[crockpot]

[ QUOTE ]
[ QUOTE ]
also you must consider that in a league with a lot of trading, teams will trend toward the mean in most categories as they search for balance in midseason.

[/ QUOTE ]

tx. does this mean that I should overpay for the top performers? ie pay more than you would expect (linearly) for 15 steals instead of 10?

[/ QUOTE ]

basically it means that in the situation i described, adding 11 steals will increase your projected fantasy points by 1 or 2, not 5. assuming a normal distribution of projections in a category, you should expect about 11.5 points when you're projected to tie for first in a category, and 4.5 when you're projected to tie for last, rather than 14.5 or 1.5. does that make sense?

the elite performers are worth more than their linear value in total, but this does not apply to single categories, where an elite performance is actually worth less. to take an extreme example, if you drafted rickey henderson or vince coleman in their primes, you were unlikely to see the full benefit of 100+ steals since you would likely win the category in a landslide, and the steals you got from further players were wasted.

[ QUOTE ]
side question - I have guys with similar PECOTA projections, that play the same position, but have different salaries.

We have a fairly restricted trading cap - our draft budget was $275, but our cap is now $285.

yearly inflation for keeping a guy is $3.

my long winded question is, if I Chad Tracy at $5, who projects better than Mark Teahan who costs $11, I should want to trade Teahan, correct? (I also have Troy Glaus and Frank Thomas, so 1 should be trade bait I think)

[/ QUOTE ]

basically, yes. if you can get a better return for teahen, you want to deal him. but if tracy is in higher demand, don't be afraid to trade him instead, assuming you get value out of it.