ask me about fantasy baseball v.2007

[crockpot]

[ QUOTE ]
so, I'm trying to come up with a model in excel which will help me evaluate trades. I suppose this is for mostly pre-season/early season trades, where PECOTA projections still are a very powerful tool.

[/ QUOTE ]

not bad. this needs to be combined with a basic understanding of how variance will affect the standings.

for example, if the team with the highest projected steals total is 10 above you in sixth place, adding 11 steals will not equate to a five point expected gain in the standings. 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.

still, this exercise will give you an idea of how to adjust standings gain points for your league setup, especially if you keep it up during the season.

[Kneel B4 Zod]

[ 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?

(that may not make sense, sorry, but I guess I'm asking if it can make sense to pay say $10 for 5 steals but $12 for 10 steals)

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)

[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.