Average Draft Position (ADP)

This article covers Average Draft Position (ADP) and discusses why it is crucial information to have at your disposal on draft day. To understand why ADP is so useful, we'll start with a brief discussion of what it can provide player values.

Player Value
The term value is often the subject of debate amongst fantasy owners:  "Am I getting good value for this draft pick?"  "Which position has the greatest remaining value?"  But does value mean the same thing to everyone and how is it determined?

Webster defines value as an assigned or calculated numerical quantity.  In fantasy football, value refers to the numerical quantity assigned to each player.  This number describes how much better (or worse) one player is than another.  It might be helpful to think of a player's value as their score or rating. 

Rank & the "Cheat Sheet"
It is important to understand that a player's value is not the same thing as their rank.  A player's rank is the spot they occupy in order from the highest to lowest valued player on a cheat sheet.  Unlike value, there is always a player ranked 1, 2, 3, 4, and so on.  It should be noted that some cheat sheets list player rank only and omit their value.  These types of cheat sheets are very limited because they do not convey the difference in value between each player.  For example, what do we know from a cheat sheet that shows Player A is ranked 1st while Player B is ranked 5th?  If there are 10 players ranked, we are left to conclude that Player B is worth half of Player A ((10-5)/10 = 50%).  In reality, this is rarely ever the case. 

A good cheat sheet expresses a non-linear difference between each player so that it conveys how much more valuable one player is than another.  Some cheat sheets attempt to accomplish this by separating players into groups or tiers.  This is a somewhat arbitrary line drawn every few players that attempts to identify when there is a significant drop-off in value.  This is better than nothing, but it tends to be more of a "gut feeling" modification to the basic cheat sheet.  Why stop at drawing a drop-off line every few players?  A good owner needs to know the difference between every player.  What is needed is an impartial method for assessing player value that naturally generates the variable difference between players and expresses how much more valuable one player is than another.

Introducing Average Draft Position (ADP)
Average Draft Position (ADP) is the average overall draft slot that has been used to draft each player in real or mock fantasy drafts.  If Player A has been drafted 1st in five drafts and 4th in five drafts, he has an ADP of 2.5 ((1*5 + 4*5)/10 = 2.5).  Note that the best possible draft slot is 1 and not 0. ADP is the simplest approach for assessing player value because all of the work is done for you by web sites (stadium.com, antsports.com).  A mock drafting web site can provide reliable ADP statistics if the following features exist:

  1. A sufficient number of drafts.  Remember that we are looking for an average.  A greater number of completed drafts by a greater number of different people will produce the most refined results.
  2. Filters for old data.  Pre-season injuries need time to be reflected in ADP results.  Hopefully, enough drafts are being completed so that data can be drawn from the past few days only.
  3. Results for different types of leagues
    1. Scoring system.  Pick a scoring system that resembles your own; otherwise, you may end up under or over-valuing certain positions. 
    2. Positions used.  Mixing drafts of leagues that use the Tight End position with leagues that don't will pollute the results.
    3. Draft depth / roster size.  Are enough rounds completed to create an ADP list deep enough for your league?
    4. Number of teams.  This is not as important as many might think.  It means much more if you are trying to line up the ADP number with your own league's draft slots instead of using ADP data for player values.
  4. No keeper league results.  Keeper league rules vary so much that analyzing draft picks made by teams with keepers is useless even for someone drafting in a keeper league.
  5. No computer picks.  Web sites often employ an automatic selection algorithm when owners fail to make a pick.  Most of the algorithms are not sophisticated enough to simulate human behavior, so these picks should not be considered when calculating the average draft position.

Reasons to use ADP
In the past, fantasy owners have been taught to use ADP as a reality check before announcing a draft pick.  This approach is typically followed by owners who use a separate, more sophisticated method for creating player values.  While it is reasonable to use ADP data to make sure a player is not being chosen earlier than necessary, it is a mistake to ignore the fact that ADP is a viable method for determining player values.  After all, a player's true value on draft day is defined by where he is drafted and not where he is listed on someone's cheat sheet!  Stop... read that last sentence again... it is very important.

To use the earlier example, when reviewing a cheat sheet that shows Player A (ranked 1st) and Player B (ranked 5th) have ADP values of 1.7 and 3.6 respectively, we suddenly understand that Player B is not such a bad catch after all.  In fact, we might be tempted to pass on Player A in order to grab a player at another needed position knowing that we can come back and get Player B without losing too much value.  Assuming that the worst player available has an ADP of 10, we can now see that Player B is only 20% inferior to Player A ((9.3-7.4)/9.3 = 20%).

With more people performing mock drafts and more web sites making real draft results available, the ADP data is becoming much more refined than it was in years past.  Many of the people participating in these drafts have done hours of research.  Take advantage of their diligence!  I'm not advocating that you shirk your responsibilities to do your own research, but if I felt unprepared for a draft or could take only one piece of data with me, it would be ADP.

Arguments against ADP
As mentioned earlier, care needs to be taken not to get ADP data from old draft results, computer generated picks, or leagues with very different scoring/roster rules.  There are also several myths that need to be addressed regarding the efficacy of ADP:

One issue that has risen from failing to recognize that ADP represents value is that software programs are using it to inaccurately predict draft picks.  This is not a criticism of ADP as much as it is of an ill-advised application of the data.  Some programs use ADP to guess which players will be drafted before your next pick.  Unfortunately, they ignore the needs and tendencies of the owners that actually own those picks.  There is no way an owner who has already drafted their defense is going to draft another one just because a defense is sitting atop the ADP list.  Please don't use ADP for this purpose.

Knowing the average draft position (ADP) of a player has long been recognized as an important piece of data to consider when making a draft selection.  However, it is time to consider it as much more than just a sounding board for informed decisions.  ADP information can also be used as a rudimentary player value that, when used by itself or in conjunction with your own player values, can give you a significant advantage on draft day.  The FFTools drafting tools are designed to make the most of ADP data.  Draft Planner has access to daily draft information that you can use to create your own cheat sheet.  You can then import this cheat sheet into Draft Predictor to predict who your competitors are likely to pick. With so many good sources of ADP data available, you'd be doing yourself a disservice not to include it in your draft day regimen. So, check out advanced applications of ADP that use it as player values, but, at the very least, be sure to consult the ADP of a player before announcing your pick.