drza wrote:Blackmill asks whether scarce talent is worth valuing, since a player cannot control if his talent is scarce. I say yes, especially in the sense that this type of scarcity is global. There are all types of variations of "you can't teach tall", and there's a reason that such an inordinate percentage of people around 7-feet tall end up in the NBA. To address the logic of the last paragraph above, a player cannot control ANY of his natural talents. You can't choose to be able to jump like Jordan, or to have Iverson's quickness, or to have Shaq's...whatever the heck you call his behemoth-ness. But frankly, whether the player can control their talent is irrelevant to their value to a team.
I understood the underlined beforehand which is why I carefully said:
neither an attribute of Mutumbo nor something he can control
and later added
Even though the number of players you can "just give the ball to and have them 'score at will'" isn't an attribute of players who are perceived to be able to do this, nor something they can control. Just wanted to extend the analogy across the board.
The above post was deleted while I wrote this, so I'm keeping the poster anonymous, but wanted to respond since there seemed to be some confusion.
Kareem's ability to "score at will" I consider the product of his attributes. Kareem's height is an attribute of his, and helps him score, but the number of other players with his height is not. The important distinction being that Kareem's height can be discerned by only observing him.
I think it's important evaluate players by their attributes which includes the mental and the physical. Attributes I fully recognize are outside the player's control.
Should Mutombo be valued because there are few elite defensive players? ABSOLUTELY. To me, that question is like asking whether LeBron should be valued because there are very few that can produce offense at his level. ABSOLUTELY. That's part and parcel of what makes him unique, and thus incredibly valuable.
[...]
Said another way, it's not that Mutombos on defense are sparse because there happened to be a particular season where there weren't many elite defensive bigs and he took advantage. No, it's because in the history of the NBA there are very few (relatively speaking) defensive monsters. There are many more players that can approximate (even roughly) elite offensive impact than there are elite defensive. I thought Micah's breakdown upthread from DocMJ's 1998 - 2012 dataset was very illuminating, putting numbers to my assertion.
We're not in disagreement here.
Blackmill says in the post above that "scarce talent being valuable talent is because non-scarce talent can't be piled up, at least not without diminishing returns". I don't believe that to be accurate. Scarce talent is valuable because not much of it exists, especially in comparison to other talents.
[...]
But if my point is to build the best team I can build, and I know that there are plenty of +5s and +6s on offense that I may be able to find, but hardly any defender that can approximate that +7? Then yeah, I might seriously consider grabbing the defensive guy. Because while his impact, in a vacuum, may be lesser...his likely net impact on my team's ability to create a positive imbalance with respect to the other teams (and thus give my team a better chance to contend) is very likely to be higher.
I almost said in my earlier post that scarce talent can be valuable for reasons other than avoiding diminishing returns. I didn't because the discussion to that point seemed (implicitly) focused on the diminishing returns aspect. However, reading your post, I must be missing something.
Could you answer a couple questions?
I assumed any offensive players mentioned were zeros on defense and any defensive players were zeros on offense. Maybe this wasn't what you had in mind. Was your point a +5 defensive player is more likely to replicate a +5 offensive player than vice versa? That could be true but I don't understand why that matters since either the defensive player does or does not replicate the +5 offensive player.
You write you would consider "grabbing the defensive guy" with the suggestion that, in the future, it would be easier to secure a +5 offensive player than a +7 (or did you mean +4?) defensive player. Is your point that getting a +5 offensive player to pair with the +5 defensive player is easier than acquiring two +5 offensive players?
If neither of the above, then we get to how I interpreted your statement, which is two +5 players don't always translate into a +10 for the team. And more specifically, that a +5 defensive player and +5 offensive player are more likely to be a +10 collectively than two +5 offensive players. Is that correct? From your response, I wouldn't think so, since this is very much diminishing returns.
Or... maybe my reading comprehension has failed me.
I said that scarce talent can provide value beyond avoiding diminishing returns. Scarce, out-of-position talent can allow for unconventional lineups that have large, competitive advantages over the field. Scarce talent such as Curry's shooting or Magic's combined vision and size can make available offensive plays, that other teams may not have experience defending or simply find difficult to counter, which would otherwise be ineffective. A similar statement could be made of defensive players. Maybe this is what you meant but this isn't what I read.
____________________________________________Updated____________________________________________
Happened to post my question a couple minutes after you posted this.
But in the same type of example, if you replace a +2 defensive player by a +5...you'll see a LOT more of the value transfer through. It'd be too simplistic to say exactly +3 will happen, but there's a LOT less potential for diminishing returns. Given the same positional considerations as above (e.g. that your +5 defender is replacing the same position), there's very little chance of negative overlap. The fear of a twin-towers is usually that, if both are centers, it'd be hard for them to coexist as they'd want to use the same spaces. But if one really is a 4 and the other really is a 5, more than likely they can coexist. And if they can...the defensive impact is darn near additive. This is true if the defenders have similar help-defense abilities (think Duncan and Robinson) or if one is a better man defender and one a better help defender (think Sheed and Big Ben).
Anyway. My point is, I believe that, in addition to being a rarer commodity, defense is much more additive than offense.
[...]
That it is both more portable (as defined by the ability of an individual player's impact to be maximized in a variety of environments) AND more scaleable (as defined by the ability of a player's individual impact to improve an already good team).
Related to both of those terms, but not exactly the same, is the [b]concept that a defensive player can add his own contribution while also allowing teammates to be all that they can be.
If only for my own credibility, I ought to point out that while some of the above concepts don't only entail diminishing returns, I think all involve diminishing returns. But whatever you wish to call it, that doesn't matter, since you made good points.
