Value-Based Draft Strategies
by Emil Kadlec (5/2002)
Mind Over Matter - Gaining A Better Perspective
There are two main ingredients to a successful draft: 1) A solid cheat sheet; and 2) A solid drafting strategy. Using one without taking advantage of the other isn't an option.
That hasn't always been the case. …
Cheat sheets were all the rage back in the early '90s. So much so that drafting strategies were largely overlooked. As a result, only a small percentage of owners actually sat down at the table with some kind of overall plan, meaning the vast majority of owners spent draft day flying by the seat of their pants. These poor drafters made it easier for the strong to thrive!
Things have changed considerably since then.
These days, everybody has a strategy. And even though some of the old, "always take two running backs in the first three rounds" conventions still tend to hold true, we know have models that validate them. More importantly, we're able to offer owners a better understanding of when those old conventions should be avoided.
As owners become better educated, draft day "steals" become fewer and farther between - a phenomenon that forces owners hoping to maintain or gain an edge over the competition to draft with more precision than ever before.
All of which brings us to the theory known as Value-Based Drafting (VBD). Although it's only one of the many theories currently making the rounds, we believe it offers owners the best perspective on the difficult task of drafting.
Determining the true origin of the VBD is difficult, but Joe Bryant of footballguys.com certainly played a major role in bringing the concepts to life. And despite a fast-growing list of variations and spin-offs, the basic principles behind the Value-Based Drafting - many of which were initially developed by Rotisserie Baseball gurus - remain unchanged.
The Theory Of Relativity
One very important part of a successful draft is determining the relative value between skill positions. Is it time to take a WR? How long can I - and should I -- wait before selecting my starting QB? How high should the first TE go?
VBD disciples answer those questions by assigning relative values - commonly referred to as X numbers - to every player in the draft pool. The X number is used to adjust the Fantasy scoring value (FSV) for each player based on your league's configuration.
For example - let's say your league's rules specify a starting lineup consisting of 1 QB, 2 RBs, 3 WRs, 1 TE & 1 PK. Let's further assume it's a 12-team league. Of course, each team needs a starting QB, meaning 12 NFL QBs will start each week's games -- just like it will take 36 NFL wideouts (12*3) to meet your league's requirements for starters and so on.
X numbers normalize the bottom starting players at each skill position. This is done by subtracting the last "starting" player's FSV from all players of that skill position. Therefore, the 12th QB has an X number of zero and the 36th WR has an X number of zero and so on. The following table provides an example of QBs normalized to the 12th player who is Brett Favre. Accentually you subtract Favre's FSV from all QBs to get their X number. The same is shown for the WRs.
As you can tell, the top WR (Randy Moss) fits between QBs Warner and Manning in an overall list even though his FSV puts him far below the 15th best QB. Why is this? The answer is basically the league structure. Let us take this for an example. Let us assume of your nine starting players each week there were no limits on how many at each skill position. If that were the case you would draft directly according to Fantasy scoring value (FSV) making quarterbacks very popular first-round picks and strategy would no longer be a factor.
But the league structure in our first example limits each team to just one starting QB. But you do need three receivers, which is why wideouts normalize to a higher value.
Einstein would be proud. … Maybe.
| QUARTERBACKS |
| Rank || Player || Bye || FSV/X# |
| 1 || Culpepper, MIN || 9 || 365/133 |
| 2 || Warner, STL || 9 || 350/118 |
| 3 || Manning, IND || 5 || 324/92 |
| 4 || Garcia, SF || 7 || 320/88 |
| 5 || McNabb, PHI || 6 || 303/71 |
| 6 || Gannon, OAK || 5 || 274/42 |
| 7 || Johnson, BUF || 6 || 253/21 |
| 8 || Green, KC || 11 || 246/14 |
| 9 || Grbac, BAL || 14 || 245/13 |
| 10 || Griese, DEN || 16 || 244/12 |
| 11 || Brunell, JAX || 6 || 237/5 |
| 12 || Favre, GB || 8 || 232/0 |
| 13 || Testaverde, NYJ || 12 || 232/0 |
| 14 || Brooks, NO || 3 || 226/-6 |
| 15 || Johnson, TB || 3 || 219/-13 ||
| WIDE RECEIVERS |
| Rank || Player || Bye || FSV/X# |
| 1 || Moss, MIN || 9 || 186/104 |
| 2 || Harrison, IND || 5 || 177/95 |
| 3 || Owens, SF || 7 || 174/92 |
| 4 || Holt, STL || 9 || 174/92 |
| 5 || Moulds, BUF || 6 || 165/83 |
| 6 || Westbrook, WAS || 10 || 154/72 |
| 7 || Smith, DEN || 16 || 153/71 |
| 8 || McCaffrey, DEN || 16 || 150/68 |
| 9 || Bruce, STL || 9 || 147/65 |
| 10 || Carter, MIN || 9 || 144/62 |
| 11 || Horn, NO || 3 || 140/58 |
| 12 || Alexander, KC || 11 || 137/55 |
| 13 || Freeman, GB || 8 || 135/53 |
| 14 || Johnson, TB || 3 || 134/52 |
| 15 || Boston, AZ || 1 || 134/52 |
| 16 || Brown, OAK || 5 || 130/48 |
| 17 || Smith, JAX || 6 || 130/48 |
| 18 || Crowell, DET || 4 || 126/44 |
| 19 || McCardell, JAX || 6 || 124/42 |
| 20 || Chrebet, NYJ || 12 || 118/36 |
| 21 || Taylor, BAL || 14 || 118/36 |
| 22 || Muhammad, CAR || 15 || 116/34 |
| 23 || Warrick, CIN || 9 || 116/34 |
| 24 || Thrash, PHI || 6 || 116/34 |
| 25 || Toomer, NYG || 13 || 114/32 |
| 26 || Robinson, CHI || 4 || 114/32 |
| 27 || Scott, CIN || 9 || 111/29 |
| 28 || Jeffers, CAR || 15 || 109/27 |
| 29 || Terrell, CHI || 4 || 105/23 |
| 30 || Galloway, DAL || 5 || 102/20 |
| 31 || Glenn, NE || 17 || 98/16 |
| 32 || Mathis, ATL || 8 || 96/14 |
| 33 || Mason, TEN || 4 || 96/14 |
| 34 || Gadsden, MIA || 7 || 96/14 |
| 35 || Hilliard, NYG || 13 || 95/13 |
| 36 || Johnson, CLE || 8 || 82/0 |
| 37 || McKnight, MIA || 7 || 82/0 |
| 38 || Jackson, SEA || 7 || 82/0 |
| 39 || Morris, KC || 11 || 80/-2 |
| 40 || Morton, DET || 4 || 79/-3 |
Where To Normalize?/Where Does Relativity Lie?
Before you get too excited, I'll remind you this isn't an exact science. In fact, instead of using the starting line to determine their normalization point - as we did in the above example - some owners use roster size.
Which means, if most managers carried four running backs on their roster, they would use 48 (4*12) as their normalization point at that position and so on. Other owners take a middle of the road approach somewhere in between. I prefer using the starting lineup to set my normalization points.
Without trying to sound too analytical, these different normalization points will usually result in different overall lists. Hence there is typically some non-linearity in the cheat sheet and real stats. If you use the roster limits as your normalization point significant shifts will occur with respect to using a starting lineup normalization point. Which one is right? I believe the starting lineup normalization is "better" because it places more emphasis on the more valuable players - your starting ones!
Potential Skill Position Saturation (SPS)
Each of us has a preference for a certain skill position when we draft. A typical tendency is to favor RBs. This falls in line with what I call Skill Position Saturation (SPS) and is why a RB happy drafter is usually successful.
SPS can affect your draft and can be determined quickly. Let's say your league has 16 managers and you have to start 2 RBs each week. That means there will be 32 RBs in your leagues starting lineup and only 32 starting in the NFL (not counting fullbacks) each week. When the number of RBs needed to fill all the leagues starting lineups is so close to the available starting players, SPS occurs. When SPS is present, the VBD theories tend to fall through. Typically runs will occur on the SPS skill position and rightfully so. This is why it is a good idea to keep the Fantasy starting players needed to NFL starting players available around a .8 ratio when setting up your league. It is a good idea to run this simple calculation on each skill position (except PK & Def. team) before setting up your drafting strategy.
This saturation can cause drafting strategy to fail in the early rounds where getting your RBs become somewhat of a free for all. If a league has 14 or more (16) owners, I suggest a starting lineup of 1 QB, 1 RB, 3 WR along with a "flex" player who can be a RB, WR or TE. This reduces the saturation point of RBs and yet still allows each owner the opportunity to start two RBs!
Flex Players/Twist Of Fate
Many leagues incorporate a flex positions typically of one player that can be a RB, WR or TE. However, it makes us have to think a little harder on draft day. If we added this flex player to our main example we could now start one of these three examples: 3 RBs, 3 WRs & 1 TE or 2 RBs, 4 WRs & 1 TE or 2 RBs, 3 WRs & 2 TEs.
Two things occur. First, it gives you some flexibility on draft day. Secondly your have a dilemma as to where to put your normalization points. I typically try and get a feel for the other owners in the league as to what they usually lean towards. Most managers are RB happy and in this cast you could pick a normalization of: RB 2.7*12 (32), WR 2.3*12 (27.5) & TE 1*12 (12). This creates a saturation (SPS) point on RBs and leaves WRs with lower pressure during the draft. With this setup you are predicting an early run on RBs and your X numbers will give you the forecasting you need to draft well. In this case I recommend looking closely at the worst-case saturation points. First, let's take a starting lineup of 2 RBs and 4 WRs in our 12 owner league. There is a saturation points of 24/32 = .67 for RBs and 48/64 = .75 for WRs. Both of these are well below 1.0, which is good. So in my opinion you could go into this draft assuming you are going to have a starting lineup of 2 RB and 4 WRs and be more successful then a 3 RB and 3 WR setup. While most owners are getting their third RB you will be picking up an undervalued WR or getting an excellent QB!
One way to help minimize this saturation problem is to find ways to have more viable RBs. The World Championship of Fantasy Football (wcoff.com) has incorporated one point per catch for all players in their rules. This brings in a few FBs and third down backs into the picture and reduces the stress on starting RBs. It also gives WRs a little boost since there SPS is much lower then the RBs.
Draft Day Fundamentals -- Looking Ahead/ Tier Analysis
A key factor in a draft strategy is what most people call tiering. As each round goes by, it is important to look ahead to not only your next pick but the following one also. Let's say it is your pick and you also have a pick in 6 more picks. You haven't taken a wideout yet and you have already taken a quarterback and a running back. The first thought is to take WR1 whom has the largest X#. However, there is a large wideout tier of six players and only a one player tier at the running back position. Hence, you take RB1 (97) and the worst you can do with your next pick is WR6 (96). That is a total X# of 193. If you take WR1 (100) you probably won't get a shot at RB1 (97) but end up with RB2 (89) totaling 189.
This makes sense unless you have already filled your starting RB quota and need two WRs to fill. Then you may want to take WR1 (100) and hope you get at least WR6 (96) totaling 196.
Tiering is not always straight forward, but there are usually good opportunities during a draft to better choices.
| Remaining QBs |
| || X# |
| QB1 || 70 |
| QB2 || 68 |
| QBs || 65 |
| Remaining RBs
| || X# |
| RB1 || 97 |
| RB2 || 89 |
| RB3 || 88 |
| RB4 || 87 |
| RB5 || 83 |
| Remaining WRs |
| || X# |
| WR1 || 100 |
| WR2 || 99 |
| WR3 || 99 |
| WR4 || 98 |
| WR5 || 96 |
| WR6 || 96 |
| WR7 || 84 |
Like tiering, watching your opponent's rosters during the draft can be very useful. I find it most helpful when drafting quarterbacks, tight ends, kickers and team defenses. These positions typically have only one player in a starting lineup. For example, let's say you have the next pick and have a pick in 6 more picks. You haven't taken a QB yet and pickings are getting slim. But there is a very short tier at RB and you have to decide if an adequate QB will be around in 6 picks if you take an RB now. This is a 12 team league and 8 managers have already taken one QB each. A quick examination of the six picks between your two is need to determine how many of these managers still haven't drafted a QB. Depending on the answer you may be able to draft the RB and still get your QB.
Simple Theories For Drafting Place-kickers & Team Defenses
One parameter that is not implemented into most VBD calculations is probabilities - or should I say consistencies. Some skill positions are easier to predict then others. Most would agree that place-kickers and team defenses are two of the hardest to forecast each year. Hence it can be argued that X numbers for these skill positions could be scaled down a little to reflect this lesser probability. The effect of this would drop these players down on the overall cheat sheet. Most drafters do this in there head by ignoring X numbers for place-kickers and team defenses and waiting to take them until late in the draft.
Not a bad move in my book and I have an even easier way of going about it. Wait until 5 to 8 kickers have been taken and then strike. Pretty simple isn't it! And the same goes for Team defenses. One never wants to take the first kicker nor the first team defense. The only way I would violate this theory is if it was the last two of three rounds of the draft and less then 5 kickers had been taken.
Bye Week Strategy
I personally believe bye week draft strategy can be overworked. There are just a few situations that can really make a difference.
The first one involves skill positions that have only one starter - like QBs, TEs, PKs and team defenses. In these cases it is important to make sure your backup has a different bye week. The quarterback is the most important, followed by the tight end. If your starting place kicker or team defense doesn't have a bye week until say after week 6, I wouldn't be as concerned about the backup's bye week. In these two positions, there is a good likelihood you will have cut your backup or starter and selected another via free agency by then anyway. If there is no way to acquire new players in your league, well it is obvious to get a different bye week backup.
Other then the situations I described above, I recommended ignoring the bye week. I believe in taking the best players available given the VBD principles. If you end up with a very large number of players on your roster with the same bye week, then you'll have one rough week and the rest aren't so bad. Now that the NFL has an even number of teams again, you don't have to worry about bye weeks interfering with your Fantasy playoffs either!
There can be very specific situations during a draft that trading down can be beneficial. The simplest situation is when there is a nice sized tier at your skill position of greatest need. If other skill positions just emptied out a tier and there is a significant drop off of talent at these positions then trading down should be considered. One persons junk is another ones treasure, so you typically can find someone to trade with. When the situations is right, trading down is one of my favorite draft principles to achieve.
Remember, the object of the draft is to fill your roster with quality player that can fill your starting lineup evenly. If you have two top ten QBs and only one good RB you may not do as well as someone with one top ten QB and two good RBs (assuming a normal starting lineup).
We also provide a FREE custom cheat sheet tool on the web that will take your league structure and league scoring system and build your cheat sheet for you. You can find our CheatSheet Live tool at cs.footballdiehards.com.
Our latest Cheat Sheets updated constantly through September.
class="headline" onClick="expanddiv('tr1')" style="cursor:hand; cursor:pointer" onmouseover="window.status=' ';return true">Packers Hoping Aaron Rodgers Returns Week 15
(10/20 6:53 PM PT)
class="headline" onClick="expanddiv('tr2')" style="cursor:hand; cursor:pointer" onmouseover="window.status=' ';return true">Marshawn Lynch Suspended One Game
(10/20 6:25 PM PT)
class="headline" onClick="expanddiv('tr3')" style="cursor:hand; cursor:pointer" onmouseover="window.status=' ';return true">Andrew Luck Will Not Practice This Week
(10/18 3:12 PM PT)
class="headline" onClick="expanddiv('tr4')" style="cursor:hand; cursor:pointer" onmouseover="window.status=' ';return true">Kelvin Benjamin Exits Practice With Knee Issue
(10/18 2:12 PM PT)
class="headline" onClick="expanddiv('tr5')" style="cursor:hand; cursor:pointer" onmouseover="window.status=' ';return true">Jameis Winston Does Not Throw During Wednesday's Practice
(10/18 1:06 PM PT)
class="headline" onClick="expanddiv('tr6')" style="cursor:hand; cursor:pointer" onmouseover="window.status=' ';return true">Sterling Shepard Returns To Practice
(10/18 12:56 PM PT)
class="headline" onClick="expanddiv('tr7')" style="cursor:hand; cursor:pointer" onmouseover="window.status=' ';return true">Charcandrick West Will Not Play On Thursday Night
(10/18 12:52 PM PT)
class="headline" onClick="expanddiv('tr8')" style="cursor:hand; cursor:pointer" onmouseover="window.status=' ';return true">Leonard Fournette Will Not Practice Wednesday
(10/18 11:08 AM PT)
class="headline" onClick="expanddiv('tr9')" style="cursor:hand; cursor:pointer" onmouseover="window.status=' ';return true">DeShone Kizer Expected To Start Week 7
(10/18 9:21 AM PT)
class="headline" onClick="expanddiv('tr10')" style="cursor:hand; cursor:pointer" onmouseover="window.status=' ';return true">Ezekiel Elliott Will Play On Sunday
(10/17 6:49 PM PT)