Here's an axiom (and confession) that might shock some folks: All
fantasy football owners are more often wrong than right when it
comes to doing player projections and rankings.
From experts to novices, and from league champions to cellar-dwellers,
this axiom is universally true. It is a demonstrable fact to any fantasy
player willing to be honest with himself. Or to any fantasy expert
who has his preseason rankings printed in magazines for hundreds of
thousands of fantasy fanatics to read.
I've corresponded with hundreds of fantasy owners who insist on their
brilliance. Every player projection is 100% right. Might as well take
their projections in August and mail them to Stats Inc. with a note:
"Don't bother waiting until the end of the year to see how it
all turns out. Just print this!"
From experts to rookies, we all need to take a reality check. Drop
the facade, take your thorazine, and make this admission: "My
name is ___ ___, and I'm a fantasy football junkie. And I'm more often
wrong than right in my player rankings."
Game Of
Inches
Now take heart, because there is a corollary to this axiom, which
is simply this: Championships are won by those who suck the least
at making projections. It might not sound as inspiring as Draft Sharks'
tag line, "Championships are won on draft day!," but it
is nonetheless true.
To draw a comparison, think of baseball hitters. What's the difference
between a guy who gets a base hit only 32% of the time and one who
gets a base hit 28% of the time. The obvious answer is "4%." But look
a little deeper. A guy who hits .320 every year will play in a whole
bunch of All-Star games, and probably wind up with a bronze bust in
Cooperstown, while the .280 hitter is destined for few accolades.
Even though both hitters produce outs far more often than they produce
hits (read: they are wrong more often than they are right) the player
who performs only marginally better is a Hall-of-Famer!
That marginal difference holds the key to winning in fantasy football,
as well. Just think back to last summer for a moment. Most everyone
thought Brett Favre, Steve Beuerlein, Steve McNair, Drew Bledsoe and
Brad Johnson would be top-10 QBs. Nobody was touting Jeff Garcia,
Donovan McNabb or Elvis Grbac. Go down the other positional rankings,
and you'll find them littered with missed predictions. Emmitt Smith,
Dorsey Levens, Stephen Davis, Jamal Anderson, Terrell Davis were pegged
as top-10 RBs. Preseason WR studs included Antonio Freeman, Jimmy
Smith, Marcus Robinson, Eric Moulds, Muhsin Muhammad and Keyshawn
Johnson. The list of bad rankings goes on and on…
That is to say, ALL of the fantasy experts were more often wrong than
right in their rankings. The same holds true for ANY fantasy owner
who wrote down their rankings from 2000, and goes back to take an
honest look at them. They'll discover that they were more often wrong
than right.
Frankly, Draft Sharks wasn't much better than the consensus in overall
prognosticating last year. In fact, we had some downright ugly picks!
However, the salient point is that we were marginally better than
the consensus, and took second place in the 2000 Fantasy Index Experts
Poll (Yes, after tweaking our rankings, we got killed in the 2000
Pro Forecasts Masters Poll. But we were the defending Champions of
the 1999 Masters Poll) Much like the .320 hitter vs. the .280 hitter,
the experts who are only marginally better than the competition rise
to the top.
The same holds true for hundreds of thousands of fantasy league champions
across America. The guy who busted his hump digging up information
during the spring was probably only marginally better than the guy
who ripped out a magazine cheat sheet the day before the draft. But
that marginal difference probably won a championship for the guy who
started thinking about his draft in late-April. Remember, football
is a game of inches -- especially fantasy football.
70%
Luck, 30% Skill
Let me draw another comparison. Fantasy football drafting is like
one of my favorite pastimes in its elements: It's like playing poker.
In fact, it is almost exactly like poker in some respects. Poker is
made up of 70% luck and 30% skill. What separates the winners from
losers is that the consistent winners focus on getting better at the
30%, while the losers whine about the 70%.
When Michael Hiban (Draft Sharks writer and business partner) and
I played a lot of poker, we would sniff out the tables where players
asked the dealer for a "deck change" because the cards were "cold."
Cards are neither hot nor cold. They are distributed randomly each
hand. If you sit at a poker table long enough, you'll get your share
of "bad" cards, and your share of "good" cards. Such is the nature
of random distribution. When guys asked for a deck change, we knew
we were playing with "fish…" players who focused on the 70% luck,
rather than the 30% skill.
It's much the same in fantasy football. Everybody will get their share
of bad luck if they play long enough. Just look at last year. Terrell
Davis limps through a season, with several game-time scratches! A
long list of WRs blow out their wheels - including Joey Galloway,
Michael Westbrook, Patrick Jeffers, and Germane Crowell. Kurt Warner
and Marshall Faulk miss some games. Brad Johnson slumps without his
top pass-catcher. Marcus Robinson struggles without a decent QB. Duce
Staley misses most of the season after showing a glimpse of greatness…
But you'll also get your share of good luck as evidenced by last year,
as well. Robert Smith stays healthy for 16 games. Veteran CFL QB Jeff
Garcia came out of nowhere and has one of the best seasons in recent
memory, with 35 total TDs and 4,278 yards passing! A low-profile rookie
named Anderson erupts into a top-5 fantasy back. And Ahman Green goes
from Mike Holmgren's doghouse to Mike Sherman's penthouse.
These "luck" players are basically spread out randomly, just like
those poker cards. And, for better or worse, they dominate fantasy
football every year. As I've said, fantasy football is 70% luck. But
the champions pull ahead on 30% skill. So what exactly constitutes
skill?
Before I define "skill" at fantasy drafting, let's start with this
historical premise: The turnover in the top-ten at each position is
greater than 50% each year. That means that if historical trends hold
up for 2001, the majority of the players at each position who were
in the top-10 last year will be replaced by other players this year.
In 2000, there were only 23 players out of a possible 50 at QB, RB,
WR, TE, and K who repeated their top-10 performance from 1999 (only
46%). In 1999, there were only 15 players out of a possible 50 who
repeated their top-10 performance from 1998 (only 30%). And in 1998,
there were were 22 players out of 50 who repeated their top-10 performance
from 1997 (only 44%). This historical trend extends well into the
early 90s.
Here comes the skill -- the 30% that will put you over the top. Your
job is to identify roughly 20 or 25 players who will either slide
out of the top-10 at their position, or climb into the top-10. (To
tell you exactly HOW to identify these players is another article).
If you're close to being right on a majority of those handful of players,
you're going to have a championship season.
This is going to take a lot of thinking, a lot of research, and a
lot of guts. You're going to look at your rankings and start to think
to yourself, "This just doesn't FEEL right." Don't back down. Look
at last year. Would it have felt right last year to drop Brad Johnson
out of the top-10? Would it have felt right to put Torry Holt ahead
of Isaac Bruce? Would it have felt right to project Ricky Williams
as a top-tier RB? Remember, fewer than 50% of the players who were
top-10 performers last season are going to repeat in 2001. Please
read that sentence again!… fewer than 50% of the players who were
top-10 performers last season are going to repeat in 2001.
Outside The Box
Are you going to be wrong with some of your bolder picks? Absolutely.
In fact, you'll be wrong with quite a few. But don't worry. And do
NOT be afraid to fail. Want to hear some bad picks from last season?
Check out these positional rankings from last summer: Marcus Robinson
at number 3 and Randy Moss at number 11 (yes, at 11!)… Steve Beuerlein
at number 2, one notch above Kurt Warner at 3… Pretty bad, huh? Well,
those were some of the bolder picks of KFFL
- the 2000 Champion of the Pro Forecast Masters Poll. But KFFL also
boldly put Terrell Davis at 16 and Jamal Anderson at 20, while Charlie
Garner made it at the tenth spot - higher than any other expert. They
even had Elvis Grbac in their top-10. In the end, they made a lot
of wrong predictions (as did everyone else). But they were bold enough
to be marginally better than all the other expert - and they walked
away the with 2000 Masters Poll Championship. Well done!
When Draft Sharks won it
in 1999, the editors at Pro Forecast offered this preface: "Draft
Sharks really mixed things up this year by having more contradictory
picks than the rest of the masters." Amen. Always bear this in mind:
You don't have to have a perfect draft (or rank players perfectly).
You only have to do it better -- marginally better -- than everyone
else. And in order to do that, you have to have some contrarian thinking.
My principle complaint of many other fantasy experts is that they
tend to play "last-year's-stats-equal-this-year's-stats" when creating
their rankings. In part, I understand why. There is a paradox in doing
player projections that most experts do not bear in mind. On the one
hand, players will roughly produce as they have in past years - unless
factors (coaching changes, personnel moves, etc.) impact that production
either upward or downward. On the other hand, we know that fewer than
HALF the players who finish in the top-ten at each position will repeat
that performance the following year.
Fantasy expert, Tony "The Prognositcator" Holm, offered this observation
in a column from last summer:
"...A magazine, or on-line draft list, often times is nothing
more than a politically correct public opinion poll. There are exceptions
of course and some of the publishers/webmasters out here really do
post what they think and do not give a snot about what other people
think. Be sure to scrutinize each list for players that look a little
odd where they are placed. If you find one, you will have found yourself
someone who truly thinks for themselves..."
Amen, brother Progno! So why don't many experts follow the historical
trends and really try to predict the major shifts in players instead
of playing it safe? There are probably several reasons.
The first is that some guys just don't want to spend time digging.
It takes a lot of research and a lot of thought to do projections,
and most "experts" do not work at fantasy football as their full-time
job.
Secondly, nobody wants to stand out from the crowd and be WRONG. Check
out this gem from this last year's Fantasy Football Index. They wrote:
"Lenny Pappano made a pick that will make him look like either a genius
or a jackass at the end of the season, ranking Kerry Collins 6th."
Ouch. As I said at the outset, I'm likely to be wrong in most of my
predictions (just like that .320 hitter), but the editors of one of
America's most widely read fantasy football magazine put me in their
cross-hairs for daring to think outside the box. Most people would
rather pass on that distinction.
Tony Holm offers a third reason, and I'm inclined to agree with him.
Money. Holm says that he had a conversation with the owner of a major
fantasy site a few years ago in which the owner admitted to following
the herd because he didn't want to chase away customers. "I can't
afford to look like an idiot because who will come to my site if I
don't have Favre #1 like everyone else?" Holm quotes the site owner
as asking. "How can I call myself an expert if I don't have the consensus
#1 pick on top of my draft board?" Holm concludes with this: "Think
I'm kidding? Think again. Welcome to the way the world works, people."
Nope -- I don't think you're kidding, Tony. Offering an analysis and
opinion that differs with the consensus of experts will indeed cost
you money. I have absolutely no doubt of it. Here is one of scores
of similar emails that I received last year alone.
"...You
call yourselves professionals? In your preliminary rankings you list
Duce Staley at 7? The Eagles still suck and can't move the ball. …
You have Brad Johnson as the 10th rated QB? Please explain… You really
shouldn't post such things especially if you expect to get paying
customer's (sic). I can get better advice from the drive thru window
at Wendy's than what you offer...."
… Plus you can get a
Frosty, dude…
Want to win a Championship this year (or at least finish in the money)?
Here's my two cents. You should be looking for a diversity of opinion.
Search out experts who are doing research and thinking for themselves
(they still exist) -- even if you don't agree with their analysis.
(or as Tony Holm said: "… scrutinize each list for players that look
a little odd where they are placed. If you find one, you will have
found yourself someone who truly thinks for themselves.") Ultimately,
a diversity of viewpoints will help you sharpen your own opinion of
players. And those opinions should differ rather frequently with the
consensus of experts.
Reasoning Behind The Rankings
My other suggestion is to make sure that someone's rankings come with
a detailed rationale. I've seen way too many rankings on the internet
with NO player profiles. Just a list. You should hold these experts
to the same standard that your elementary school math teacher used.
You have to "show your work" to get credit. Only by reading an expert's
rationale will you be able to sharpen your own opinions. Check out
the magazine your holding. Every player (even the 30-something Kicker)
gets a nice beefy write-up. That means that the editors broke a sweat
in putting together their rankings. Look for that same depth of research
when assessing any expert's work.
I can make you one guarantee. If you follow the consensus of experts
this year, you will not win a championship. No way. You're a lamb
being led to the slaughter, not a shark in search of prey. As for
me, I would rather have Draft Sharks go broke than to bend our opinions
to be more palatable to potential subscribers. Frankly, far too many
experts are doing a disservice to our hobby by regurgitating last-year's-stats-equal-this-year's-stats.
