Hacker News Reader

Comment by Alex3917 | Hacker News Reader

Those aren't the stats that really matter though. At the end of the day, the question isn't whether the pool is fair or not, but whether the people watching the event on TV still go out and eat McDonalds or whatever afterwards.

Most sports aren't really 'fair' per se. Making them fair wouldn't be cost effective, and wouldn't make for good TV. E.g. everyone knows that some lanes are faster than others in rowing due to the lack of wind screens around the venues. And that's not even to mention the people who spend ten years training only to get violently ill and unable to compete due to the water quality. The reality is that even at the highest levels of competition there is still a lot of luck.

View on HN

> The reality is that even at the highest levels of competition there is still a lot of luck.

It really depends on the sport. For instance, football (soccer) is very random, and this is amplified by the tournament format. But in many other sports, there is no suspens at all. Actually, it would be interesting to somehow evaluate the random factor for each sport.

Even in sports with less variance in performance (e.g. weightlifting), we're only looking at one or very few data points. There are definitely random factors in the final outcome, but it may even add excitement to the events.

That said, I still think fairness or at least perception of reasonable fairness is very important in high-level sports. No one is going to watch the games (or the ads as the GP pointed out) if they think the game's rigged.

Some even argue that the randomness of football (which Americans know by riddiculous name "soccer") is the main factor that makes it the single most popular sport in the world. Some unlucky events end up being remembered for generations.

For example people in Poland are still discussing over the beer whether we would have been world champions if only it hasn't been raining so heaviliy during Poland - West Germany semi-final in 1974.

P.S English is not my language and I'm pretty sure I messed up with conditionals, hope you still get what I meant in the last sentence.

The name “soccer” is just an old British slang abbreviation of “Association Football”, to differentiate it from other kinds of football (that is, games played on foot, as opposed to horseback—compare “footrace”). You can find old references to “Rugby football” and “rugger” as well. Americans kept the term, Brits didn’t.

http://www.todayifoundout.com/index.php/2010/06/the-origin-o...

Exactly. And America isn't the only country that kept the abbreviation. Australia, Ireland, Canada and some countries in Southern Africa use the same term.

http://deadspin.com/map-what-every-country-calls-the-thing-w...

Soccer and rugger survive in parts of upper-class society in the UK, FWIW.

> P.S English is not my language and I'm pretty sure I messed up with conditionals, hope you still get what I meant in the last sentence.

Change "hasn't" to "hadn't" and your conditional is perfect.

(If you're curious, you've made what I would consider a definiteness error three times -- I would say 'the ridiculous name "soccer"', 'discussing over beer', and 'the Poland - West Germany semi-final in 1974'. Those won't affect understanding and you give the impression of being very fluent; there are no other errors.)

For sports are head-to-head match-ups with scores, and have leagues with sufficient games per season, you can evaluate the amount that luck is involved by comparing the points-scored/points-allowed to win/loss over a season.

Merely empirically determine the appropriate exponent to input for the pythagorean win percentage[1]. Baseball uses an exponent slightly under 2, while basketball uses an exponent slightly under 14. The higher the exponent, the less likely a poor team is to defeat a good team.

In fact this can be generalized to sports without points if you have some other way of estimating a team/player's "quality" normalized to 1 being "average" which is all the points-scored/points-allowed ratio does.

[edit]

Note that the Jamesian (points-scored/points-allowed) quality metric doesn't work as well in games where draws are common (e.g. soccer/football), and different quality metrics will yield different exponents, so do not allow apples-to-apples comparisons. For what it's worth, applying James's quality metric to league football yields an exponent of about 1.3[2], which is in line with the intuition of ranking these 3 sports from most to least luck per win:

Soccer, Baseball, Basketball

1: https://en.wikipedia.org/wiki/Pythagorean_expectation

2: http://www.soccermetrics.net/soccer-pythagorean-theory/why-t...

I'm annoyed by NBC's approach to streaming online where you have to login. I suppose they don't care about me, because I don't eat at McDonalds.

Edit: Also you are right.

Reply on news.ycombinator.com