• Welcome to OGBoards 10.0, keep in mind that we will be making LOTS of changes to smooth out the experience here and make it as close as possible functionally to the old software, but feel free to drop suggestions or requests in the Tech Support subforum!

Butler

I doubt he lucked into TWO consecutive final fours, but hell, I'd rather be lucky than [Redacted].
 
deacvision7 said:
The simple idea that an opponent missing a layup makes you "lucky" is just so damn absurd to me.

Because if you are "lucky" that they missed then that means the opponent was "unlucky" to miss, correct? How is somebody "unlucky" to miss a layup or shot? They aren't. They simply weren't good enough.


This whole notion of "luck" in sports is absurd. Especially from somebody that treats statistics like they are his Bible.

The sooner we eliminate the idea that people are "lucky" when their opponents fuck up, the better.
Click to expand...

Luck is just a layman's term for variance or noise.

What's absurd to me is how anyone could not understand this simple concept.

If you are down by one and foul a 80% free throw shooter, but he misses both free throws an you win, that is luck. It's luck because it's not repeatable. And if it's not repeatable then it has no predictive value, and if it has no predictive value, then it's fairly useless to use as an evaluator.
 
Luck has a minimal amount (if any amount) to do with any contained success, IMO.
 
IAppreciateIt said:
Luck is just a layman's term for variance or noise.

What's absurd to me is how anyone could not understand this simple concept.

If you are down by one and foul a 80% free throw shooter, but he misses both free throws an you win, that is luck. It's luck because it's not repeatable. And if it's not repeatable then it has no predictive value, and if it has no predictive value, then it's fairly useless to use as an evaluator.
Click to expand...

That isn't "luck" at all. Then you are suggesting the guy shooting the FTs got "unlucky." He didn't

He simply wasn't good enough on those two free throws. Luck has nothing to do with it.

It is a lazy term to throw out.
 
deacvision7 said:
That isn't "luck" at all. Then you are suggesting the guy shooting the FTs got "unlucky." He didn't

He simply wasn't good enough on those two free throws. Luck has nothing to do with it.

It is a lazy term to throw out.
Click to expand...

:rjkarl:
 
How is a guy missing two free throws "unlucky" ??



You explain that to me, and I'll come to your viewpoint.
 
deacvision7 said:
How is a guy missing two free throws "unlucky" ??



You explain that to me, and I'll come to your viewpoint.
Click to expand...

Because he's an 80% free throw shooter. A true 80% free throw shooter will only miss 2 free throws 4% of the time. The free throw shooter missing those shots tell you nothing about the true ability of the team in question, which is the entire point.

Nearly all advances in sports statistics, across sports, have been about trying to minimize the impact of luck and instead focus on information with predictive value, which is what matters. And this all starts with the basic understanding that variance, or statistical noise, or luck plays a giant part in what we observe, especially over small sample sizes.

I know that you follow at least a bit of sabermetrics, and those are the basic concepts -- thats why we avoid wins for starting pitches (too many confounding factors, small sample size = too much noise/variance/luck). Same thing with FIP/DIPS vs. ERA, etc.
 
Is it really "luck" if a 80% free throw shooter misses 2 free throws if he made the last 8 or makes the next 8? No. Your "luck" is just poor sampling.

Second, equating luck with variance is incorrect. Variance is what isn't accounted for by stats. Luck is what CAN'T be accounted for by stats.
 
Agree to disagree then.


Nobody that misses a free throw is "unlucky" to miss that free throw. Regardless of their percentage at the line before hand. Independent trials. They just weren't good enough on that particular shot.

Luck, to me, had nothing to do with it.
 
Nothing enrages me more when a soccer player hits a great looking shot and it cannons off the crossbar or one of the uprights.

"Oh, he got so unlucky there." Is always the response by the announcers? Ummm...no he didn't. His shot wasn't good enough and hit the post instead of going in the net. Luck had nothing to do with it! Zero.
 
What about the 39 minutes and 50 seconds that put the team in a position to win if the shooter misses?
 
deacvision7 said:
Nothing enrages me more when a soccer player hits a great looking shot and it cannons off the crossbar or one of the uprights.

"Oh, he got so unlucky there." Is always the response by the announcers? Ummm...no he didn't. His shot wasn't good enough and hit the post instead of going in the net. Luck had nothing to do with it! Zero.
Click to expand...

Perfect example.

One player hits the lazer that bounces off the uprights. Another softly kicks the ball towards the goal that the goalie misplays for a goal.

Which player is, all else equal, more likely to make a goal in the future? Based on the above example, it would be the guy who hit the cannon that barely. Why? Because that has more predictive value -- a guy who kicks like that is more likely to score than the guy who relied on a goalie misplay in that one occurrence.

Call it noise, not luck.
 
deacvision7 said:
Wow. You are clearly insane on this topic.
Click to expand...

I don't know how you could possibly disagree with anything in that post.

What do you think the logic behind the kenpom rating system is?

I also don't understand how these concepts are accepted in baseball, but the exact same concepts aren't embraced in other sports.
 
Last edited:
I guess the difference is that in a game situation it isn't "luck" that determines the outcome.

Just because something is statistically likely to happen doesn't mean if it doesn't that "luck" played a part.


You are "lucky" that a ball bounced to you off a refs backside. You are not "unlucky" that you missed a free throw.
 
deacvision7 said:
I guess the difference is that in a game situation it isn't "luck" that determines the outcome.

Just because something is statistically likely to happen doesn't mean if it doesn't that "luck" played a part.


You are "lucky" that a ball bounced to you off a refs backside. You are not "unlucky" that you missed a free throw.
Click to expand...

How many times do I have to post that "luck" is just a term that means the same as statistical noise or variance?

When something has a 1% chance of happening, that doesn't mean it should never happen -- it means it should happen about once every hundred times. If it doesn't happen once every hundred times, that means that it doesn't have a 1% chance of happening.

If you are an 80% shooter and you miss, that means it fell under the 20%. If we ran that same scenario 100 times, you would make the free throw 80% of the time. We should not give someone credit (or discredit someone) because in a single situation, the unlikely result happened.

I think the soccer example is the best to illustrate the difference between what's predictive and what could be considered noise, or even an outlier result.
 
Last edited:
Luck was my golf ball bouncing off that tree this am and back into bounds. Saying that of course ignores the fact that I hit a shitty shot to begin with and thus needed to get lucky.
 
IAppreciateIt said:
How many times do I have to post that "luck" is just a term that means the same as statistical noise or variance?

When something has a 1% chance of happening, that doesn't mean it should never happen -- it means it should happen about once every hundred times. If it doesn't happen once every hundred times, that means that it doesn't have a 1% chance of happening.

If you are an 80% shooter and you miss, that means it fell under the 20%. If we ran that same scenario 100 times, you would make the free throw 80% of the time. We should not give someone credit (or discredit someone) because in a single situation, the unlikely result happened.

I think the soccer example is the best to illustrate the difference between what's predictive and what could be considered noise, or even an outlier result.
Click to expand...

Generally, I agree with your assessments in this thread, but I would like to make two points.

Just because John Doe has been an 80% free-throw shooter up to a point, does not mean that he will absolutely continue to be an 80% free-throw shooter for his entire career. At this moment in time, he may shoot 80%, but what if he misses his next 10 free throws? His percentage would change. There is no guarantee that the John Doe will continue to perform at an 80% level just because he has to this point.

This ties into another factor - can a human basketball player's performance be accurately predicted? The idea that a player got "lucky" or not kind of assumes that a player will act a certain way regardless of situation. John Doe may be a great free-throw shooter, but what if he has never felt the pressure of an Elite Eight? He might be a 20% free-throw shooter in this situation, but we just don't have the data yet to play it out.
 
PaddyWF said:
Generally, I agree with your assessments in this thread, but I would like to make two points.

Just because John Doe has been an 80% free-throw shooter up to a point, does not mean that he will absolutely continue to be an 80% free-throw shooter for his entire career. At this moment in time, he may shoot 80%, but what if he misses his next 10 free throws? His percentage would change. There is no guarantee that the John Doe will continue to perform at an 80% level just because he has to this point.

This ties into another factor - can a human basketball player's performance be accurately predicted? The idea that a player got "lucky" or not kind of assumes that a player will act a certain way regardless of situation. John Doe may be a great free-throw shooter, but what if he has never felt the pressure of an Elite Eight? He might be a 20% free-throw shooter in this situation, but we just don't have the data yet to play it out.
Click to expand...

Your first point is absolutely correct. I usually try to refer in these hypotheticals as a "true X% shooter" because we can never absolutely be absolutely sure about anything if we base them on our empirical observations.

And that ties into your second point -- it's true nothing can be perfectly predicted, because we are only estimating based on our observations. It is possible, although I'd say very unlikely, that a true 80% free throw shooter in a regular season game might be a true 20% shooter in a clutch elite 8 scenario.

However, the underlying point stands - - said 20% free throw shooter (in a particular situation) only has 20% as his mean, and will perform around that, with the variances getting bigger the smaller the sample size.
 
You must log in or register to reply here.
Back
Top