Lies, damn lies and statistics (split)

HighFlyingMonkey said:
Incredibly, somebody was able to calculate Real +/- stats from Summer League action. Moreland ranked third in the entire tournament. Small sample size of course, but underscores how well he played and what kind of impact he made when he was on the court.

http://nyloncalculus.com/2014/07/22/2014-las-vegas-summer-league-mvps/

View attachment 4922
Click to expand...

That would mean so much more to me if "real +/-" weren't basically the most worthless stat yet invented by the stat brigade. What a scam.

Fortunately I have this handy test I like to employ where I actually watch guys play, and amazingly enough, actually watching Moreland play you could see he was dynamic at that level. Wondering what the games are about with him and this signing. Not like somebody is going to call a press conference for the signing of an undrafted guy.
 
Bricklayer said:
That would mean so much more to me if "real +/-" weren't basically the most worthless stat yet invented by the stat brigade. What a scam.

Fortunately I have this handy test I like to employ where I actually watch guys play, and amazingly enough, actually watching Moreland play you could see he was dynamic at that level. Wondering what the games are about with him and this signing. Not like somebody is going to call a press conference for the signing of an undrafted guy.
Click to expand...

Ick, don't bash something you don't understand. RAPM has it's flaws certainly, but it's certainly a useful statistic. In this case, not so much because it's 100 minutes, but on a larger scale, it's a good one.
 
HighFlyingMonkey said:
Incredibly, somebody was able to calculate Real +/- stats from Summer League action. Moreland ranked third in the entire tournament. Small sample size of course, but underscores how well he played and what kind of impact he made when he was on the court.

http://nyloncalculus.com/2014/07/22/2014-las-vegas-summer-league-mvps/

View attachment 4922
Click to expand...

Kinda silly someone took time to do so, but cool to see nevertheless. Backs up what we saw out of the impact players in SL
 
HighFlyingMonkey said:
No need to trash a stat you don't understand. Real +/- isn't some panacea but its useful. Though in this situation probably not much given the sample size.
Click to expand...

The_Jamal said:
Ick, don't bash something you don't understand. RAPM has it's flaws certainly, but it's certainly a useful statistic. In this case, not so much because it's 100 minutes, but on a larger scale, it's a good one.
Click to expand...

lol...
 
The_Jamal said:
Ick, don't bash something you don't understand. RAPM has it's flaws certainly, but it's certainly a useful statistic. In this case, not so much because it's 100 minutes, but on a larger scale, it's a good one.
Click to expand...

Oh yes, quite a useful stat. :lol:

Did you know Channing Frye is better than Kevin Love?

Or that Jeff Withey is better than DeMarcus Cousins? Heck so is Shavlik Randolph, Kyle O'Quinn, and Jason Thompson for that matter, so no surprise there. Although Cuz can be forgiven for not equaling the greatness that is JT, since JT is also > Anthony Davis.

You guys and your clinging to every geek's muddled formula just because it is a geek's formula is just goofy. I can only imagine its the same urge that makes people go in for invisible sky wizards looking for somebody to tell them what ultimate truth is. I however don't have that urge. Stat reflects observable reality over a large enough set, its a useful, although perhaps still not necessary stat. Stat produces junk lists unreflective of reality, then it does no produce a single ounce of pain for me to call it, and its adherents, out for being foolish in the extreme.


Here, and just to prove to you that stats geeks are not all sheep everytime a major website pushes a fad stat to drive up membership:
http://www.boxscoregeeks.com/articles/rpm-and-a-problem-with-advanced-stats
 
Bricklayer said:
Oh yes, quite a useful stat. :lol:

Did you know Channing Frye is better than Kevin Love?

Or that Jeff Withey is better than DeMarcus Cousins? Heck so is Shavlik Randolph, Kyle O'Quinn, and Jason Thompson for that matter, so no surprise there. Although Cuz can be forgiven for not equaling the greatness that is JT, since JT is also > Anthony Davis.

You guys and your clinging to every geek's muddled formula just because it is a geek's formula is just goofy. I can only imagine its the same urge that makes people go in for invisible sky wizards looking for somebody to tell them what ultimate truth is. I however don't have that urge. Stat reflects observable reality over a large enough set, its a useful, although perhaps still not necessary stat. Stat produces junk lists unreflective of reality, then it does no produce a single ounce of pain for me to call it, and its adherents, out for being foolish in the extreme.


Here, and just to prove to you that stats geeks are not all sheep everytime a major website pushes a fad stat to drive up membership:
http://www.boxscoregeeks.com/articles/rpm-and-a-problem-with-advanced-stats
Click to expand...

RPM and RAPM aren't the same thing. They're variations of each other. I don't like RPM because it skews it's rankings far more towards players on winnings teams, even though it's supposed to be "minute and team independent". The basic idea of the stat is to take a look at ranking the value of the role players, moreso than the star guys.

And ESPN has ALWAYS been about 3-4 years behind on statistics. Same thing happened with baseball with "WAR" and ESPN started spouting it off as gospel.
 
Of course, these stats are not very reliable, but I think it's obvious that we were a much better team when Moreland was on the floor!
 
Bricklayer said:
Oh yes, quite a useful stat. :lol:

Did you know Channing Frye is better than Kevin Love?

Or that Jeff Withey is better than DeMarcus Cousins? Heck so is Shavlik Randolph, Kyle O'Quinn, and Jason Thompson for that matter, so no surprise there. Although Cuz can be forgiven for not equaling the greatness that is JT, since JT is also > Anthony Davis.

You guys and your clinging to every geek's muddled formula just because it is a geek's formula is just goofy. I can only imagine its the same urge that makes people go in for invisible sky wizards looking for somebody to tell them what ultimate truth is. I however don't have that urge. Stat reflects observable reality over a large enough set, its a useful, although perhaps still not necessary stat. Stat produces junk lists unreflective of reality, then it does no produce a single ounce of pain for me to call it, and its adherents, out for being foolish in the extreme.

Here, and just to prove to you that stats geeks are not all sheep everytime a major website pushes a fad stat to drive up membership:
http://www.boxscoregeeks.com/articles/rpm-and-a-problem-with-advanced-stats
Click to expand...

You don't have to work so hard to prove you have no idea how it works and how to use it, we believe you just fine :p

Its amusing watching people (especially around these parts) rail on and on about these mystical "advanced stats"; its like watching somebody hammer away at a nail in futility with a screwdriver and complaining about how useless it is. Only response is to pat them on their senile little heads and move on. "If its such a smart phone why can't it change my car battery?!?"

Hey look, here's a stat that ranks JT above Cousins. Wow, it also ranks DeAndre Jordan above Lebron James and Kevin Durant! Also, Samuel Dalembert is ranked higher than the entire Western Conference All-Star team. What a garbage stat! Lets get rid of it! Wait, we're talking about FG% :p

Though, its true we don't know the exact formula for RAPM, which limits a rigorous evaluation of its merit. We do know what it measures in theory. It didn't just pop into existence by itself and create its own formula through a series of random events, although some people might tell you thats how the Universe got here :D
 
HighFlyingMonkey said:
You don't have to work so hard to prove you have no idea how it works and how to use it, we believe you just fine :p
Click to expand...

No no, do not confuse having no need to use it or desire to do so with no idea how to. RPM stats are noisy to the point of ridiculous. No scientist would ever rely on such a messy corrupted bundle of nonsense. But hey, basketball fans are supposed to. Because we're stupid. RAPM can't even reliably produce lists that its supposed to produce, let alone the ridiculous error of promoting it as some be all stat.

Advanced stats are often anything but. They are Stats for Dummies. Stat baby food. They are not real stats at all. They are preprocessed formulas made from real stats for basketball neophytes to suck down and spit back up at inappropriate times in big smelly blobs. And they pretty much require you to have a follower's mentality, since you are laying your trust entirely on the dipcrap who designed the thing in the first place to have truth you do not. Worse yet, you suck down all their biases that went into mushing the stats in just that way. Because I can mush stats very nicely to feature anything that I think is important too. So they come up, I look at the most recent dips stat baby food, and immediately I check to see if his cute little formula looks at all like reality. It does? Hey, he gets a cookie for representing statistically what his basketball betters already knew without his little B.S. formula. Flawed old battlehorses like PER and WS/48 do pretty well on that front. Blind and crude, but still at least can be useful for descriptive purposes. But now some little dip's formula spits out nonsense, then into the trash it goes.

And we are now entering a new era when we are streaming in are a huge swath of truly, if not advanced, at least real stats. And the latest stupid baby food formulas pale in comparison to the actual real stats flowing in from the cameras. Advanced shot charts, time of possession, hell behind the scenes apparently they even have dribbles per possession, whatever. Even these stats will continue to pale in the face of someone who knows their stuff -- a human brain can balance 1000 factors around the edges in an instant that a formula is dumb and deaf to. But the new stats give us extra data points. "Advanced metrics" do nothing of the kind. They don't provide new data. They regurgitate mushed up basic stats.

And to keep this argument from getting circular, I now drop the gauntlet:

Given that an almost infinite variety of statistical formulas can be derived by combining, recombining, balancing etc. etc. the dozen or so most common basic stats, if I spit out 100 such formulas right now, producing 100 different lists, how exactly would you go about determining which ones had value and which ones did not other than looking at the lists and comparing them to what you personally already knew about the league and its players? The answer is there is no other way. You either determine their validity based on what you know to be true, or you play dumb and are forced to accept them all as having intrinsic value. Including my times he picks his nose/0.25FTA * 317.1TO * stl/min formula. It too will produce a list of players. If you refuse to engage your brain and dismiss it out of hand, then refuse to engage your brain and dismiss it once you see the list it produces, then there you go. It too is an "advanced" formula. And hey, unlike the RAPM guys, I'm not even too humiliated at my crap to let you see the formula itself.
 
Last edited:
Bricklayer said:
Given that an almost infinite variety of statistical formulas can be derived by combining, recombining, balancing etc. etc. the dozen or so most common basic stats, if I spit out 100 such formulas right now, producing 100 different lists, how exactly would you go about determining which ones had value and which ones did not other than looking at the lists and comparing them to what you personally already knew about the league and its players? The answer is there is no other way. You either determine their validity based on what you know to be true, or you play dumb and are forced to accept them all as having intrinsic value. Including my times he picks his nose/0.25FTA * 317.1TO * stl/min formula. It too will produce a list of players. If you refuse to engage your brain and dismiss it out of hand, then refuse to engage your brain and dismiss it once you see the list it produces, then there you go. It too is an "advanced" formula. And hey, unlike the RAPM guys, I'm not even too humiliated at my crap to let you see the formula itself.
Click to expand...

The irony to me is that the stats that eventually prove to be the most valuable are those that agree with what my eye sees.
 
Glenn said:
The irony to me is that the stats that eventually prove to be the most valuable are those that agree with what my eye sees.
Click to expand...
But if you assign a number to what your eyes see, that makes it more valuable, right? ;)
 
Spike said:
But if you assign a number to what your eyes see, that makes it more valuable, right? ;)
Click to expand...

If the number actually represents something and is valuable, the number is easier to use in an argument than to argue all the underlying factors that your eyes see.

For example, Moreland is great because his number is the third best in the SL is easier to say than Moreland is great because he seems like a legitimate defensive player, he is probably one of the few players that are available who could play opposite Boogie because of yackity yack, he is this, he is that, and furthermore he is this and blah! blah! blah! Stats are shorthand.

brick's question is important. How do we know which stats to trust and which are more valuable than others? I do not know what stats are reliable as I am getting too damn old to care that much and simply want to watch the Kings play and occasionally educate you youngsters. :) I mean, heck, look at what bricklayer has learned after being around me for over 10 years.

I think we are fortunate to have forum members who are great with numbers.
 
Last edited:
Glenn said:
If the number actually represents something and is valuable, the number is easier to use in an argument than to argue all the underlying factors that your eyes see.

For example, Moreland is great because his number is the third best in the SL is easier to say than Moreland is great because he seems like a legitimate defensive player, he is this, he is that, and furthermore he is this and blah! blah! blah! Stats are shorthand.

brick's question is important. How do we know which stats to trust and which are more valuable than others? I do not know what stats are reliable as I am getting too damn old to care that much and simply want to watch the Kings play and occasionally educate you youngsters. :) I mean, heck, look at what bricklayer has learned after being around me for over 10 years.
Click to expand...

See, I don't think eyes and stats are independent of each other. They should be used to supplement each other. So like the Moreland example. We see with our eyes that he had a tremendous impact when he was in the game. And the most statistics that people care to go find (like the one above) back up that assertion. Then, we know our eyes aren't lying to us and we have a clearer picture of the player. The same could be true of the reverse. Say, I had no idea who Moreland was, but saw the above RAPM stat of his impact in the game. That'd make me want to watch him play and I'd be able to see that the numbers backed up what he was doing on the court.

Another example would be if you started watching Jeremy Lin when he went crazy in NYK for those 2 weeks, but didn't look at his numbers for the rest of that season, when he fell back down to earth. You'd have a very different perception of Jeremy Lin than who he actually was.
 
The_Jamal said:
See, I don't think eyes and stats are independent of each other. They should be used to supplement each other. So like the Moreland example. We see with our eyes that he had a tremendous impact when he was in the game. And the most statistics that people care to go find (like the one above) back up that assertion. Then, we know our eyes aren't lying to us and we have a clearer picture of the player. The same could be true of the reverse. Say, I had no idea who Moreland was, but saw the above RAPM stat of his impact in the game. That'd make me want to watch him play and I'd be able to see that the numbers backed up what he was doing on the court.

Another example would be if you started watching Jeremy Lin when he went crazy in NYK for those 2 weeks, but didn't look at his numbers for the rest of that season, when he fell back down to earth. You'd have a very different perception of Jeremy Lin than who he actually was.
Click to expand...

I'm guess I'm just old-fashioned. I'm going to believe my lying eyes over dry combinations of numbers every day of the week to determine my opinion of a particular player. "We see with our eyes that he (Moreland) had a tremendous impact when he was in the game." I'll agree in total with that statement. But, unlike you, I do not need "statistics that ... back up that assertion" to "know our eyes aren't lying to us."

Using statistics to determine whether or not to watch a particular player might be valid. Using statistics, however, instead of trusting your own impressions is just silly - unless you (not meaning you specifically) are doing so to mask the holes in your knowledge of basketball.

Just sayin'
 
Small sample size or not, in this instance Moreland's high RAPM does jibe with my eyes told me.

Inexpcably Oriakhi started (side note, despite being a only moderate fan of the LOTR movies and books every time I saw Oriakhi's name I thought of the Fighting Uruk-hai) but the momentum turned noticeably in most games when Moreland dubbed in.

And part of why I'll lend credence to the stat despite the small sample size is the consistency Moreland showed. He put up nearly the same rebounding and blocks per minute in each game. It wasn't like he feasted on weak teams and struggled with NBA caliber slashers or bigs.

What I saw was a defensive energizer that made the Kings' summer league team significantly better everytime he came into the game.

All stats are just pieces of info. The key is knowing how much weight to give them and understanding why they are high or low for a given player. And to me that means using the eye test.
 
The +/- stat item is a relative indicator of a players impact. Not absolute, not a factual reference point but a relative indication of what happened when that player was in the game. Good for one view of a players performance but for Moreland his 2.7 blk and 8.8 feb per game are the "real" stats for him. Since the Kings generally struggled in the first quarters in Vegas, the bench would tend to show better +/- numbers as a relative indicator of out scoring opponents . Then when PDA/Malones "positionless basketball" comes into play and the +/- gets a bit more vague but still a good talking point for summer league.
 
I guess like VF21, I'm old fashion. Doesn't mean I don't look at stats, but the one stat that I pay little or no attention to is the +/- stat. I could spend a long time explaining why, but since I think the stat is irrelevant, I'm not going to waste my time. I watch more basketball than any human being should watch, unless their getting paid to do it. And when I'm watching, I'm not looking at stats.. Let me see a player play in at least 10 games, and I can tell you whether he's any good or not, and what his real pluses and minuses are. If you took the time to watch Moreland play, you didn't need to look up his stats to confirm what you saw. During one of the games I turned to Uncia03 and said, man Moreland must have a lot of boards, he seems to be grabbing everything. At the time, I didn't know how many he had. Turned out he ended up with 12 boards and in 18 minutes of play. You know what, I don't need any stat breakdowns to tell me that's pretty damm impressive.

What I don't need is a stat that gives a minus rating to a player that comes into a game for 12 minutes, scores 12 points, grabs 5 boards and blocks a shot, but still gets a minus rating because everyone else on the floor with him played like crap.
 
VF21 said:
I'm guess I'm just old-fashioned. I'm going to believe my lying eyes over dry combinations of numbers every day of the week to determine my opinion of a particular player. "We see with our eyes that he (Moreland) had a tremendous impact when he was in the game." I'll agree in total with that statement. But, unlike you, I do not need "statistics that ... back up that assertion" to "know our eyes aren't lying to us."

Using statistics to determine whether or not to watch a particular player might be valid. Using statistics, however, instead of trusting your own impressions is just silly - unless you (not meaning you specifically) are doing so to mask the holes in your knowledge of basketball.

Just sayin'
Click to expand...

For sure, and that's totally up to you, as an NBA fan. The stat geek in me wants to always know more. If my assumptions about players are correct. Get as much information as possible. There's no one right way to enjoy basketball :)
 
bajaden said:
I guess like VF21, I'm old fashion. Doesn't mean I don't look at stats, but the one stat that I pay little or no attention to is the +/- stat. I could spend a long time explaining why, but since I think the stat is irrelevant, I'm not going to waste my time. I watch more basketball than any human being should watch, unless their getting paid to do it. And when I'm watching, I'm not looking at stats.. Let me see a player play in at least 10 games, and I can tell you whether he's any good or not, and what his real pluses and minuses are. If you took the time to watch Moreland play, you didn't need to look up his stats to confirm what you saw. During one of the games I turned to Uncia03 and said, man Moreland must have a lot of boards, he seems to be grabbing everything. At the time, I didn't know how many he had. Turned out he ended up with 12 boards and in 18 minutes of play. You know what, I don't need any stat breakdowns to tell me that's pretty damm impressive.

What I don't need is a stat that gives a minus rating to a player that comes into a game for 12 minutes, scores 12 points, grabs 5 boards and blocks a shot, but still gets a minus rating because everyone else on the floor with him played like crap.
Click to expand...
That's what the adjusted "real" plus/minus stats are supposed to augment - it takes the +/- stat and adjusts based on the quality of the other players on the court.

It's also probably a garbage stat compared to the eyeball test, but I believe it is supposed to supplement what the eye sees rather than the other way around.

*Not a stats geek. Don't care one way or the other.
 
I'm not a stat geek, but I AM a math guy. Any time regularization and overfitting come into play, especially in some unknown formula, well, that raises a red flag.
Without the ability to study it, I can just assume quite a bit of fudging came into play in order to make the current numbers work. That doesn't even take into account future reliability. Unfortunately, we don't have the option of studying the formula (why?) and are being asked to take it as gospel. No thanks.

It's like buying a gluten free cake just because the chef says it is.
"Can you tell me the ingredients?"
"Nope. Trust me."
Then you spend all night on the crapper. Huh. It's actually quite apt, now that I think of it.
 
Bricklayer said:
No no, do not confuse having no need to use it or desire to do so with no idea how to.
Click to expand...

You keep saying so but unfortunately have proven to have no idea what these particular stats actually say.

See, the problem is that you view these advanced statistical formulas as if they are supposed to be an end-all, be-all, end-all-debate formula. They're not. The purpose of advanced stats is to be MORE descriptive, not less so. This is a concept that, for whatever reason, you either fail to grasp or refuse to do so. Its at the point of hilarity how insistently you are striking down this straw man. And I frankly can't understand why someone would deliberately place themselves at such a competitive disadvantage so as to

Advanced stats will never replace GMs. But GMs who use advanced stats are very rapidly replacing GMs who don't. Its an irreversible trend that all of the stubborn old fogies who refuse to adapt to the newest, most accurate tools are being rightfully kicked to the curbside. Deliberately sticking your head in the sand is the fastest way for the game to zip right on by you and render you utterly irrelevant.

And to keep this argument from getting circular, I now drop the gauntlet:

Given that an almost infinite variety of statistical formulas can be derived by combining, recombining, balancing etc. etc. the dozen or so most common basic stats, if I spit out 100 such formulas right now, producing 100 different lists, how exactly would you go about determining which ones had value and which ones did not other than looking at the lists and comparing them to what you personally already knew about the league and its players? The answer is there is no other way. You either determine their validity based on what you know to be true, or you play dumb and are forced to accept them all as having intrinsic value. Including my times he picks his nose/0.25FTA * 317.1TO * stl/min formula. It too will produce a list of players. If you refuse to engage your brain and dismiss it out of hand, then refuse to engage your brain and dismiss it once you see the list it produces, then there you go. It too is an "advanced" formula. And hey, unlike the RAPM guys, I'm not even too humiliated at my crap to let you see the formula itself.
Click to expand...

Hey look, you inadvertently found a statistic measuring Boogers Produced Per 36 minutes. Neat :p
 
You must log in or register to reply here.

Similar threads

CruzDude
OK SO WHAT DID WE LEARN ABOUT THE KINGS THIS PAST SEASON?
Replies
18
Views
2K
Gilles
G
Y
Sim Bhullar's coming to the Sacramento Kings, and that's a pretty big deal (Ball Don't Lie)
Replies
0
Views
484
Yahoo! Sports - NBA -
Y
'89 Kings Fan
Year in Review: Ryan Hollins
Replies
11
Views
2K
Spike
Spike
Y
Meet Sim Bhullar, the Kings' 7-foot-5 center project and the NBA's first player of Indian...
Replies
0
Views
328
Yahoo! Sports - NBA -
Y
Y
The Mamba in flight: A Kobe Bryant Dunk History retrospective (Ball Don't Lie)
Replies
0
Views
582
Yahoo! Sports - NBA -
Y
Back
Top