5thhorseman wrote: ↑Mon Oct 17, 2022 4:52 pm
Ronning's Ghost wrote: ↑Sun Oct 16, 2022 8:52 pm
I have always been interested in the theory of optimized cap distribution in the abstract. In a hypothetical league wherein any level of talent could be matched to a fair market cap hit, what proportion of the cap do you think should go to each of:
#1 C
#2 C
#3 C
#4 C
?
You describe an assumption where every contract is fair value, and presumably value correlates 100% with performance. Putting aside other intangibles such as how well players gel with one another, any allocation will do. The total value (performance) will always be the same. No-one has an edge.
Unless I misunderstood your problem statement.
It seems to me like you understood where I was going generally, but it appears that I didn't explain what I meant by 'any level of talent at fair market cap hit' clearly.
I meant that every level of talent was available: If the GM wanted to spend on McDavid level talent, he could get that, but if he preferred a Trocheck level of talent, that was available, too. And those contracts were at the going rate in the current NHL market for that kind of player (so not an ELC, or a hometown discount, or a pleasant surprise on a late bloomer), and the players were playing up to the expectations of their contracts, with the value GMs place on intangibles at contract time correctly accounted for, too. But in the NHL market, GMs have to pay different amounts per point (or per Corsi For, SV%, leadership value, or whatever other performance characteristics interest him) at different points of the talent scale, as well as different amounts for different types of contributions (offence, defence, toughness, etc.)
I didn't mean for that to imply that contract value would correlate 100% with performance in the sense of contribution to team success per dollar, since I don't think anyone knows -- quantitatively, anyway -- exactly what that is, either on an individual or team basis, and that is the effect I was trying to explore.
That is, how should a GM balance out his roster? So, for example, knowing exactly what fairly-paid high-scoring wingers go for, how many do you want at each scoring level, and at what part of your lineup would you rather start spending on other types of contributions from your wingers, instead? The implication being that if he had too many, or not enough, a smart GM would trade from some part of his lineup to derive the optimal balance. To choose perhaps the most obvious example, no GM would appear to believe that it is worth spending as much on his team's back-up goalie as their starting goalie.
We discuss this sort of thing here in the specific all the time, in the sense of the Canucks should trade from this point of strength to shore up this weakness (largely forwards for D, in the current roster composition), but I was trying to get a sense of the ideal blueprint to which posters were striving.
5thhorseman wrote: ↑Mon Oct 17, 2022 4:52 pm
But the assumption is flawed. Contending teams have players on ELCs or on hometown discounts, which by definition are not fair value. Or sign players who outperform their contracts. Therein lies the edge, not in allocation of cap to fair value contracts.
It is, of course, true that this league of all fair value contracts does not exist. It is a hypothetical I described (or attempted to) for the sake of a thought experiment.
But do you believe this deviation from fair market value is the principle edge a GM can obtain in team composition? Performance above fair market value?
I suppose part of the premise of my question is that team balance is also a significant factor, but maybe it just isn't.