James Harden trade

[mystic]

My projections are here: http://bit.ly/Y41J5n

I had them as a -4.0 PD team, so 29 wins.

Thanks. But 3 wins difference or 1 PD to my projection warrants a statement: overly generous?

I'll write a collective table and adjust for this trade sometime tonight. However there's a hurricane going on so I may not pump it out before the season starts!

Puh, I hope everything goes well for you and your family/friends/neighbours/etc.

[Bobbofitos]

My projections are here: http://bit.ly/Y41J5n

I had them as a -4.0 PD team, so 29 wins.

Thanks. But 3 wins difference or 1 PD to my projection warrants a statement: overly generous?

You are right, my language was over the top. I suppose simply "generous" applies

I'll write a collective table and adjust for this trade sometime tonight. However there's a hurricane going on so I may not pump it out before the season starts!

Puh, I hope everything goes well for you and your family.

Thanks, nothing to worry about. Went for a walk earlier in fact.

[DSMok1]

mystic, what I did to generate the marginal value:

-Calculated median production for the first 4 years
-Converted production to dollars, based on league-wide average value (3.8 million per point of VORP)
-Compare to rookie wage scale

I do need to update for the latest CBA, but shouldn't change that much...

I saw the thread in which you explained it. But I really think that there is not a linear relationship between "production" and "price". Especially the veteran minimum players increased their value under the new CBA and at the first glance the mid to late 1st rounders became less valuable. As a contending team I ask myself: Do I want to develop a young player or can I just sign a veteran instead? I guess that also implies that the value of a draft pick changes with the different strength of teams.

Also, why are you using median and not average? Isn't it more likely to get an average player for the position? Assuming average is lower than median value.

All of the derivations for value vs. price that I have seen indicate that, if you set replacement level right, Production over Replacement scales linearly with price. I know for a fact that has been established for baseball.

[mystic]

All of the derivations for value vs. price that I have seen indicate that, if you set replacement level right, Production over Replacement scales linearly with price. I know for a fact that has been established for baseball.

I guess, that is the important part. I can see that being linear for a correct replacement level. But that draws up a question for me which is not really answered to my satisfaction: What is the replacement level?

Personally, I can handle things based on league average level easier.

[DSMok1]

All of the derivations for value vs. price that I have seen indicate that, if you set replacement level right, Production over Replacement scales linearly with price. I know for a fact that has been established for baseball.

I guess, that is the important part. I can see that being linear for a correct replacement level. But that draws up a question for me which is not really answered to my satisfaction: What is the replacement level?

Personally, I can handle things based on league average level easier.

I've done a lot of work with replacement level, and I'm pretty confident it is somewhere between -3.0 and -3.5 pts/100 possessions.

Check out this visualization to get at least a qualitative perspective: http://public.tableausoftware.com/views ... l?:embed=y

[bchaikin]

Anyone who believes the Thunder are still a contender are mistaken. The #1 offense owed a huge portion of credit to Harden. He can't simply be replicated, ESPECIALLY not by KMart.

why not kevin martin?...

considering that they had made up their minds to trade harden once he turned down their latest contract offer, if they were looking in particular to replace harden's offense at SG, they certainly could have done a whole lot worse...

just 2 years ago (2010-11) martin was a top scoring SG that was also very efficient on offense. he scored 23.5 pts/g and 28.8 pts/40min. the only SG that year to score at a better per minute rate was kobe bryant. and his offensive efficiency of 2.51 pts/0ptposs (points scored per zero point team possession personally responsible for) was best among all SGs scoring at least 17 pts/g that season...

not only that - but for the 5 years of 06-07 to 10-11 his scoring rate was 22.4 pts/g and 25.5 pts/40min. the only SGs with a better per minute scoring rate were kobe bryant and dwyane wade, and the only SG more efficient was ray allen. these 5 seasons he averaged close to 10 FTA/40min and shot 87% on FTs, and scored 1/3 of his total points from the FT line - this despite the fact that he also took 1/3 of his FGAs as 3pters...

if anything martin's 2011-12 season of poorer offense could simply be an outlier. since they only have to pay him for one more year, if they were in fact looking to replace harden's offensive production at SG knowing that they had to trade him, at just 29 years of age kevin martin is a very good gamble. if he reverts back to his high scoring/high efficiency, they can resign him. if he doesn't, they can let him walk, and they still got three 1st round draft picks (jeremy lamb plus two future picks)...

[Bobbofitos]

Anyone who believes the Thunder are still a contender are mistaken. The #1 offense owed a huge portion of credit to Harden. He can't simply be replicated, ESPECIALLY not by KMart.

why not kevin martin?...

considering that they had made up their minds to trade harden once he turned down their latest contract offer, if they were looking in particular to replace harden's offense at SG, they certainly could have done a whole lot worse...

just 2 years ago (2010-11) martin was a top scoring SG that was also very efficient on offense. he scored 23.5 pts/g and 28.8 pts/40min. the only SG that year to score at a better per minute rate was kobe bryant. and his offensive efficiency of 2.51 pts/0ptposs (points scored per zero point team possession personally responsible for) was best among all SGs scoring at least 17 pts/g that season...

not only that - but for the 5 years of 06-07 to 10-11 his scoring rate was 22.4 pts/g and 25.5 pts/40min. the only SGs with a better per minute scoring rate were kobe bryant and dwyane wade, and the only SG more efficient was ray allen. these 5 seasons he averaged close to 10 FTA/40min and shot 87% on FTs, and scored 1/3 of his total points from the FT line - this despite the fact that he also took 1/3 of his FGAs as 3pters...

if anything martin's 2011-12 season of poorer offense could simply be an outlier. since they only have to pay him for one more year, if they were in fact looking to replace harden's offensive production at SG knowing that they had to trade him, at just 29 years of age kevin martin is a very good gamble. if he reverts back to his high scoring/high efficiency, they can resign him. if he doesn't, they can let him walk, and they still got three 1st round draft picks (jeremy lamb plus two future picks)...

About the bolded: I don't think last year is an outlier. I think he is the most affected player by the abolishment of the rip rule, and the data follows suit. (His foul rate was split in half)

He is an inferior playmaker (relative to Harden) so I can't imagine a RWB-Kmart-Durant lineup working nearly to the extent it did w/ Harden. If they are planning on using KMart as a 3pt specialist, he's an above average shooter, but nothing special. (37.7 career, 34.7 last year)

[Crow]

just observing:

37.7% 3 pt FG% career is 101st all-time
About 50th best about actives

[bchaikin]

About the bolded: I don't think last year is an outlier.

looking at the last 6 seasons, since 06-07, in 2011-12 martin:

- shot his worst overall...
- got fouled the least per touch (despite similar touches/min last year to 06-07 to 10-11)...
- had his lowest per game and per minute scoring...
- had his lowest per minute offensive rebounding rate...
- had his lowest per minute defensive rebounding rate...
- had his lowest steal rate...
- had his 2nd lowest shot blocking rate...

that's 7 lowests or worsts (well 6 at least) in the same 1 season of the last 6, and the rip rule had nothing to do with 4 of those. that's pretty much the definition of an outlier. so you can look at it in one of two ways - either he has completely lost it at the age of 29, or the odds are high he will improve in at least some of these areas next season. since most players do not "lose it" as early as the age of 29, and he got his 3rd highest annual touches/min on offense last season (in an 8 year career), then again, considering that they decided they had to trade harden, getting kevin martin to replace harden's offense is a very good gamble...

He is an inferior playmaker (relative to Harden) so I can't imagine a RWB-Kmart-Durant lineup working nearly to the extent it did w/ Harden.

i fail to see the relevance here - last year the thunder were dead last in the league in ast/fgm ratio, and dead last in ast/g. despite this they finished 2nd in the league in offensive efficiency (pts/poss scored) and 3rd in the league in eFG%. if anything, they have clearly shown they do not need many assists to have one of the league's best offenses...

[Bobbofitos]

About the bolded: I don't think last year is an outlier.

looking at the last 6 seasons, since 06-07, in 2011-12 martin:

- shot his worst overall...
- got fouled the least per touch (despite similar touches/min last year to 06-07 to 10-11)...
- had his lowest per game and per minute scoring...
- had his lowest per minute offensive rebounding rate...
- had his lowest per minute defensive rebounding rate...
- had his lowest steal rate...
- had his 2nd lowest shot blocking rate...

that's 7 lowests or worsts (well 6 at least) in the same 1 season of the last 6, and the rip rule had nothing to do with 4 of those.

I don't know why you'd reference oreb, dreb, steal rate, or block rate as pertinent to a discussion about his offensive prowess and how he affects the Thunder's offense. I suppose OReb may be slightly relevant, but if he reverts back to his historic norms, that's literally a 0.3/36 difference.

So the conclusion about his offensive impact stays the same.

that's pretty much the definition of an outlier. so you can look at it in one of two ways - either he has completely lost it at the age of 29, or the odds are high he will improve in at least some of these areas next season. since most players do not "lose it" as early as the age of 29, and he got his 3rd highest annual touches/min on offense last season (in an 8 year career), then again, considering that they decided they had to trade harden, getting kevin martin to replace harden's offense is a very good gamble...

They could have done worse as far as Kevin Martin vs others. I'm not that surprised you see him as a better than expected replacement, as box score wise Kevin Martin (historically - not so much last year specifically) grades out far better than other measures would have you believe. (+/-, PBP)

He may even thrive in this role; I suppose that's why they play the games. And you're right, most don't "lose it" by 29. But we have some reasonable data that suggests HE has, namely the change in rules, but also his litany of injuries, which surely has robbed him of *some* athleticism.

He is an inferior playmaker (relative to Harden) so I can't imagine a RWB-Kmart-Durant lineup working nearly to the extent it did w/ Harden.

i fail to see the relevance here - last year the thunder were dead last in the league in ast/fgm ratio, and dead last in ast/g. despite this they finished 2nd in the league in offensive efficiency (pts/poss scored) and 3rd in the league in eFG%. if anything, they have clearly shown they do not need many assists to have one of the league's best offenses...

Sure, they don't need assists to have a great offense. But don't you suppose there's a dropoff, as far as going from a [better passer] to a [less skilled passer]? In other words, we can assume assists are a nonzero positive force. How nonzero is not important. But going from more to less assumes a loss, across the board, of that nonzero force.

Unless you want to somehow argue that his lack of passing is a good thing, I'd say the lack in playmaking is relevant.

[bchaikin]

I don't know why you'd reference oreb, dreb, steal rate, or block rate as pertinent to a discussion about his offensive prowess and how he affects the Thunder's offense.

simply showing the season as a whole - for him - was an outlier statistically pretty much across the board...

So the conclusion about his offensive impact stays the same - I think he is the most affected player by the abolishment of the rip rule, and the data follows suit. (His foul rate was split in half)...

why - because he saw a huge drop in his rate of FTs? what about players like randy foye, richard hamilton, kirk hinrich, richard jefferson, tyrus thomas, martell webster, zaza pachulia, amir johnson, and others? they all saw large drops percentage-wise in their rate of FTs - were these large drops also all due to this one rule change? or is it just kevin martin because he saw the largest drop in FTA/g?...

these drops can be due to any number of factors - the rule change, injuries, age, new teammates, etc. the league as a whole saw a very large drop of 1.9 FTA/g in just the one year from 10-11 to 11-12, partly due to the rule change. but the league has also seen drops (or increases) in FTA/g almost as large (1.0-1.5 FTA/g) from one season to the next without any rule changes...

He may even thrive in this role;

yes - if 2011-12 is an outlier for him (and was not due say to a still currently lingering injury)...

And you're right, most don't "lose it" by 29. But we have some reasonable data that suggests HE has, namely the change in rules, but also his litany of injuries, which surely has robbed him of *some* athleticism.

if you are so sure of this, are you also this sure of the other players mentioned above who also saw large drops in their FTs?...

Sure, they don't need assists to have a great offense. But don't you suppose there's a dropoff, as far as going from a [better passer] to a [less skilled passer]?

of course not - why would there be? you just said it yourself - they don't need assists to have a great offense...

Unless you want to somehow argue that his lack of passing is a good thing,

absolutely - since 06-07 martin has been one of the highest scoring and most offensively efficient players in the league, almost as high scoring/efficient as dirk nowitzki. from the ages of 23-28 (06-07 to 11-12) martin scored at a 25.1 pts/40min rate (efficiency of 2.48 pts/0ptposs), while passing for 2.8 ast/40min. dirk nowitzki in the same age range scored at a 25.9 pts/40min rate (efficiency of 2.65 pts/0ptposs) while passing for 3.0 ast/40min. so why would you want him passing more? would you say nowitzki should have passed more?...

I'd say the lack in playmaking is relevant.

relevant based on what?...

[J.E.]

37/12/6/4 on 14-25 in his first game. Houston's FO has to be feeling good right now

[DSMok1]

37/12/6/4 on 14-25 in his first game. Houston's FO has to be feeling good right now

Miami's Big 3 are less offensively than they were separately. I think the same will happen with KD/Westbrook/Harden, in reverse--redundancies eliminated.

[KAN]

I've long wondered how big of a factor this is. As they say there is only one ball.

I guess it would be easy to test. Compare expected Ortg of 5 man unit versus actual based on offensive RAPMs and see how it scales as the expected gets higher and higher.