|
| Draft Strategy: Numerical Crunch Time Services vs. Value Based Drafts |
| Posted on September 1,
2004 |
|
Paul Baitinger
Senior Editor
|
In a nut shell, using Value Based Drafting (VBD)
strategies can be both time consuming and error prone, but
still, it's certainly worth a look in terms support
information on draft day. I find that VBD tools alone are
most helpful when you are trying to formulate an overall
strategy OR when you are trying to determine your strategy
for your next few picks. It is also a decent supporting tool
when trying to choose between two players. However, I prefer
our Numerical Crunch Time Services with the Point
Differential Reference Guide, Generic Comparative Analysis
Matrix (GCAM), and Customized - Comparative Analysis Matrix
(C-CAM)
and most specifically the C-CAM works better rather
than a pure VBD. One look at my 2004
Writer's League Draft and you'll notice that I was very
fond of selecting a mix of players that would yield the most
fantasy points. I snatched up a lot of quality RBs
that I had ranked fairly high and waited on positions like
WR a bit because I saw very little difference between
players after the top few WRs were taken.
Why VBD tools on other sites may be time consuming and
error prone
First, you probably need to adjust the fantasy
scoring rules provided by whatever VBD tool you are using.
This can take some time and the tool may not account for
your exact rules, although it will usually be close enough
especially if you use our (Point Differential Reference
Guide).
Second, you need to get statistics over the last 3
or so years AND plug in projections for the upcoming year
end totals. Most tools provide you with figures from the
last three year's and give you their projections, but you'll
want to update many of the projections yourself.
Third, you need to rely on subjective opinions from
the tool creator or your own personal subjective opinions to
input projected year end stats for EVERY "draftable" player
(probably 200 players). This is highly error prone AND very
time consuming. Additionally, if you're a homer, you may
overvalue players from your favorite team based on how you
"hope" they will perform.
Fourth, you need to analyze past performance for
trends and adjust your projections per the trends. If you
play in the stock market you likely understand that past
performance is not always an indication of future results.
This alone doesn't always take into account trends in which
offenses can take off and fly. Value Based Drafting would
have undervalued many offensive players from the the Bills
in 2002 unless you adjusted their projected stats
accordingly. Bledsoe, Travis Henry, Eric Moulds, and
Peerless Price were all fantasy studs that year as we
predicted. Guys like Santana Moss and Matt Hasselbeck
took off last year and the Lions' offense shows a lot of
potential this year.
Fifth, you have to look at the players schedule and
formulate the ease or difficulty of their schedule. Most VBD
tools only look at the projections based on how the
opposition performed last year. If you follow the NFL, you
would have noticed that the Bills and Patriots made some
upgrades to their defenses before the 2003 season, thus you
would have needed to adjust the projections for the players
that played them accordingly. This year we look for
the Raiders run defense to be better and we see the Seahawks
and Vikings making improvements to their pass defense.
Lastly, most people don't truly understand that they need
to look at the BIG picture when using VBD tools. For
example, you really need to compare the difference between a
QB that you will get in Round 2, versus a RB or a WR in
Round 2 with an eye towards the "Value" you will get at RB
or WR if you wait to draft them until later rounds.
When you add all of these factors together you are subject
to multiple points of failure or multiple error controls.
The AskTheCommish.com staff drafted in a number of expert
leagues over the last few years. A number of the experts in
these leagues stuck by their guns and went with their VBD
formula. Meanwhile, we used our AskTheCommish.com rankings,
our expert judgment, and sprinkled in our Average Value
Based C-CAM Rankings and our Point Differential Reference
Guide to draft. A few years ago the VBD only experts loaded
up on RBs left and right early, but we steered our course
and drafted Harrison and Manning in the first two rounds and
then loaded up on RBs in the middle rounds. After that, we
looked at the players that we had ranked the highest and
started drafting them. We snagged players later than we had
him ranked because the VBD did not account for their
potential to have a "breakout" year, thus, they were
overlooked by the VBD experts. Needless to say, we have won
many leagues and have enjoyed much success in our fantasy
leagues while reaping the benefits of taking players that
other owners were leaving behind.
Additionally, some folks that use VBD take forever to
select their picks as they may continually need to perform
mathematical calculations to determine their best value.
While we understand that fantasy football is about winning
and the best way to win is to make the right picks, we
ourselves get annoyed with the guy who takes upwards of 15
minutes to make their selection of a backup kicker, rather
than a 5th WR.
Why VBD tools are useful
The VBD theory helps give each fantasy drafter a snapshot
of value of players at any position with an apples to apples
comparison using YOUR fantasy league rules. The VBD tools
like our Point Differential Reference Guide are used to
determine the value of drafting a RB versus a QB, WR, or TE
at any point in the draft. It can also be used to indicate
when potential runs will occur on a particular position. We
feel that it is best to be on the back end of a run rather
than starting a run. Therefore, if you are drafting in the
middle of a round you are best suited to let others start
the run as you can likely get better value at another
position.
Customized - Comparative Analysis Matrix
Our
Customized Comparative Analysis Matrix (C-CAM),
which was developed by Senior Editor Al Lackner, allows you
to quickly compare your ranked players at one position to
your ranked players at other positions. Now you can compare
apples to apples to determine if a RB is more valuable than
a QB, WR, or TE in terms of fantasy points. For instance,
Holmes was the RB1 in 2002 and 2003, while Faulk was the RB1
in 2001. Regardless of whom you have ranked as your RB1 this
year, the Average Fantasy Point Value for the RB1 is 299. If
you use these rankings, notice that you need look at the BIG
picture and compare the value that you’ll likely get at
other positions later in the draft.
You should also use these rankings to determine the drop
off in production when debating between picking two players.
Say you are picking 9th, 15th, 33rd, and 40th in a draft.
With the 9th pick you are debating between WR1 (205.3) or
RB9 (167), with the 15th pick you look at QB2 (230), RB13
(140.7), or WR3 (176.2), with the 33rd pick you look at QB4
(220), RB19 (123.6), or WR6 (136.7), and with the 40th pick
you are simply looking for the best value. You’ll see that
waiting before picking your QB would likely yield the most
value. Remember, you are not comparing just two players, but
rather the difference in many players!
Notice that our C-CAM isn't time consuming for the fantasy
drafter. The matrices are simply "Cheat Sheets" that
provides the average slotted value for a player ranked at
the corresponding position. More importantly, if you do not
have the time and inclination to enter your own league rules
and player projections to generate a true C-CAM, we have
also provided a Generic Comparative Analysis Matrix (GCAM)
based on purely objective numerical analysis and standard
league rules.
Whether using the GCAM or C-CAM, either can be used to
compare values across multiple positions. For instance, you
can quickly see that RB10 holds roughly the same value as
QB15 and WR3.
If you take this one step further you see a difference of
almost 203 points when comparing the RB1 to RB26, a
difference of about 138 points when comparing QB1 to QB26,
and a difference of 126 points when comparing WR1 to WR26.
Our subjective rankings shows us that there is very little
difference between RB1 and RB6, but our C-CAM shows a
difference of about 85 points and a marked difference
between RB7 to RB12 of about 39 points. Given that most
leagues start 2 RBs, they will surely fly off the board
quickly with such drop-off in production.
Our subjective rankings shows us that there is very little
difference between QB1 and QB6 and this is supported by our
C-CAM with a difference of 36 points. Furthermore, we see
little difference between QB7 and QB12 (20 points). You can
likely wait for your QBs unless one falls into your lap at
the right spot.
As far as WR's go, after the first few WRs, they are
typically a dime a dozen. Players such as Anquan Boldin,
Santana Moss, Chris Chambers, Keenan McCardell, and Steve
Smith weren't considered top tier WRs and many were likely
waiver wire pickups, yet they all finished as top 15 or so
WRs. Our C-CAM shows that there is usually a marked
difference between the top 3 WRs and everyone else, but last
year we saw a big difference from WR1 to WR2 and another
sizable difference from WR2 to WR3. However, after
that there is little difference as the difference between
WR7 and WR12 is only about 12 points.
Now onto TEs, you typically used to notice a difference of
only about 22 points between the top TE and the 4th TE, but
last year after Gonzalez and Sharpe the drop-off was
incredible (the Point Differential from the 1st to the 3rd
TE was approximately 50 points). There is 1 clear cut top TE
this year. If you don't land him, you are usually better
served waiting awhile. As such, you notice that top TE is
generally drafted near the end of the 3rd round or early 4th
round in a 12 team league. The reason is that the difference
between RB21 and RB24 is only about 12 points, while the
difference between QB7 and QB9 is 8 points, and the
difference between WR6 and WR 9 is only 16 points.
Now a note... If you're a Premium Member, then
you’re probably wondering why we show the GCAM for only the
top 32 players at QB, RB, WR, and TE, especially for RB and
WR. First off, after the first few rounds of the draft, you
will likely need to draft players that will fill positions
of need on your team. Second, after you get past the 32
players at RB you’re mostly dealing with reach picks,
sleepers, or RB by committee players and there is very
little difference between WRs 32-48 or even 60-75. NOTE:
Kickers, Team Defenses, and Individual Defensive players do
not belong in such a ranking.
For more information on our Numerical Crunch Time
Services check out our Premium Services page:
http://www.askthecommish.com/services/
|
|
| Sound Off! |
| Click here to discuss this article on our message boards! |
| |
|
|
