In Absentia: An Autopick/Draft-By-Proxy Primer

  Guest Submission posted 8/25 by Colin McGinnis, Exclusive to Footballguys.com

Let's face it: life happens outside fantasy football. While draft day may be the most wonderful time of the year, a conflict may pop up unexpectedly. You might have a work emergency, a surprise family visit, or a first date with your favorite supermodel. Whatever the reason, you need someone/something on your behalf to draft for you, and your agent has done zero preparation. Fear not: you can still dominate your draft without even being there.

Of course, drafting live is the best way to go. As you proceed through a live draft, you'll be able to draft reserve players who complement your starters and ensure bye week coverage. You can wait on kickers and defense until a certain number are taken. You can monitor other teams' drafting trends, notice their positional needs, and choose accordingly. Being there gives you more knowledge and, as a result, more competitive advantage. While you can't leverage any of this knowledge by autopicking, your list can still be competitive if you've done your homework.

Start with the Basics

Whether you're drafting live or not, draft preparation is fundamental to success. Initial preparation for an autopick list will be nearly identical to the work for any other league. As you create any draft list, you will be considering the same things:

How many players at each position are draft-worthy?

If you're in a Yahoo league, you'll notice that your autopick list has a finite limit based on your league size and settings. For a 12-team league with 18 roster spots, you'll be able to enter (12 * 18) 216 players on the list. Even if your league doesn't have this limit, you should have a good idea of the list's general makeup by position. There's no point in figuring out the 80th best WR if your league doesn't have room for him.

Your league might have position-specific roster limits (2 QB + 4 RB + 6 WR + 2 TE + 2 PK + 2 DEF = 18 slots), which will make this task really easy. Simply multiply the team roster limit by the number of teams. With 12 teams, your list will have 24 QBs, 48 RBs, and so on. Your roster limits might not be that rigid, so you have some room to experiment. Mock drafts are an excellent tool to explore these possibilities. Using Draft Dominator's default settings, a mock draft may give you something closer to 24 QBs, 60 RBs, 71 WRs, 24 TEs, 15 PKs, and 22 Defs. The initial results may work for you, or you may tweak it from there.

Who are the top # players at each position? Top players overall?

Now that you've determined your list's makeup, you need to figure out the players you actually want. Use whatever research methods you wish (expert rankings, projections, VBD, et al). When you're done, you'll have a ranking list for each position along with an overall list. You might be tempted to use the overall list as your autopick list and be done with it, but you'd definitely be shortchanging yourself.

When will each player be taken?

The key to success in any draft is to get the best available player at the lowest draft pick possible. There is no point in getting the player you want at your current pick if you can get him at the next one. There's no accounting for this in a simple overall ranking, so you will need to figure out the expected draft pick for each player. Mock draft results and Average Draft Pick (ADP) values are the best tools available for this research. While your values may not reflect an average per se, it's common to still refer to this value as ADP. Once you have assigned the ADP value to every player on your list, you're ready to get started.

Using Tiers (No Mock Drafts Required)

This method is very straightforward and is easily and quickly done using any spreadsheet application:

1. Sort your player pool by overall ranking.
2. Separate your pool into tiers of equal value.
3. Sort each tier by ADP.
4. Adjust as desired/needed.

The potentially hardest step is the second one. The key here is to group players based on how satisfied you would be substituting any of them at a given pick. Chances are your top players will be miles ahead of the rest and deserve their own tier. If your ranking system is quantitative, mathematics can easily set the tiers for you. One popular method when using VBD is dividing the VBD amount by 16. Theoretically, all players with the same (VBD / 16) value will be within 1 fantasy point per game of each other, so they belong in the same tier.

Here's the top of an overall list as an example:

Notice the blank 'Tier' column. Since a VBD amount is present, populate the Tier with VBD / 16, rounded up. Always rounding up simplifies the value for your spreadsheet as VBD approaches zero.

Now sort the list by Tier (descending) and ADP (ascending).

Notice the order of players has changed slightly but still somewhat close to the original rank. Using this sort for an autopick list allows you to account for the player's ADP while minimizing the value-loss risk.

Let's use the list in a live draft example. You've drawn the ninth pick in a 12-team league with a serpentine method, giving you picks 1.09 and 2.04. If players are taken as expected, you'll have Steve Smith as the best available player on your overall sheet. Since do you don't expect him to be available for your next pick, drafting him here would be appropriate. Looking more closely, you've noticed the draft started with an RB rush, which should continue for the rest of the first round. By sacrificing less than .25 points per game, you can ensure a quality RB (Travis Henry) and still get a WR of equal value in the second round.

The tier/ADP sort executes the above example automatically. When a number of equal-value players are available, it will take the one least likely to last through the next round. This allows you to maximize total team value by getting the most players in any given tier.

There are some limitations to this approach, especially when using VBD or another quantitative method for overall ranking. Since most team defenses and place kickers have similar fantasy point totals, they will bunch up toward the middle of the list. This means that backup PKs and Defs. will be placed above many of the reserve RB and WRs, and most owners prefer it the other way around. There are a few ways to accommodate this.

Using Tiers within a Mock Draft Shell

By running a mock draft, you can get a better overall picture of when certain positions are being taken and then fill our autopick list accordingly. Once again, a spreadsheet application is very useful for handling the data.

1. Sort your player pool by overall ADP and your position lists by tiers/ADP.
2. Perform a mock draft where each team draft based on the overall ADP list and positional need.
3. Store the mock draft results: note which position was taken with each pick.
4. Create an autopick list using the draft results as a shell and filling the players in using your positional tier/ADP lists.
5. Adjust as desired/needed.

Here's the sample list. In this example, we've sorted this by tier/ADP and overall ADP ranking, so you can see how the sorting affects the list.

If you performed a mock draft using overall ADP, you would draft eleven running backs before selecting a player from any other position. The draft as a whole may look like this.

Remember when recording the mock drafts results that the individual players being drafted are not important. You just need to note which position was taken for each pick of the entire draft. This position data will be the shell of your autopick list. To fill this shell, you'll need your tier/ADP list broken out by position. Sample RB and WR lists are shown below.

From here, you'll combine the draft results with your position lists. Since the first eleven picks were running backs, start the autopick list with the first eleven players on your running back list. Following the draft results, add the first wide receiver, the first quarterback, the second wide receiver, your next two running backs, and so on. Continue this process for the entire draft. If the 58th pick of the draft (pick 5.10 in this example) was the third tight end taken, then the third player on your tight end list will be the 58th player on your autopick list. Using the sample data, the autopick list for the first two rounds of the draft would look like this.