1 percentile of people. So in a way not only 95 percentile is no good, a lot of times even 99th percentile may not be that good either in grand scheme of things.

If we take two examples of chess and boxing. In boxing there is huge difference between top 5 level fighter in a weight class, compared to top 20 level fighter, and huge difference between top 20 and top 100, and another huge difference between top 100 and regular pro boxers. And there can be 1000s of pro boxers for a weight class. And there is another huge difference between a pro boxer and random guy who trains boxing in a gym, and another huge difference between random guy who trains boxing and just a random Joe from the street.

The level differences are so huge that, if a top 5 fighter is fighting someone from top 20 it is generally considered a one sided beat down and a mismatch, with 10:1 odds for a favourite.

There is very similar situation in chess. A Fide Master can crush people who have studied chess and play as a hobby while just playing blind folded. (Not even talking about average Joe, even I can beat most average people playing blind folded, and I am not any kind of Master). International masters (IM) crush Fide Masters like its nobodies business. At next level, Grand Masters are a level above most IMs. Being a Grand Master is considered a huge accomplishment, not just top 1% but probably top 0.001%. No matter how hard you work and how dedicated you are, unless you are also an exceptional talent, you can't become a GM in chess. However top 10 players will crush below average GMs with ease. That is how crazy the level difference is in this sport.

I think this type of phenomenon is more prevalent in individual sports/games compared to team sports. (Especially larger team sports). I think that is another reason why this phenomenon is not always as visible in "real life" situations, because most of the real work is done in groups and sometimes in very large groups which averages out potential huge differences between individual abilities. In a team/group environment there are faster diminishing returns on how much more a single individual can do as part of a group. Of course there are still big differences but in that case top 95th percentile is probably good enough to get you in the top team, especially if you have other skills to complement your abilities (like better communication, better team work, etc)

Re: 95th percentile isn't that hard to reach

#69

Post by eshyong »

gizmo wrote: post #46
lordleft wrote: post #40

I like that Dan acknowledges that he sometimes get ridiculed for his ideas, but continues to post whatever is on his mind. I think that's admirable, because he might be occasionally very right despite the opprobrium of the world. I sometimes feel that my generation (Millenials and younger) are super cagey and obscure their real views because of a fear of censure.

He literally says the opposite: > I have about 20 other posts on stupid sounding ideas queued up in my head, but I mostly try to avoid writing things that are controversial, so I don't know that I'll write many of those up.

I don't think the fact that Dan posts the least controversial ideas detracts from his point; publishing one of them is admirable in itself. Putting yourself out there is better than not posting anything at all.

Re: 95th percentile isn't that hard to reach

#70

Post by crispyambulance »

5cott0 wrote: post #45

There is something so depressingly robotic about the obsession with being the best at everything all the time. Career, income, diet, exercise, lifestyle, hobbies; can't even have fun anymore without spending days holed-up researching esoteric hyper-optimizations. Some of you need to relax and stop sucking the fun out of everything.

> relax and stop sucking the fun out of everything.

Yep. It's important to try new things and put yourself "out there". That means you're going to fumble on a regular basis.

The best way to master something, in the absence of a god-like mentor, is through endless repetitions of effort, failure, reflection and exploration. The concept is as old, tried, and true as the story of Odysseus.

Post reply on HN