How much worse is Nelson than CP3?

[Secretly Space Jesus 670]

quote:

Originally posted by iicybershotii:

 

Finally some stats. First of all, forget +/- and therefore also Lenovo and Roland. +/- is nearly useless in basketball. Every member of bad teams is going to be dead last in each one of those statistics. +/- is better for comparing within a single team. So then PER doesn't account for the time a player has the ball in his hands. Win shares... are a complete joke and are basically proportional to total wins. You can't compare a team with more wins to another team with less wins and expect that statistic to mean anything. The maximum number of win shares Curry could have is 26, where as rose can have 41. Why not look at win share %? And which efficiency stat are you talking about?

 

On the other hand, the individual statistics all favor curry. Using Hoopdata (and/or basketball-reference, EFF excluded), for advanced offense, Curry has a higher:

 

TS%, %AST, AR, DRR, TRR, EFF, WS and AWS.

 

So that leaves Rose with better ORR, TOR and the PER's.

 

So he's better at offensive rebounding only, turnovers, and the amazing PER... sounds good.

 

http://www.hoopdata.com/player...name=Stephen%20Curry

http://www.hoopdata.com/player...?name=Derrick%20Rose

 

Oh yea, and http://basketballmonster.com/Help.aspx

 

Well, you just demonstrated that you have no idea what adjusted +/- or the Roland rating are, since you seemed to just assume they're variations on the Lenovo stat that don't normalize for team wins(even though, they do). Given that both of those stats are measurements of how an individual team does with a player on the court versus how the team does with that player off the court, it's, you know, relevant.

 

Also, if you're going to say that PER is a joke, but you're then going to list Win Score, or any variation thereof, as being a meaningful statistic, you should probably just go ahead and kill yourself. Or keep pushing it in your case, since Win Score is exactly the kind of stat I'd expect from someone pushing a fantasy statistical site as actual player analysis, since it's literally just a sum of box score stats with virtually no multipliers at all(which is why it so blatantly favors centers and power forwards), and is exactly the kind of stat that would inflate the numbers of, for instance, a point guard like Curry who played on a team lacking even a single competent rebounder.

 

Also, the fact that you've STILL not acknowledged that Curry's numbers are inflated because he was playing on a team full of D-leaguers is just sad.

 

Tracy McGrady was putting up some of the best advanced stats EVER in '03. Which do you think is more likely: that Tracy McGrady had a top 5 season of all time in '03, or that his numbers were inflated because he played on a team that played Jeryl Sasser 1000+ minutes?

 

Also, I didn't include Win share % because the only way to do so fairly, given that Golden State played about 71 different players last season, would be to use win shares per48, and it seemed cruel to do so when Curry finishes 14th on his own ******* team when doing that(he finished second on his team overall, behind Maggette).

 

Also, real quick: TS% proves that Curry shoots 3s better than Rose which we already knew, Ast% is easily explained away by the fact that one played in a spread offense and one didn't(and the difference was minimal anyway, since player's Ast% can swing upwards of 10% from year to year depending on the offense they play in), you can't really take rebound rate seriously when one played without a competent rebounder and the other played with several(and again, the difference was minimal anyway), I already explained hoopdata's EFF stat, and I already explained win score.

 

Oh, and by the way you arrogant little troll: I saw the basketballmonster help page. I also saw this on the first page you linked to:

 

"Player values are calculated using our own valuation method based on objective statistics and experienced subjective input."

 

I know how to set a standard score to zero. I'm not a moron, nor am I someone who dropped out of high school. However, given that their valuation method is "based on objective statistics and experienced subjective input", but they never bother explaining anywhere on their site how those statistics are being calculated together nor are they explaining what subjective input they're adding to it, there's no way to take a stat like that seriously, and only a total moron actually would.

 

You know why we know about all the flaws that PER has as a statistic, such as deflating point guards and over-pumping players who are volume shooters, especially volume 3pt shooters?

 

Because Hollinger wrote a ****ing book about how he was calculating it, and put it out in the aether for criticism. That's what statisticians DO.

 

That's why Hollinger can call himself a statistician with a straight face, and why I'd never heard of basketballmonster until you came here to shill for them.

 

I can't believe I'm having to actually explain why using a fantasy basketball stat metric, a metric which says that Dwight Howard is like the 35th best player in the NBA, is retarded to someone. Wow.