Do you notice the silence?

[SexSymbol]

Of this forum. Curry had a great performance, 26 in one Q, and nobody's talking, because it just solidifies him more as a runaway MVP.
Top 3 performance of the season by the most popular player in the league and 1-2 threads on the front page, LeBron/Harden stans shook

[navy]

Or because it was a late ass game nobody stayed up for.

[Cocaine80s]

Maybe its because no one gives a shit

[3ball]

Of this forum. Curry had a great performance, 26 in one Q, and nobody's talking, because it just solidifies him more as a runaway MVP.
Top 3 performance of the season by the most popular player in the league and 1-2 threads on the front page, LeBron/Harden stans shook
of course bro, anything that might diminish losers like lebron and harden or whoever, and everyone becomes ghost itt or forum.

no one cares about the way the game actually is, they only care about what allows them to continue believing their pre-conceived notion of the game.

what's that GIF from the wire? yeah, that one.

[RRR3]

Lmao @ Curry being the "most popular player". IDGAF how many AS votes he got. LeBron and Kobe (in no order either one could be argued as first) are still the most popular players. Asinine to think otherwise IMO.

[RoundMoundOfReb]

So you bitch about nobody talking about it and then call him the most popular player? Which one is it?

[J Shuttlesworth]

What's scary is that's not even the best performance put up by a Warrior this year. Klay put up 52 in about 3 quarters. Just shows how stacked that team is and why Harden = MVP

[JT123]

So you bitch about nobody talking about it and then call him the most popular player? Which one is it?
:lol. I noticed that too. Kobe stans stay taking L's

[aj1987]

So you bitch about nobody talking about it and then call him the most popular player? Which one is it?
OP is dumber than a box of rocks. Why does it surprise you?

I honestly can't believe than Curry is gonna win MVP. I actually expected KD to repeat (before his injury) or AD to win it. But Curry? :coleman:

[buddha]

Curry is not even that good. Hassan Whiteside has a higher per.

[RoundMoundOfReb]

Curry IS that good. He leads the league in real-plus minus by a fair margin...In fact it's higher than Durant's was last year.

Fact is Curry has been a top 5 player in the league for 3 years now. People are just realizing it.

[Inferno]

Curry IS that good. He leads the league in real-plus minus by a fair margin...In fact it's higher than Durant's was last year.

Fact is Curry has been a top 5 player in the league for 3 years now. People are just realizing it.

:rockon:

[Practice?]

OP is dumber than a box of rocks. Why does it surprise you?

I honestly can't believe than Curry is gonna win MVP. I actually expected KD to repeat (before his injury) or AD to win it. But Curry? :coleman:

AD was never a solid bet unless you actually believed Pelicans would be near the top of the west.

[Dresta]

Curry IS that good. He leads the league in real-plus minus by a fair margin...In fact it's higher than Durant's was last year.

Fact is Curry has been a top 5 player in the league for 3 years now. People are just realizing it.
Yep. Don't know if other people are just hating because they're stans, or whether they are actually that blind.

[DukeDelonte13]

Lebron, Harden, and AD are all more valuable to their teams than Curry is but Curry is somewhat of an NBA darling child, and Lebron already won a bunch. AD hasn't been in the league long enough.

[3ball]

Lebron, Harden, and AD are all more valuable to their teams than Curry is but Curry is somewhat of an NBA darling child, and Lebron already won a bunch. AD hasn't been in the league long enough.
b-b-b-b-but curry leads the league in RAPM

http://espn.go.com/nba/statistics/rpm/_/sort/RPM


that means he's better than lebron

[Dro]

Or because it was a late ass game nobody stayed up for.
This, I fell asleep on it.

[FKAri]

Most popular player in the league?

I doubt more than 1 in 50 people in this country have even heard his name before. He isn't on ESPN's agenda like Bron is.

[Dro]

:lol. I noticed that too. Kobe stans stay taking L's
All stans stay taking L's, just for being stans of another grown man:confusedshrug:

[insidious301]

b-b-b-b-but curry leads the league in RAPM

http://espn.go.com/nba/statistics/rpm/_/sort/RPM


that means he's better than lebron

That is RPM, not RAPM.

You can find this seasons RAPM here:
http://stats-for-the-nba.appspot.com/ratings/RAPM_2015.html

[3ball]

That is RPM, not RAPM.

You can find this seasons RAPM here:
http://stats-for-the-nba.appspot.com/ratings/RAPM_2015.html
whats the difference

[3ball]

whats the difference
oh wait, here's the difference:

RPM reflects enhancements to RAPM by Engelmann, among them the use of Bayesian priors, aging curves, score of the game and extensive out-of-sample testing to improve RPM's predictive accuracy.

http://regressing.deadspin.com/just-what-the-hell-is-real-plus-minus-espns-new-nba-s-1560361469


so RPM is an improvement to the out-dated RAPM technique.

[3ball]

oh wait, here's the difference:

RPM reflects enhancements to RAPM by Engelmann, among them the use of Bayesian priors, aging curves, score of the game and extensive out-of-sample testing to improve RPM's predictive accuracy.

http://regressing.deadspin.com/just-what-the-hell-is-real-plus-minus-espns-new-nba-s-1560361469


so RPM is an improvement to the out-dated RAPM technique.
gonna be honest - RPM looks a lot better to me.

if i was going to use RAPM to determine who the best player was, RAPM would say Kawhi Leonard is #2 player in the league, Iguodala is #5, Draymond Green and Ricky Rubio are #7 and #8 in the entire league, Jrue Holiday > Westbrook, Mike Conley > Wall, and on and on and on.... like literally many, many more weird things.

RAPM has far more holes, so many that it's untenable, which why RPM is such an improvement... The RPM data doesn't have these kinds of gross mis-rankings, or at least not nearly as many.

[insidious301]

whats the difference
RPM contains box score elements, while RAPM is based only on scoring margin, and quality of teammates and opponents.

RPM is basically an advanced version of +/- (good stat, I use it, but personally prefer RAPM).

[3ball]

RPM contains box score elements, while RAPM is based only on scoring margin, and quality of teammates and opponents.

RPM is basically an advanced version of +/-
nothing against your opinion, but the experts say RAPM is an outdated technique, and that RPM improves on RAPM with the use of Bayesian priors, aging curves, score of the game and extensive out-of-sample testing.

the stat bots at ESPN who calculate RPM get paid a lot more than the guys who came up with RAPM for their "science fair project" as it has been called... all the RAPM websites look like science fair projects done in their mother's basement... its true... for my money, i'll go with ESPN and their guys.
.

[$LakerGold]

Lmao @ Curry being the "most popular player". IDGAF how many AS votes he got. LeBron and Kobe (in no order either one could be argued as first) are still the most popular players. Asinine to think otherwise IMO.
You go to China, only 45% of the people you ask knows Curry. You ask about Kobe or Lebron, 95% of em will know how they are. I guarantee you that.

[insidious301]

nothing against your opinion, but the experts say RAPM is an outdated technique, and that RPM improves on RAPM with the use of Bayesian priors, aging curves, score of the game and extensive out-of-sample testing.

the stat bots at ESPN who calculate RPM get paid a lot more than the guys who came up with RAPM for their "science fair project" as it has been called... all the RAPM websites look like science fair projects done in their mother's basement... its true... for my money, i'll go with ESPN and their guys.
.

J.E., the "expert" on ESPN who created RPM, also hosts the RAPM website I linked you to. He dabbles with both, and even with regressed versions of RAPM (xRAPM/NPI-RAPM).

RPM/RAPM are both great stats, but like all stats, you need to use them with caution and context.

[HOoopCityJones]

Curry will be burdened with basically the same problem Kobe was post Shaq , most of his games will be Televised late night in comparison to the East Coast and half the country will have been asleep during his standout games.

Then you'll have a bunch of know it alls , who were probably knocked out around the time a Klay Thompson dropped 37 in a quarter, coming out the woodwork to claim a one of a kind performance as overrated just because they went ****in night night while it was happening. "But did you see Kyrie or John Wall tho?"

It's not just the Cali Teams either, the Texas teams suffer from the same shit. I can guarantee you half of ISH was treating Dirk's name like second hand talent before the 2011 post season where he went off. Duncan in general doesn't get the respect he deserves, but it's common knowledge what playing in San Antonio did to his career as far as flying under the radar.

[Akhenaten]

gonna be honest - RPM looks a lot better to me.

if i was going to use RAPM to determine who the best player was, RAPM would say Kawhi Leonard is #2 player in the league, Iguodala is #5, Draymond Green and Ricky Rubio are #7 and #8 in the entire league, Jrue Holiday > Westbrook, Mike Conley > Wall, and on and on and on.... like literally many, many more weird things.

RAPM has far more holes, so many that it's untenable, which why RPM is such an improvement... The RPM data doesn't have these kinds of gross mis-rankings, or at least not nearly as many.

you ninjas with these goofy azz "advanced" stats man :facepalm

none of you gooftroop azz ninjas even now what the hell a Bayesian prior is.

so stupid

[3ball]

J.E., the "expert" on ESPN who created RPM, also hosts the RAPM website I linked you to. He dabbles with both, and even with regressed versions of RAPM (xRAPM/NPI-RAPM).

RPM/RAPM are both great stats, but like all stats, you need to use them with caution and context.
I know it's the same guy, but that same guy uses a garbage, homemade looking site for his RAPM data, but posts his RPM data on ESPN.

as for the context - that's just people's subjective opinion.

seriously, like which data should someone use? RPM or RAPM? They both yield wildly different results. I'd go with the stuff that the creator of both was willing to post on ESPN (assuming i was the type of fan that needed stats to inform me and give me an opinion about the game).

[Akhenaten]

nothing against your opinion, but the experts say RAPM is an outdated technique, and that RPM improves on RAPM with the use of Bayesian priors, aging curves, score of the game and extensive out-of-sample testing.

.

Can you explain what a Bayesian prior is please?

[gts]

you ninjas with these goofy azz "advanced" stats man :facepalm

none of you gooftroop azz ninjas even now what the hell a Bayesian prior is.

so stupid

3balls just copy and pasting from a deadspin article from last year...

http://regressing.deadspin.com/just-what-the-hell-is-real-plus-minus-espns-new-nba-s-1560361469

ESPNs stat has already been proven to be pretty much worthless

[insidious301]

I know it's the same guy, but that same guy uses a garbage, homemade looking site for his RAPM data, but posts his RPM data on ESPN.

as for the context - that's just people's subjective opinion.

seriously, like which data should someone use? RPM or RAPM? They both yield wildly different results. I'd go with the stuff that the creator of both was willing to post on ESPN (assuming i was the type of fan that needed stats to inform me and give me an opinion about the game).

I use a number of advanced metrics, personally. Not just "one". I think we can both agree that watching film and game tape (aka the "eye test") beats anything metric related with even the smallest margin of error.

Watching as many games possible will yield THE best results.

[3ball]

I use a number of advanced metrics, personally. Not just "one". I think we can both agree that watching film and game tape (aka the "eye test") beats anything metric related with even the smallest margin of error.

Watching as many games possible will yield THE best results.
the eyetest will always be best, heck even for guys that don't know much - even they can see something that the stats can't tell you, and it will probably be much more acutely specific and relevant to whatever circumstance is being contemplated.

personally, i've only looked at such data a few times, and only to reconcile two pieces of the data - i.e. love's RPM was very high last year, but this year it has cratered.. i would try to come up with some theories as to why... that's one of the drawbacks of RPM - one year's RPM does not predict a future year's RPM, especially if playing with different teammates.

but nonetheless, i'd still prefer to look at data that the creator thinks is his latest, best work and a finished product and that he posted on ESPN (which would be RPM), rather than the pre-cursor, 1st and 2nd draft-type stuff he was messing around with and put on his blog or homemade site (RAPM).

[Hey Yo]

I know it's the same guy, but that same guy uses a garbage, homemade looking site for his RAPM data, but posts his RPM data on ESPN.

as for the context - that's just people's subjective opinion.

seriously, like which data should someone use? RPM or RAPM? They both yield wildly different results. I'd go with the stuff that the creator of both was willing to post on ESPN (assuming i was the type of fan that needed stats to inform me and give me an opinion about the game).
Use the safest bet, LBJ.

[3ball]

Use the safest bet, LBJ.


it's duncan... by a mile

[Chadwin]

only 33 mpg and his team is stacked

Harden for MVP

[Jasper]

Of this forum. Curry had a great performance, 26 in one Q, and nobody's talking, because it just solidifies him more as a runaway MVP.
Top 3 performance of the season by the most popular player in the league and 1-2 threads on the front page, LeBron/Harden stans shook
run away if he keeps his numbers up -
point on : a player winning the MVP has to be versatile as well.
Even when Rose one it , he had some assists per game ...
Durant won it , shooting shots as well as 3 pt attempts as a 7' er.

[IMObjective]

Can you explain what a Bayesian prior is please?
In Bayesian statistical inference, a prior probability distribution, often called simply the prior, of an uncertain quantity p is the probability distribution that would express one's uncertainty about p before some evidence is taken into account. Bayes' theorem relates current probability to prior probability. It is important in the mathematical manipulation of conditional probabilities.

I hope that clears it up for you :oldlol: