Bonn1997 wrote:

Wrong because I found an actual threshhold of FGA with the particular player. Like why didn't I choose 19 or 22 or 26. The 20FGA were thrown out there by the number crunchers because they found how KOME's FGA affect the team's W-L column and it's no different say with a player at the point guard position and looking at Ast/To ratio or say when a team scores a certain amount of points or allows a certain amount of points, you can discern a lot more from these statistical measures in regards to end results.

You can do that for many players. Over the past four years, it appears that Durant's team's lose when he takes over 19 shots and win when he takes under 19. I guess 19 is the super magical threshold. I can imagine a game where he's already taken 19 shots and it's tied with 4 seconds to go and you have to diagram a play. So you conclude, we're going to win as long as anyone other than Durant shoots!

No Bonn you still don't get it....

It doesn't mean Kobe should never shoot above 20FGA/gm, it means he should focus less on shooting and doing the other things to help his team win. Naturally the numbers would reflect he's doing this. I knew you were thinking along these lines. So predictable!

Interesting to note although Durant has shot over 20FGA in games many times in his career his career FGA/gm average is usually below 20/gm. Durant wouldn't necessarily be the player to look at, it would be the player who takes the highest volume or has the highest usage of possessions and see what the results are.

Westbrook would be that player for OKC and more than likely hon in on his ast totals

NYKMentality wrote:I couldn't care less what a worthless "win share" has to say, because the fact of the matter is this... The Knicks are only 3-3 without Carmelo Anthony. Struggling to even play .500 ball without Melo, but yet, we're 20-7 and soon to be 21-7 with Melo starting/playing. Melo's won us a lot more games than what some "win share" says on a piece of paper.

Watch the games.

The 3-3 is not bad of a stat you have to consider injuries overlapped oh and that would mean Lin's 9-1 record last year without Melo(MVP candidate and Amar'e Max Player) was rather significant.

The 3-3 record shows me we have a better team than given credit for sans Melo.

I do agree with you about Win Shares and Wins Produced relevant but.....

O-V-E-R-A-T-E-D...Clap Clap....Clap Clap Clap

Bonn1997 wrote:

mrKnickShot wrote:

Bonn1997 wrote:

mrKnickShot wrote:

Bonn1997 wrote:

Bonn1997 wrote:

mrKnickShot wrote:

Bonn1997 wrote:If every case is different, MrK, you need to identify the variables differentiating them so that we know when to trust and when not to trust the sabermetrics. Otherwise, you're just saying "they're not trustworthy when they don't confirm my pre-existing beliefs," which basically just means you think they are useless.

Did I say useless? Please provide evidence that I said this since I do not recall this.

I believe that they are a work in progress and therefore not to be relied on.

And yes. I can take it for what it is and use it as an assistant especially when the numbers are alarmingly bad. However, nothing beats the eye test.

If you don't watch games then stats is all you have.

If you watch Kobe play and see the glaring issues that don't make it into saberville you start seeing that numbers are good but often can lie

I didn't say that you said they were useless. I think that follows from the way you're using them though.
If nothing beats the eye test, why are GMs so clueless about how much to pay players?

Basically since you're not open to the possibility that your eye test is wrong and the stats are right, all you're really doing is using an eye test. If we simplify the scenario by evaluating players simply as either "good" or "bad," we have the following scenarios

A) Eye test: player is good; stats: player is good. Your conclusion: Player is good, stats are right
B) Eye test: player is good; stats: player is bad. Your conclusion: Player is good, stats are wrong
C) Eye test: player is bad; stats: player is good. Your conclusion: Player is bad, stats are wrong
D) Eye test: player is bad: stats: player is bad: Your conclusion: Player is bad, stats are right

This both an exaggeration and simplification but I believe it fits how you're using the stats. It doesn't take a genius to see that your just doing an eye test and the stats have no bearing on your evaluation of the player. Of course, if you can provide examples of when you thought the stats were right and your eye test was wrong, it will invalidate my A through D statements.

Simple Bonn (please don't make me take out the puppets)

Data will tell me who to scout. Eyes will finalize my decision.

Watch 20 laker games (with an open mind) and come back to me.

Why use the data to determine who to scout? If you've got tens or hundreds of millions of dollars, why not just send several scouts to each game and have them write up reports based on their eye tests (which you say are superior to the #s)?

Because I believe in data and respect statistics.

I am also aware of their potential to be wrong. Or at least misleading.

Watch me turn it around. Why have scouts? Just use stats. Why even go to a game? Just play fantasy GM from the caman islands while enjoying da hooka

That's a strawman argument. (I wish I had a dime for every time people made those arguments against me!) I never said to give 100% of the weight to the stats.

I never said to give 100% of the weight to scouts.

Wow! Did that just happen again?

Bonn1997 wrote:

mrKnickShot wrote:

Bonn1997 wrote:

mrKnickShot wrote:

Bonn1997 wrote:

mrKnickShot wrote:

Bonn1997 wrote:

Bonn1997 wrote:

mrKnickShot wrote:

Bonn1997 wrote:If every case is different, MrK, you need to identify the variables differentiating them so that we know when to trust and when not to trust the sabermetrics. Otherwise, you're just saying "they're not trustworthy when they don't confirm my pre-existing beliefs," which basically just means you think they are useless.

Did I say useless? Please provide evidence that I said this since I do not recall this.

I believe that they are a work in progress and therefore not to be relied on.

And yes. I can take it for what it is and use it as an assistant especially when the numbers are alarmingly bad. However, nothing beats the eye test.

If you don't watch games then stats is all you have.

If you watch Kobe play and see the glaring issues that don't make it into saberville you start seeing that numbers are good but often can lie

I didn't say that you said they were useless. I think that follows from the way you're using them though.
If nothing beats the eye test, why are GMs so clueless about how much to pay players?

Basically since you're not open to the possibility that your eye test is wrong and the stats are right, all you're really doing is using an eye test. If we simplify the scenario by evaluating players simply as either "good" or "bad," we have the following scenarios

A) Eye test: player is good; stats: player is good. Your conclusion: Player is good, stats are right
B) Eye test: player is good; stats: player is bad. Your conclusion: Player is good, stats are wrong
C) Eye test: player is bad; stats: player is good. Your conclusion: Player is bad, stats are wrong
D) Eye test: player is bad: stats: player is bad: Your conclusion: Player is bad, stats are right

This both an exaggeration and simplification but I believe it fits how you're using the stats. It doesn't take a genius to see that your just doing an eye test and the stats have no bearing on your evaluation of the player. Of course, if you can provide examples of when you thought the stats were right and your eye test was wrong, it will invalidate my A through D statements.

Simple Bonn (please don't make me take out the puppets)

Data will tell me who to scout. Eyes will finalize my decision.

Watch 20 laker games (with an open mind) and come back to me.

Why use the data to determine who to scout? If you've got tens or hundreds of millions of dollars, why not just send several scouts to each game and have them write up reports based on their eye tests (which you say are superior to the #s)?

Because I believe in data and respect statistics.

I am also aware of their potential to be wrong. Or at least misleading.

Watch me turn it around. Why have scouts? Just use stats. Why even go to a game? Just play fantasy GM from the caman islands while enjoying da hooka

That's a strawman argument. (I wish I had a dime for every time people made those arguments against me!) I never said to give 100% of the weight to the stats.

I never said to give 100% of the weight to scouts.

Wow! Did that just happen again?

I was pointing out that your "turn it around" made no sense.
Anyway, you did say you give 100% weight to the eye test when it conflicts with the stats, though, right?

Did I say that? Again you will have to show me evidence since I suspect that you are mis-remembering again.

I never said I give 100 percent to either.

Bonn1997 wrote:

mrKnickShot wrote:

Bonn1997 wrote:

mrKnickShot wrote:

Bonn1997 wrote:

mrKnickShot wrote:

Bonn1997 wrote:

mrKnickShot wrote:

Bonn1997 wrote:

Bonn1997 wrote:

mrKnickShot wrote:

Bonn1997 wrote:If every case is different, MrK, you need to identify the variables differentiating them so that we know when to trust and when not to trust the sabermetrics. Otherwise, you're just saying "they're not trustworthy when they don't confirm my pre-existing beliefs," which basically just means you think they are useless.

Did I say useless? Please provide evidence that I said this since I do not recall this.

I believe that they are a work in progress and therefore not to be relied on.

And yes. I can take it for what it is and use it as an assistant especially when the numbers are alarmingly bad. However, nothing beats the eye test.

If you don't watch games then stats is all you have.

If you watch Kobe play and see the glaring issues that don't make it into saberville you start seeing that numbers are good but often can lie

I didn't say that you said they were useless. I think that follows from the way you're using them though.
If nothing beats the eye test, why are GMs so clueless about how much to pay players?

Basically since you're not open to the possibility that your eye test is wrong and the stats are right, all you're really doing is using an eye test. If we simplify the scenario by evaluating players simply as either "good" or "bad," we have the following scenarios

A) Eye test: player is good; stats: player is good. Your conclusion: Player is good, stats are right
B) Eye test: player is good; stats: player is bad. Your conclusion: Player is good, stats are wrong
C) Eye test: player is bad; stats: player is good. Your conclusion: Player is bad, stats are wrong
D) Eye test: player is bad: stats: player is bad: Your conclusion: Player is bad, stats are right

This both an exaggeration and simplification but I believe it fits how you're using the stats. It doesn't take a genius to see that your just doing an eye test and the stats have no bearing on your evaluation of the player. Of course, if you can provide examples of when you thought the stats were right and your eye test was wrong, it will invalidate my A through D statements.

Simple Bonn (please don't make me take out the puppets)

Data will tell me who to scout. Eyes will finalize my decision.

Watch 20 laker games (with an open mind) and come back to me.

Why use the data to determine who to scout? If you've got tens or hundreds of millions of dollars, why not just send several scouts to each game and have them write up reports based on their eye tests (which you say are superior to the #s)?

Because I believe in data and respect statistics.

I am also aware of their potential to be wrong. Or at least misleading.

Watch me turn it around. Why have scouts? Just use stats. Why even go to a game? Just play fantasy GM from the caman islands while enjoying da hooka

That's a strawman argument. (I wish I had a dime for every time people made those arguments against me!) I never said to give 100% of the weight to the stats.

I never said to give 100% of the weight to scouts.

Wow! Did that just happen again?

I was pointing out that your "turn it around" made no sense.
Anyway, you did say you give 100% weight to the eye test when it conflicts with the stats, though, right?

Did I say that? Again you will have to show me evidence since I suspect that you are mis-remembering again.

I never said I give 100 percent to either.

You said you go with the eye test when they conflict, which sounds like it means you are giving the stats zero weight when they conflict. If it's not zero, then how much weight would you give the sabermetrics when evaluating Kobe's season?

Did I say that? Evidence please!

43-45 percent - depending on how many drinks I had that can blur my eye test

mrKnickShot wrote:

Bonn1997 wrote:

mrKnickShot wrote:

Bonn1997 wrote:

3G4G wrote:

Bonn1997 wrote:

3G4G wrote:

Bonn1997 wrote:

mrKnickShot wrote:

3G4G wrote:When you place Kobe higher than he should be it makes Melo's case weaker. Not sure how you guys don't understand this.

Kobe should not be ranked in the top 30. He is killing his team with his selfishness.

Kobe and Melo are going in opposite directions.

Kobe is hitting shots at an excellent rate and passing very effectively. Old age is killing his team.

Kobe is getting his numbers at the expense of team success. You can be personally efficient and yet not play in a team aspect. Kobe takes unnecessary heat checks, still shoots over double triple teams, still trying to pull off trick shots. He's hitting a good portion of them at the expense of team chemistry which Dwight Howard says the team has none. Kobe needs to let Nash be Nash, let Howard be Howard, get Gasol the guy who really helped him get rings post Shaq era involved instead of telling him to put big boy pants on. And while blame can be placed on D'AnToni or Brown or whoever Kobe has been there the longest he knows what it takes to win but I'm pretty positive he doesn't care this year. Even James Worthy has called his play disgusting of late. Kobe endorsed D'AnToni over Phil which begs to question why? Kobe is loving his personal scoring accomplishments this year and to come no doubt about it.

I mean he recently commented to the press about how well he's playing and constantly big upping in state rival Clippers, while calling his team old. ROTFLOL @ this clown.

Nothing says Top 5 MVP about the above from him, nothing.

Bonn sorry Kobe is killing your stance and causing you to set a double standard.

Sorry but you have not provided any evidence that my claim is wrong. You're merely stating opinions but presenting them like facts.

So I ask you to present more to the table than just scoring efficiency

How about win shares and wins produced?

That's where it fails. I buy into these stats at some degree but the fact that it gives the same weight to last three shots of a game as it does to the first three shots of a game and that it does not/cannot account for team chemistry makes this fail in this scenario (IMHO).

This is the 80-20 rule where it is in the 20. Eye test wins out for kobe.

Or perhaps your view that 2 points later in the game count more than 2 points earlier in the game is flawed. Anyway, are you saying you give 80% weight to the stats and 20% to your personal observations (or the reverse)? If so, I would not object at all. It doesn't sound like you're giving the stats 80% of the weight in your evaluation of Kobe though.

If the stats are correct 80 percent of the time and incorrect 20 percent of the time, then kobe is part of the 20 percent.

As far as valuing the points, yes. I think the a last second basket should have a higher value than the first shot of the game. The fact that WS and WP don't account for clutch stats makes it flawed in this areas.

If you want to reduce it to just game winning shots, he's still pretty close to his career average (48.1%) and among the leaders but you're also just talking about only a couple dozen shots spread out over an entire career.
http://www.82games.com/gamewinningshots.htm

For the record, I don't believe points in "clutch" moments are more valuable than they are early in the game but I wanted to bring in data relevant to your points