Disclaimers:

1.)    The following article will be long, boring, and generally not useful to anyone at all.

Draft Shark.  This term is thrown around quite often in today’s world of fantasy football information.  With the instant access to information that the Internet provides, all fantasy owners can be draft sharks.  Every owner has access to more lists, statistics, and analyses than he could possibly have imagined ten years ago.

An owner’s ability to apply the available information to his draft management is the “stepping stone” to a successful season.  It all starts with the draft!  This point can not be overstated.  While it is possible to recover from a poor draft, this is not anyone’s ideal situation (unless that person is especially warped).  Your goal entering a draft should be to get ahead of the game, not fall as far behind as possible.  With that goal in mind, I present you with these three simple words (okay two if you want to be technical):

Value-Based Drafting

Value-Based Drafting, or VBD, is a commonly used tool of draft management.  The basic principle behind VBD is that you should pick the player who has the most value to your fantasy team with every pick in your draft.  Your goal is to maximize the scoring potential of your team by selecting the most valuable player available with every pick.

Value, as it pertains to fantasy players, has numerous meanings.  However, the most valuable pick at any point in a draft should be the player who will score the most points for your team, in your system, in comparison to others at his position.  The most valuable player is not the highest scorer, but the highest scorer relative to the other players drafted at the same position.  Also, when determining which players are the most valuable, you must factor in both the needs of the other owners in your league and the dynamics of the draft.

There are two sample scenarios that I often use to illustrate “value” of draft picks:

#1 Suppose you and I are in a 2-team league with a serpentine draft.  Each of us is going to draft 1 QB, 1 RB, and 1 WR, from a pot of 2 QBs, 2 RBs, and 2 WRs.  Further, suppose that we know exactly what each of the players in the pot will score, and those numbers are:

QB #1 = 74 points scored and QB #2 = 71 points scored

RB #1 = 56 points scored and RB #2 = 53 points scored

WR #1 = 52 points scored and WR #2 = 42 points scored.

Given the first pick, whom would you take in this scenario? Why?

#2 Suppose you and I are in the same draft as above, but I have the #1 pick. 

If I take WR #1, whom will you take with your first two picks?  Why?

The answers to all of these questions should be immediately obvious.

In Scenario #1 you MUST take WR #1 with your first pick.  He is the player with the most value.  If you draft him, you can’t lose.  He outscores the alternative at WR by 10 points, and no other player in this pool can match that.   Even if I get the #1 QB AND the #1 RB, I can not beat you.

In Scenario #2 you should take QB #1 and RB #1.  This should also be obvious, although the reasoning behind the selections might not be so obvious.  Since I have selected my WR already, there is no need for you to spend an early pick on one.  I will not be drafting another WR, so you can get WR #2 at a later point in the draft.  Therefore, you maximize your value by selecting the better scorers and getting the WR later.

Obviously these two scenarios are very basic, but they can be applied to an entire draft.  Actually, they should be applied to all drafts.  The combination of these two basic principles is the foundation for Value-Based Drafting: In every round you should draft the player who has the most value to your team among players still available in that round.  By drafting a player even one round too early or too late, you are reducing the value of your draft, and, therefore, reducing the value of your team.

I am not going to discuss in detail what I consider to be the obvious weaknesses of the commonly used VBD methods: the Joe Bryant X-Value system and the tiered-reduction system.  Instead, I will offer you an alternative to these methods.  My system is based upon these two common systems, but also covers for their weaknesses.

Value-Based Drafting – The War Room System.

1. Ignore all Kickers and Defenses.  These are creatures unto themselves, and they should be treated independently.

2. Project season statistics for all of the players you expect to be drafted.  You must project ALL meaningful statistics.  Meaningful statistics are all statistics that have been assigned point values in your scoring system.  Remember that the players you consider worthy of being drafted are not necessarily the same as those that other owners deem worthy.  Your lists of players should include at least as many players as the number of picks that will be made.

  **My own personal thoughts on projections: Players tend to perform similarly from year-to-year, based on talent, UNLESS there are NEW outside factors which will enhance/decrease their performance. Some of these include new team, new coach and system, new role (i.e. now a starter when player was a backup, etc) with team, injury, returning from injury, etc.

There are always exceptions, but generally the players will be pretty consistent from year-to-year, unless there are these external influences**

**My advice on projections: Find 3 web-sites/magazines that you trust and analyze their projections.  Their statistical projections will probably generally be similar, and they will often give reasoning behind their projections.  Use them as guides, remembering that they are merely the GUESS of someone else, and alter them based on YOUR opinions of what will occur.  Remember that your projections MUST reflect your beliefs for the upcoming system.  If they don’t, modify them as needed**

3. Now take your projected statistics for each player and enter them into your league’s scoring system to determine how many points the player will score this season, according to your projections.  For example, if you project Kurt Warner to pass for 3,900 yards and 37 TDs with 15 INTs, and your league gives 3 points for a Pass TD, 1 point for every 25 passing yards, and –1 for every INT, Warner will score 252 points according to your projections.

4. Sort all of the players, BY POSITION, from highest number of fantasy points scored to lowest.  Keep the players on separate lists by position.

Ex. Your QB list might look like this: 1. Warner 252 points 2. Favre 247 points

3. Beuerlein 244 points 4. Brad Johnson 229 points and so on…

5. Determine the number of starters in the league at each position.  If you have a 12-team league that starts 1 QB and 2 RBs, there will be 12 starting QBs and 24 starting RBs in your league.  If your league allows variations, use the maximum allowable for all positions.

6. Average the fantasy points projected for the two worst starters at each position in your league.  Using the above league example, this would be the #11 and #12 QBs and the #23 and #24 RBs.

Ex. Assume your #11 scorer at QB is projected for 142 points and the #12 QB is 138 points.  The average of these two would be 140 points.  Also, assume that the #23 RB is projected to score 77 points and the #24 RB is projected to score 59 points.  The average of the worst starting RBs is then 68.

7. Repeat option 6 above using the middle two starters at each position.  In the same sample league, you would then average the projected fantasy points for the #6 and #7 QBs and the #12 and #13 RBs.

Ex. Repeat step #6 with the middle-of-the-road starters.  So, if the #6 QB is projected to score 191 points and the #7 is projected to score 188 points, the average of the average (or middle-of-the-road) QBs would be 189.5.  Also, assuming that the #12 RB is projected to score 167 points, and the #13 RB is projected to score 158 points, the average of the two would be 162.5 points.

8. Average the numbers you calculated in options 6 and 7 above, for every position.  This will be your value modifier.

Ex. Using the numbers from 6 and 7 above, the Value Modifier for QBs would be the average of 140 and 189.5, or 164.75, and the Value Modifier for RBs would be the average of 68 and 162.5, or 115.25.

9. To calculate the Value of each player, subtract the value modifier for his position from his projected fantasy points scored for the season.  Place this resulting number on your list in a column labeled Value.  At this point, you have completed a variation of the X-Value system of VBD.

Ex. Still using the numbers from above and mythical point projections, here would be a Sample of your Value Charts:

QB #1 K Warner 252 points Value = 87.25 (252 – 164.75)

QB #2 B Favre 247 points Value = 82.25 (247 – 164.75)

QB #3 S Beuerlein 244 points = 79.25 (244 – 164.75)

QB #12 Rob Johnson 138 points Value = -26.75 (138 – 164.75)

RB #1 E James 250 points Value = 134.75 (250 – 115.25)

RB #2 F Taylor 239 points Value = 123.75 (239-115.25)

RB #3 M Faulk 238 points Value = 122.75 (238 – 115.25)

RB #24 JJ Johnson 59 points Value = -56.25 (59 – 115.25)

**REMEMBER, track positions on SEPARATE lists at all times**

10. Next players must be grouped into positional tiers.  Divide the projected fantasy points for every player at a position by the number of projected fantasy points for the top player at that position, and multiply the result by 100 to yield a percentage (round off to whole percent).

Ex. Using the Values listed in 9 above, group those 8 players into positional tiers.

QB #1 K Warner 252 points Value = 87.25 Tier % = 100 (252 / 252 * 100)

QB #2 B Favre 247 points Value = 82.25 Tier % = 98 (247 / 252 * 100)

QB #3 S Beuerlein 244 points = 79.25 Tier % = 97 (244 / 252 * 100)

QB #12 Rob Johnson 138 points Value = -26.75 Tier % = 55 (138 / 252 * 100)

RB #1 E James 250 points Value = 134.75 Tier % = 100 (250 / 250 * 100)

RB #2 F Taylor 239 points Value = 123.75 Tier % = 96 (239 / 250 * 100)

RB #3 M Faulk 238 points Value = 122.75 Tier % = 95 (238 / 250 * 100)

RB #24 JJ Johnson 59 points Value = -56.25 Tier % = 24 (59 / 250 * 100)

11. The top player on each list should be 100% and the percentage of points scored will decrease as you move down the list.  Draw breaks at every 10% down the list.  The top break will be right under the top player at each position.  The second break should be drawn between the last player at 90% or above and the first player below 90%.  The final break will be below the last player at 10% or above.

**NOTE: If the breaks are done in bold distinguishable colors, you will easily be able to identify your tiers**

Ex. Using the chart from 10 above, I have inserted spaces where there are breaks.  Obviously a full listing would have more breaks (1 per every 10%) and more players on each level (i.e. grouping between breaks).

QB #1 K Warner 252 points Value = 87.25 Tier % = 100 (252 / 252 * 100)

QB #2 B Favre 247 points Value = 82.25 Tier % = 98 (247 / 252 * 100)

QB #3 S Beuerlein 244 points = 79.25 Tier % = 97 (244 / 252 * 100)

QB #12 Rob Johnson 138 points Value = -26.75 Tier % = 55 (138 / 252 * 100)

RB #1 E James 250 points Value = 134.75 Tier % = 100 (250 / 250 * 100)

RB #2 F Taylor 239 points Value = 123.75 Tier % = 96 (239 / 250 * 100)

RB #3 M Faulk 238 points Value = 122.75 Tier % = 95 (238 / 250 * 100)

RB #24 JJ Johnson 59 points Value = -56.25 Tier % = 24 (59 / 250 * 100)

12. Now you have a tiered list that displays a numeric Value for every player, by position.  You have accounted for points scored, but not consistency.  To account for consistency, utilize the previous season’s statistics.  Calculate the average points per game scored by the player during the previous system (most leagues have this data available without the need for calculations). Then count the number of regular season games in which that player scored at least 85% of his average points per game.  Write this number of games on your list under the heading of Consistency Factor.  A player who was both healthy and consistent will have 16, as this is the maximum value possible.  The higher this value is the better.

Ex. Using the same players, I will use a mythical number of games and average points in 1999.  Assume this chart is accurate for 1999:

Kurt Warner 287 points, 22.1 points per game, 12 games at least 18.79 points (18.79 is 85% of 22.1)

Brett Favre 229 points, 17.61 points per game, 7 games at least 14.96 points (17.61 * .85 = 14.96)

Steve Beuerlein 259 points, 19.9 ppg, 9 games at least 16.92 points

Rob Johnson 28 points, 14 ppg (only 2 games played), 1 game at least 11.9 points

Edgerrin James 301 points, 23.15 ppg, 10 games at least 19.67 points

Fred Taylor 156 points, 17. 3 ppg, 5 games at least 14.7 points

Marshall Faulk 274 points, 21.08 ppg, 11 games at least 17.92 points

JJ Johnson 71 points, 7.9 ppg (9 games), 5 games at least 6.72 points

So your draft charts (using only these 8 players) would look like this:

13. In order to finalize your positional draft lists, you should add a Comments column.  I add various comments in this column.  Some that I would recommend are “I have him higher rated than most” or vice-versa, “I know [insert owner name] is very high on him”, “he is recovering from injury”, etc.  Anything you add here is just meant to serve as reminders during the fast and furious draft.

14. The final step in pre-draft preparation is to create a “Need list” for every team.  Create a chart that lists all team names and has all of the positions that you draft in the league.  As a team drafts a position, cross it off of the need list.  This will help you track the potential for players to slip to you based on the lack of need of your opponents.

Ex. Need List for Team X: 

QB1 QB2 RB1 RB2 RB3 WR1 WR2 WR3 WR4 K1 K2 Def1 Def2

Once Team X picks a player at one of these positions, merely cross the position off the list.

Now that your preparation is complete, you can sit back and relax until draft day.  Once that wondrous day arrives, you will be able to use your lists to maximize the value of every selection you make.

During the draft, you should use the “Need List” to track the positions selected by every team in the league.  Once you are on the clock, you use your positional lists to make the value pick.  To use the positional lists, write down the Value of the top player still available at each position.  Using your “Need List” and tiers, try to determine which players at each position will be drafted between your current pick and your next selection.  Then, subtract the Value of the player top player on each list that you anticipate being available at your next pick from the Value of the top player at his position that is currently available.  This will yield a Value Differential between the top players available to you in this round and your anticipated top players available to you in the next round.  You then merely pick the player with the HIGHEST Value Differential, although you must pay attention to the Consistency Factor when two players are very similar.

Ex. Assume you are picking 3^rd in the first round and there are no keepers.  Your top QB still available is Warner, with a Value of 87.25.  Your top RB available is Faulk, with a Value of 122.75, and your top WR available is Moss, with a Value of 91.  Assuming there will be 18 picks made before you select again, you must use determine what players will be selected before you pick again.  If, on your list, there are a large number of QBs and WRs in the upper tiers that you defined, you can estimate that more RBs will be selected early, then there will be a run on these positions.  Try to make accurate assessments of which players should be gone before your next pick.  Assuming you expect the next 9 RBs, next 5 WRs, and next 4 QBs on your list to be off the board when you pick in Round 2, you must look at the Value of the #10 RB, the #6 WR, and the #5 QB.  Suppose Duce Staley is the #10 RB with a Value of 77, Steve McNair is the #5 QB, with a Value of 59.25, and Eric Moulds is your #6 WR with a Value of 78.5.  So, subtracting each of these Values from the Values of the relative “best” available players at each position will yield the following Value Differentials: Warner has a Value Differential (VD) of 28 (87.25 – 59.25), Faulk has a VD of  45.75 (122.75 - 77), and Moss has a VD of 12.5 (91 – 78.5).   Therefore, Faulk has the highest VD and should be the pick, as long as there is not a player with a similar VD and a much higher Consistency Factor (see Below).

FINAL NOTE: Use of the Consistency Factor- One of the final additions that you made to your list was the Consistency Factor.  I only use this to separate two players that are very similar in Value Differential.  If two players have Value Differentials that are practically identical (say within 8 points, or less than 1/2 point a game), I will ALWAYS take the player with the higher Consistency Factor.  I will also consider extending the differential in Values up to 10 or 12 if there is a LARGE discrepancy in Consistency Factors among the top couple of players available at a position. 

Ex. Using the chart in 12 above, even though Fred Taylor has a higher Value, I would take Marshall Faulk over him if they were the two most valuable players on the board when I was set to pick.  Give me that Consistency Factor of 11 over the 5 any day!

Now you have everything you need to be another shark in the waters…

I would like to thank the legendary Bill James, Joe Bryant, and Greg Kellog for laying the foundations for this system.