Obsession with stats?

[kuniva_dAMiGhTy]

stats never lie, its only when people (intentionally or not) misuse them, or ignore context, that you get into trouble.

This.

[ILLsmak]

stats never lie, its only when people (intentionally or not) misuse them, or ignore context, that you get into trouble.

lol, the reason stats never lie is because of what I stated. It already happened. So, it's the same as saying, "One time I dunked." Even "I can usually dunk." "I've dunked every day in my life."

That doesn't mean it will happen again. So, how can we use stats to compare players or predict outcomes? How can we use them to see who is good? We even have used stats to compare cross generation. CONTEXT is what stats lack, as you said.

But there are no stats without context. To me, context is more important than outcome. I will say, again, that everything that leads up to a recordable result is NOT recorded. That makes stats seem pretty useless other than "this is what happened on that day." Nobody can deny that. Especially keeping in mind that they most often reference end-game stats or season stats, so what I said about the end result is even more magnified.

And whenever something deviates from stats (which happens nearly every game) they call it an anomaly.

Then you've got players saying "I won't shoot this shot because it will have a negative effect on my stats..."

-Smak

[DMAVS41]

This is untrue. ALMOST everyone does. And for an "unbiased" person, it would come down to something like... overvaluing one aspect of the game that they feel is important. Like, say, pick and roll defense. You could say X player can't guard shit because he can't defend a pick and roll. But then someone would say BUT HE GETS 3 STEALS AND 3 BLOCKS. Nothing is proved on either side.

Rebounding is also really out of wack. I don't really care about overall rebounding numbers. Plenty of dudes get like 6+ bullshit rebounds a game. A rebound, to me, is something you have to fight for. That's why I believe offensive rebounds are a much better way to measure, but if you are missing your own shot and rebounding it, again, that's misleading.

The ability to look at something and understand it is extremely rare. We have a tendency to break everything down to smaller pieces, but you can't always do that. A stat like PER is a failure because it doesn't account for people TRYING to be efficient at the expense of their team. Most people can tell when a player is good. He just shines above all of the others. that's how you can tell through a lot of mix tapes of players who is really good... because most mix tapes look the same, doing the same shit... but you can still observe differing nuances.

Nuances can not be measured, mentality can not be measured... by stats.

Stats are like history. It is the result. Not what happened. If that makes sense...

-Smak

You started with..."this is untrue"

That is wrong. Sorry, we all have individual biases that cloud our judgment when watching a game. You simply aren't paying attention enough to notice everything. And, most likely, you are pulling for one team over another...even if it's just because you want a close game or something. And those biases cloud judgment...that is an accepted fact in the psychology community.

That doesn't mean that stats are the only way to evaluate basketball, but you can't ignore them.

Sorry, you simply can't ignore rebounding stats because you value offensive rebounding more.

You have it correct. Stats merely tell you what happened...getting 14 rpg vs 12 rpg does not mean one player is a better rebounder for getting more. Context is needed.

But that surely is not a problem with the stats...it's a problem with someone trying to use rpg or ppg to tell you something.

[josh99]

Not many people watch all 1,230 games of the season and so its quite hard to quantify how good players are from teams that you aren't a fan of. Plus personal bias comes into account as well.

For example if you just watched a couple of Golden State games and they happened to be ones where Steph Curry went off and you didn't look at the stats you would think that he is the best player in the league. After looking at stats you would realize that he doesn't play like that every night.

That being said stats are often misleading and deceptive so you need to be careful when using them to evaluate players.

[Horatio33]

American sports seem to be more stat obsessed, although football in Europe seems bo be catching up. We are a year behind here in Europe,and media don't use stats as much here as American media does. Not knocking America, just saying that's how it is.

[DMAVS41]

Not many people watch all 1,230 games of the season and so its quite hard to quantify how good players are from teams that you aren't a fan of. Plus personal bias comes into account as well.

For example if you just watched a couple of Golden State games and they happened to be ones where Steph Curry went off and you didn't look at the stats you would think that he is the best player in the league. After looking at stats you would realize that he doesn't play like that every night.

That being said stats are often misleading and deceptive so you need to be careful when using them to evaluate players.

Yea. That is the other side of it. Even if you watched all 82 games for one team...you still are at most watching 10 games of other teams...honestly, probably less.

So your sample size is so damn small it's useless to begin with...let alone the personal biases we all have.

[ILLsmak]

You started with..."this is untrue"

That is wrong. Sorry, we all have individual biases that cloud our judgment when watching a game. You simply aren't paying attention enough to notice everything. And, most likely, you are pulling for one team over another...even if it's just because you want a close game or something. And those biases cloud judgment...that is an accepted fact in the psychology community.

That doesn't mean that stats are the only way to evaluate basketball, but you can't ignore them.

Sorry, you simply can't ignore rebounding stats because you value offensive rebounding more.

You have it correct. Stats merely tell you what happened...getting 14 rpg vs 12 rpg does not mean one player is a better rebounder for getting more. Context is needed.

But that surely is not a problem with the stats...it's a problem with someone trying to use rpg or ppg to tell you something.

But what are stats worth? You said you can't notice everything that's happening, why not? How many official score keepers are there? How are stats measured? It could be said, then, that stats are equally misleading because stat-keepers can't notice everything.

I really don't have biases. I don't give a shit who does what. I think that, like I said, I value certain parts of the game more than others... am impressed by things that people don't think are impressive. Might not be what produces "statistical results." Is that a bias, though? You say it as if I would watch two players and like one better... because of some preconceived idea not because of what I saw. I definitely disagree with that. I would say most people who do that are of lower intelligence.

There are also scouts, you know?

The underlying problem is that if you can't compare 12 rpg to 14 rpg and say someone is the better rebounder because he had better stats, then what use do stats have? To say that someone who gets 14 rpg is better than someone who gets 6rpg? You don't think by watching a game you could pick that up?

I don't ignore stats. I have seen many stats. When I was a kid, I used to look at stats a lot... I was autistic with stats at like age 12-13. Then I reached ENLIGHTENMENT.

but yea, I can see why you wouldn't understand because whenever someone says something against what is written in print, they are labeled crazy or, at least, irrational.

-Smak

[DMAVS41]

But what are stats worth? You said you can't notice everything that's happening, why not? How many official score keepers are there? How are stats measured? It could be said, then, that stats are equally misleading because stat-keepers can't notice everything.

I really don't have biases. I don't give a shit who does what. I think that, like I said, I value certain parts of the game more than others... am impressed by things that people don't think are impressive. Might not be what produces "statistical results." Is that a bias, though? You say it as if I would watch two players and like one better... because of some preconceived idea not because of what I saw. I definitely disagree with that. I would say most people who do that are of lower intelligence.

There are also scouts, you know?

The underlying problem is that if you can't compare 12 rpg to 14 rpg and say someone is the better rebounder because he had better stats, then what use do stats have? To say that someone who gets 14 rpg is better than someone who gets 6rpg? You don't think by watching a game you could pick that up?

I don't ignore stats. I have seen many stats. When I was a kid, I used to look at stats a lot... I was autistic with stats at like age 12-13. Then I reached ENLIGHTENMENT.

but yea, I can see why you wouldn't understand because whenever someone says something against what is written in print, they are labeled crazy or, at least, irrational.

-Smak

You do have biases. Everyone does.

In terms of the rebounding thing. You could look at rebounding rates...on both sides. You could look at the rebounding rate of his team while he's on the floor versus off. You could do so many things all of the rebounding stats...my point is that you can't just pick out one and run with it.

But looking in depth at rebounding while using all the stats is far better than you just claiming that someone is better than the other because "you watch the games"

Because you really don't watch the games. You have an absurdly small sample size even if you are a die hard NBA fan that spends 5 hours watching a night. There simply isn't enough time to watch it all.

[ILLsmak]

Not many people watch all 1,230 games of the season and so its quite hard to quantify how good players are from teams that you aren't a fan of. Plus personal bias comes into account as well.

For example if you just watched a couple of Golden State games and they happened to be ones where Steph Curry went off and you didn't look at the stats you would think that he is the best player in the league. After looking at stats you would realize that he doesn't play like that every night.

That being said stats are often misleading and deceptive so you need to be careful when using them to evaluate players.

Well, by watching Steph Curry go off, you can get an assessment for his psyche... and know that he would likely have games where he shot like shit. and maybe even say that in games he shoots like shit, he would play more poorly overall.

I always look at box scores and try to imagine what "really happened." It's just sad that people believe it's the Bible truth or that its irrefutable when everything happens because of something. It's a chain of events that leads to the end result. Basketball skill, like anything else, is intangible. That's why people talk about empty stats.

-Smak

[ILLsmak]

You do have biases. Everyone does.

Just as everyone can't notice nuances? Do you have biases when you watch a ping pong game? I think you are misusing the word. I have a point of view. And my point of view is defined by what happens. If I watch a player for the first time and he chucks up a 30 footer... I might be like wow why did he do that? I would need more information to judge him, though. I wouldn't be biased because of what color jersey he was wearing, though.

If I was in the 99.9th percentile of "most unbiased people" since "everyone is biased", do you think that it would still cloud my judgment so much that I couldn't make a rational observation?

-Smak

[DMAVS41]

Just as everyone can't notice nuances? Do you have biases when you watch a ping pong game? I think you are misusing the word. I have a point of view. And my point of view is defined by what happens. If I watch a player for the first time and he chucks up a 30 footer... I might be like wow why did he do that? I would need more information to judge him, though. I wouldn't be biased because of what color jersey he was wearing, though.

If I was in the 99.9th percentile of "most unbiased people" since "everyone is biased", do you think that it would still cloud my judgment so much that I couldn't make a rational observation?

-Smak

Biases can mean any number of things. You already showed yours. You think offensive rebounding is more important than defensive rebounding. That is your bias coming through. I think defensive rebounding is more important.

How are you going to prove to me that offensive rebounding makes a better rebounder without resorting to some type of evidence or statistic?

I'm not talking about someone being a huge fan of a player and it clouding their judgment. I'm talking about the things we all have somewhat hard wired into us just like the above example.

Honest discourse with the use of as much information as possible is the best way to discuss things in life. Not using information because you don't think it's valuable is just yet another one of your biases. You are against stats...LOL

But I can't help you if you think your opinion on the best rebounder in the game trumps what using all the data combined with opinion and honest discourse and debate can give you.

[Kovach]

Advanced stats, per etc etc, I am from England, and we don't really bother with stats when stating who we think are the best player in footbal/soccer for example, so is this part of American culture or is it specific to basketball, I don't follow other sports in America.

Yes it is pretty much part of their culture, it was derived from the baseball stats obsession. It also feeds the desperate urge of an average human to have an opinion on everything, including things they know nothing about or have never actually seen.

[cotdt]

It should never be all or nothing, but individual biases just come into play too often unless some objective measures are used.

And a lot of time the biases are things we aren't even aware of.

I argued for years on here that Kobe actually doesn't excel in game winning situations to near the extent his fans claimed. We went back and forth and back and forth...and then the stats came out and ended it. Because they are objective measures and once you tell someone that their "guy" shoots 30% in those situations while other players shoot much better....that's it.

So in a case like that, stats are great.

Where they get poor, is when someone would want to use a single stat to quantify how good a player is. Basketball has been and always will be a mixture of art and science. There are certain things that simple stats can't quantify.

For example, Dirk's impact on the floor on offense without him doing anything but standing in the corner can be hugely beneficial to a team. Sure, you might capture that in offensive efficiency with and without him, but when comparing players...usually stuff like that doesn't come up.

So you have to be careful to try to define a player overall just by stats. But using a stat like ft% or game winning shot % to debate who is better at those specific things is fine.

what stats are you looking at? the stats this year say kobe is the best performer in the clutch in close games.

[josh99]

Well, by watching Steph Curry go off, you can get an assessment for his psyche... and know that he would likely have games where he shot like shit. and maybe even say that in games he shoots like shit, he would play more poorly overall.

I always look at box scores and try to imagine what "really happened." It's just sad that people believe it's the Bible truth or that its irrefutable when everything happens because of something. It's a chain of events that leads to the end result. Basketball skill, like anything else, is intangible. That's why people talk about empty stats.

-Smak

Wait so making lots of assumptions is more accurate than stats? I'm not saying that stats are Biblical truth but they are extremely useful. You cannot watch every single player every single game, if you watch every single game of one team how are you going to compare players from that team to players from teams that you hardly ever watch?

[chips93]

what stats are you looking at? the stats this year say kobe is the best performer in the clutch in close games.

i dont want to get off topic and turn this into a 'kobe' thread, or a 'clutch' thread, but i think when you look at his career stats in the clutch, they arent very good.