RealGM

Every where I look, people are arguing that one player is better then another because he has a better TS% or his PER is higher. Well, if thats not the case. You have to actually watch the players and form an opinion. Stats are a smokescreen for people who don't know basketball to make it seem like they do. Straight forward stats like PPG, RPG, etc. are good because they tell you the direct effect that the players have on a game, but you can't just judge players on that.

Don't cry because you don't understand advanced statistics. They can be very helpful for someone who knows how they are measured and uses them to assert observations that they gather from watching the guys play.

Stats are right 60% of the time, everytime.

tracey_nice wrote:Stats are right 60% of the time, everytime.

You just hate stats because McGrady hasn't had any worth mentioning since 2004.

Malinhion wrote:

tracey_nice wrote:Stats are right 60% of the time, everytime.

You just hate stats because McGrady hasn't had any worth mentioning since 2004.

lol, dude, brah, I am not a Mcgrady fan....soo, is this awkward now?

we use stats because we all have lives outside the game and can't watch all 600 games or so a season in depth.

^ don't speak for Malinhion

It depends on what stats and how they are used. TS% is an useful stat. PER can be useful in the rare occasions when its actually used correctly. Then there are stats like opponent PER (or any stat that attempts to calculate defense) and such which are pretty much worthless.

tracey_nice wrote:

Malinhion wrote:

tracey_nice wrote:Stats are right 60% of the time, everytime.

You just hate stats because McGrady hasn't had any worth mentioning since 2004.

lol, dude, brah, I am not a Mcgrady fan....soo, is this awkward now?

More awkward than meeting a transvestite you hooked up with the night before, and now the balls are hanging free in sweatpants and she's got a 5 o'clock shadow.

Malinhion wrote:Don't cry because you don't understand advanced statistics. They can be very helpful for someone who knows how they are measured and uses them to assert observations that they gather from watching the guys play.

Ok, I agree with this. I'm more talking about people that use stats as their only way of arguing something. Instead of actually making legit points, they go to 82games.com or whatever. John Hollinger types who only look at stats and not the game.

Yes. I agree that stats need to be used as supplementary information and not as primary sources for inference. But sometimes this is the best info we can get without seeing a lot of a player, and it is possible to get insights from advanced stats that you won't get from looking at box scores.

durka wrote:

Malinhion wrote:Don't cry because you don't understand advanced statistics. They can be very helpful for someone who knows how they are measured and uses them to assert observations that they gather from watching the guys play.

Ok, I agree with this. I'm more talking about people that use stats as their only way of arguing something. Instead of actually making legit points, they go to 82games.com or whatever. John Hollinger types who only look at stats and not the game.

I'm pretty sure most people use stats and analysis. If you think a player's D is overated, you go to 82games and find out that his team actually defends better without him, you can come back and say, "he could be a great defender if he put more effort into it but he has been coasting on reputation for the past 3 or 4 years, his team actually defends a little better when he is a spectator."

Capiche?

I trust stats that have been compiled with every single game the player's played without bias than some guy on the internet who claims to have seen his favorite player all season or have watched some college player in the tournament who went 5/7 from 3!! ZOMG!

So basically what the OP is trying to say is "base your opinion on a player not on their stats but on their youtube mix."

There are some stats I don't like to cite like PER and what not, but I'm now a firm believer in eFG% / TS%...

Pace is another thing that often goes unnoticed so pace-adjusted stats are always useful...

The Main Event wrote:So basically what the OP is trying to say is "base your opinion on a player not on their stats but on their youtube mix."

No, I'm actually saying base your opinion on the players play not his stats. If i haven't seen a player play then I don't pretend like I know what I'm talking about. Starts are a good way to support your opinion but not form it.

People don't like stats because 1) They don't understand it and 2) It doesn't agree with what they think. I have seen VERY few people who actually understand what PER is and what context to use it in. Instead most go, IT SAYS MY FAVORITE PLAYER IS NOT AS GOOD AS THIS OTHER PLAYER SO I HATE IT!

I think the opposite. People like to use stats to make their favorite players look good and disregard things like hustle and intangibles and the little things that make players that much more valuble. If stats told us how good players were, Zach Randolph would be one of the best bigs in the leauge for the past few years.

Any stat that is weighted is most likely garbage. Like PER. Weights are arbitrary and at the full discretion of the maker of the stat. Weights are only useful if they used in some type of historical significance. Some people take stats too far like Hollinger, who tried to say Kevin Love is a better prospect than OJ Mayo because of their collegiate stats (he obviously ignores lost of other factors, but I digress)

Also individual defensive rating are very flawed.

The only stats I trust are team offensive and defensive ratings, individual offensive rating, usage, and pace. Percentages of rebounding, assists, and steals seem to make sense too.

You should read Dean Oliver's book "Basketball on Paper".

obinna wrote:Any stat that is weighted is most likely garbage. Like PER. Weights are arbitrary and at the full discretion of the maker of the stat. Weights are only useful if they used in some type of historical significance. Some people take stats too far like Hollinger, who tried to say Kevin Love is a better prospect than OJ Mayo because of their collegiate stats (he obviously ignores lost of other factors, but I digress)

Also individual defensive rating are very flawed.

The only stats I trust are team offensive and defensive ratings, individual offensive rating, usage, and pace. Percentages of rebounding, assists, and steals seem to make sense too.

You should read Dean Oliver's book "Basketball on Paper".

But it's so much better than not weighing though... otherwise you end up with garbage like EFF and Assist to TO ratio.