Scoring Assist Differential - Breaking Down The Kobe Laker E

[Malinhion]

No way man.

People have been arguing for years about whether or not Kobe shoots his teams out of games. Broadcasting companies notoriously present statistics on how the Lakers fare when Kobe takes x shots or less. This has happened for years.

The argument always then becomes whether or not he's shooting more to keep them in games that they're losing, or whether he's shooting them out of games. Really, you would only be able to tell by watching every contest. Its probably a little of both.

[TrueLAfan]

indiefan is taking down 1 kobe homer at a time.

TrueLAFan has him on lockdown.

This is not my intention. Indiefan is welcome to his opinions. He does not, however, have any statistical justification for what he is saying. From a functional and variance standpoint, there are too many mitigating factors about how players score and pass differently within various offensive systems. From an analysis standpoint, he has far too small of sample population(s) to derive meaningful conclusions. I'm not disagreeing with the ideas or conclusion at all...but I know, for a fact, that there is no "stat that analyzes assist production with respect to scoring" in a comparative sense, and what indiefan23 has done is not anything close to statistical analysis and cannot be taken as such.

[Malinhion]

indiefan is taking down 1 kobe homer at a time.

TrueLAFan has him on lockdown.

This is not my intention. Indiefan is welcome to his opinions. He does not, however, have any statistical justification for what he is saying. From a functional and variance standpoint, there are too many mitigating factors about how players score and pass differently within various offensive systems. From an analysis standpoint, he has far too small of sample population(s) to derive meaningful conclusions. I'm not disagreeing with the ideas or conclusion at all...but I know, for a fact, that there is no "stat that analyzes assist production with respect to scoring" in a comparative sense, and what indiefan23 has done is not anything close to statistical analysis and cannot be taken as such.

Can you explain what part of his analysis is affected by the statistical variance?

[WesWesley]

Can you explain what part of his analysis is affected by the statistical variance?

I think he's referring to the sample size having too much of a margin for error for it to be a reliable stat.

[TrueLAfan]

^^^

Think of it this way. Let's say you interviewed/polled people for a survey, and your confidence interval was 16% to have a 95% confidence level. From a statistical point of view, nothing you derive from your poll will be usable. The margin for error is too great. You may be thinking of it in Vegas-type way...as in, "Well, if I'm right 84% of the time, then that's pretty damn good!" From an analytic point of view, that's just not true. It doesn't work like that.

Here is a quote from a site that has a nice calculator for this sort of thing.

The confidence interval is the plus-or-minus figure usually reported in newspaper or television opinion poll results. For example, if you use a confidence interval of 4 and 47% percent of your sample picks an answer you can be "sure" that if you had asked the question of the entire relevant population between 43% (47-4) and 51% (47+4) would have picked that answer.

The confidence level tells you how sure you can be. It is expressed as a percentage and represents how often the true percentage of the population who would pick an answer lies within the confidence interval. The 95% confidence level means you can be 95% certain; the 99% confidence level means you can be 99% certain. Most researchers use the 95% confidence level.

http://www.surveysystem.com/sscalc.htm

I'm going to kind of cap the math talk right now. If people want to know more, or better acquaint themselves with this type of analysis, I suggest some advanced studies in statistics, particularly variance and standard deviation. There are plenty of websites, professors, statisticians, etc. that can help explain this for you. But it is not my job--or the job of this website--to teach sampling, variance, correlation, and other areas of advanced mathematics. I encourage people to go brush up and experiment with statistical analysis...but hope they will recognize the necessity of a firm grounding before coming up with a "stat" or "formula" that they claim shows and proves something. It's (a lot) harder than it sounds.

[Tim Lehrbach]

Accuracy Rating: 6

LOL, what?

[kflo]

even if the margin of error was extremely small, his analysis would still tell us very little about kobe. it's a junk made up stat that says nothing about player performance. did kobe play well in those games? poorly? hurt his team? it says nothing.

[Heyvoon24]

some people just try too hard

[Malinhion]

You're not understanding what I'm asking. I'll just drop it.

I know what a confidence intreval is.

[microfib4thewin]

Lakers were 17-3 when Kobe shot under 40%, Cavs were 3-7 when Lebron shot under 40% is obvious Kobe doesn't really impact his team. See what I did there?

Lakers are 3-5 when Kobe shoots 30+ shots, Cavs are 3-0 when Lebron shots 30+ shots. There are many more stats that support Kobe's ball hogging ways.

Yes, and it's because Kobe was able to elevate his teammates even though he shot badly by being a threat anywhere on the floor and making his teammates better. Lebron's teammates shot badly because Lebron cannot elevate his team's play. If Lebron can develop a jumpshot and have his teammates do the same then they can shoot better.

Of course I don't believe this, but that's indiefan23's line of logic when he argued that Kobe doesn't make his team better but Lebron does. All I did was switch the bias between those two players.

Accuracy Rating: 6

LOL, what?

Maybe it's on a scale of 7.

[Malinhion]

The only accuracy rating in the NBA is FG%.

[indiefan23]

This is not my intention. Indiefan is welcome to his opinions. He does not, however, have any statistical justification for what he is saying. From a functional and variance standpoint, there are too many mitigating factors about how players score and pass differently within various offensive systems. From an analysis standpoint, he has far too small of sample population(s) to derive meaningful conclusions. I'm not disagreeing with the ideas or conclusion at all...but I know, for a fact, that there is no "stat that analyzes assist production with respect to scoring" in a comparative sense, and what indiefan23 has done is not anything close to statistical analysis and cannot be taken as such.

You know what, I do have a totally founded method here and it took me a bit to realize what you did with your sample size argument. Been a few years since I did the old stats courses. Gotta admit I was scratching my head. Something didn't seem right and it took me a bit to figure it out.

You said that for 'Nique my sample size was only 45 games out of thousands. But you could not be more wrong. My entire data set is the 45 games. The sample sizes are actually the 5 point scoring quantiles vastly increasing the interval confidence. Vastly. I'm in no way making a stat that is predictive of 1000 games and I never once claimed it offered any insights. The stat is not saying that in 1000 games Kobe will have 8 games over 60 points with 4 dimes. Not one single thing like that.

Its reflective of games over 40 points and the behavior that actually occurred in those games and the trends that occured in dimes over those 5 point intervals. The >40 point interval is inclusive of the entire data set.

The interval confidence weakens as you approach the last interval, >55 points for some players because sample sizes do become an issue (which I duely noted in the column), but its still nothing near what you're saying, that the stat is meaningless and moot. I'm guessing you mistakenly misunderstood what my stat was trying to show because you don't seem to be the type that would intentionally mis-represent someone's hard work. Hope not anyway.

[indiefan23]

Accuracy Rating: 6

LOL, what?

I call it the Kobe Bryant argument corollary. In any given Kobe argument the accuracy rating of your statements varies proportionally to the number of personal insults you receive in response instead of people addressing the argument itself. Last I checked, I was at 6. I'm sure its higher now.

[indiefan23]

Yes, and it's because Kobe was able to elevate his teammates even though he shot badly by being a threat anywhere on the floor and making his teammates better. Lebron's teammates shot badly because Lebron cannot elevate his team's play. If Lebron can develop a jumpshot and have his teammates do the same then they can shoot better.

Of course I don't believe this, but that's indiefan23's line of logic when he argued that Kobe doesn't make his team better but Lebron does. All I did was switch the bias between those two players.

Hmm... yea, that's not the argument I made at all. Lebron has a history of taking mediocre players/teams way, way past what they should be doing. Kobe has never carried a team beyond it's expectations. The inverse is true however, and he has led them to underachievement. Gratz to the Lakers, and gratz to Kobe, but he's on the most talented team in the league and they should have won the championship. Before the playoffs even started they were clear favorites. What they should 'not' have done was be taken to 7 games by a 25 million dollar payroll after 3 games. They got bailed out, big time. They also should not have looked lost for half a series in Denver either.

To Kobe's credit, he turned it around and stopped chucking outside shots, started driving the lane, and started passing the rock to create easy buckets for his teammates. Prior to the last two games though, Kobe would spent entire quarters without even trying to score in the paint and just shot his team out of games after over dribbleing on the perimeter. Its the truth guys, not hate. I'm honestly happy for Kobe. He looked so relieved to get Shaq's a$$taste out of his mouth I felt good for him.

[kflo]

did kobe play well in the 96 games your "stat" is based on? and how does your "stat" answer that question?

[wallflower]

To be honest its not that stat you used that bugged me, but the conclusions you drew from them. Your stat could be accurate for all I know (am terrible at understanding stats) and within normal confidence interval but that doesnt mean it says half the stuff you claim it does. All it says is that when Kobe scores a lot there is a tendency for his box score to have less assists. That has nothing to do with his character (selfish? lol) or what happened during the game, or his motivations for making the plays he does. If you want a scientific approach with mathematical conclusions, then fine make them dont confound your findings with anything psychological because it clearly doesnt show any of that.

[Volcano]

I agree that Kobe's overrated..but you don't need stats to tell you that. Just compare observation to the hype. If you're ignoring his flaws or you're unable to tell what makes a great b-ball player, then you can't form a proper opinion. He's still a top 20 player of all time, but just overrated a bit by a whole lot of people.

[Asianiac_24]

Its funny how every hater try to discredit Kobe's scoring records. "BUT OMGZ ITS THE RAPTORS!" Yeah because Lebron never play them? Kobe plays the Raps TWICE a year, Lebron plays them FOUR TIMES! No one was discrediting Lebron because he had his triple doubles against the freakin' Knicks. Lebron, Wade, Arenas, MJ, etc etc, they all get to play against bottom feeders too every year. As for Kobe "stat padding", did you know the Lakers were down 15 at halftime before Kobe went off for 55 in the second half against the Raptors?

If Kobe wanted to stat pad, his scoring record would be higher than 81. Kobe could of easily scored over 81 against Dallas when he dropped 62 in 3 quarters. Or 56 in 3 quarters against the Grizzlies, etc. The article mentioned MJ, saying MJ would of been close to Wilt's record had he "stat-padded" like Kobe because he dropped 69 on the Celtics. That is just wrong. He failed to mention that game went into overtime, and MJ actually lost that game. Selective memory here.

The stat of "When Kobe shoots more, his team record is bad" is crap. If his teammates are playing well, Kobe feeds them the ball, thus shooting less. But when his teammates are bricking everything, this is when Kobe starts shooting. I don't know why this is so hard to understand.

Lastly, the biggest myth of all Kobe hate posts, "Kobe does not make his teammates better". Funny because almost EVERYONE who played with Kobe avged career high in FG% in LA. Smush, Cook, Odom, Gasol, Turiaf, you name it. While players around LeBron, the supposedly unselfish player, got worse statistically in Cleveland. Larry Hughes, Ben Wallace, Ilgauskus, etc, all became worse since arrival to Cleveland. Cleveland fans might bring up the age factor, and thats totally legit. However, that didn't stop Fisher from posting up career high FG% last year as a 32 year old.

[Asianiac_24]

This article should shut the kobe fans up. Kobe is a stat padder, not until this year did he kind of understand how to win.

And Lebron understands how to win? Last time I checked he has as much rings as you and me in his 7 year career

[Asianiac_24]

People will always find some fault in Kobe no matter what he does. He just won a freakin' ring, give him a break and learn to appreciate