Draft strategy: Move up or move down?

The phones in NFL war rooms generally get a pretty good workout during the draft. Some teams are looking to move up and grab a player they covet before another team can select him. Other teams do not see anyone exciting at its draft spot and would just as soon move down and add an extra pick or two. Trades happen when either there are two teams with complementary strategies (e.g., one wants to trade up and one wants to trade down) or a team looking to trade up offers enough of a premium to entice a potential trade partner.

Typically, some variation of the Trade Value Chart is used to determine the fair value for a trade. DRAFTMETRICS is not a fan of the trade value chart but at least it brings a semblance of structure to the trade process.

In this article, DRAFTMETRICS addresses the issues surrounding move up/move down decisions. DRAFTMETRICS will first establish the relative value of draft choices and then determine, given the consideration involved, whether recent trades favor the team moving up or the team moving down. The Value Groups previously defined by DRAFTMETRICS are used to establish the relative values. To refresh your memory, the seven Value Groups are as follows:

The premise of the Value Group is that every selection within that Value Group has the same chance of success. The 13th selection, for example, has the same chance of success as the first selection. DRAFTMETRICS reviewed the relative value of draft choices for each of five possible metrics:
• The probability of at least a five-year career
• The probability of being at least a three-year starter
• The probability of being at least a five-year starter
• The probability of being selected to the Pro Bowl at least once
• The probability of being selected as an All Pro at least once.

Data was used from the previously published “Digging Deeper into Draft Probabilities” to establish the relative values under each metric, as shown in the tablesin Appendix 1 to this article. The most appropriate metric depends on the Value Group. For selections in Value Group 1, for example, a team really is not hoping for a player to have a five-year career without reaching starter status. The team is looking for an impact player, so the Pro Bowl and All Pro metrics may be the most appropriate. For a selection in Value Group 7, though, a team may be content to draftsomeone who can be a backup and the five-year career metric can be the most appropriate.

As a side note, multiple All Pro selections were excluded from the metrics because there is such a dominance of the measure by selections in Value Group 1 (e.g., 24 ofthe 50 three-time All Pros drafted from 1993-2012 were from the first 13 picks) that the comparative values were completely distorted. The relative values establish the minimum consideration that should be acceptable to the team trading down. Appendix 1 shows, for example, that when moving from Value Group 2 to Value Group 3 and using five-year starter as the operativemetric, a team moving down should receive at least a selection in Value Group 4 in addition to the Value Group 3 selection.

DRAFTMETRICS identified 82 instances from 2010 through 2012 where a team traded up on or somewhat before the day of the draft. DRAFTMETRICS divided the trades into four categories:

• Trade-ups within the same Value Group (“Same VG”
• Trade-ups from one Value Group to another (“New VG”)
• Trade-ups with future year draft choices involved (“Future”)
• Trade-ups with players as well as draft choices involved (“Mixed”)

The 82 trades can be summarized as follows:

DRAFTMETRICS did no further analysis of the “future” and “mixed” categories because they introduce a level of complexity that would take the article off on a tangent.

Trade-Ups In the Same Value Group

Exactly half of the 82 trade-ups occurred in the same Value Group (e.g., selection #50 was traded for selection #45 with both selections having the same value). In those cases the team trading down always has the advantage as they would typically receive a premium as an inducement to make the trade. The team trading down almost always has the most leverage in any trade up scenario.

The next table summarizes the consideration received in the 41 trade-ups in the same Value Group. The five situations where no or multiple consideration was received typically involved multiple draft selections, all within the same Value Group, where a team moves up in the selection order but not to a different Value Group.

Moving down within a Value Group is pretty much a no brainer unless a team bypasses a player it really wants. The team receives a selection with the same chance of success as the one they surrendered plus a premium, so why wouldn’t they make the trade?

Moving from One Value Group to Another

The 20 trades where a team traded up from one Value Group to another are summarized in Appendix 2. Appendix 2 shows both the minimum value that should have been received (as reported in Appendix 1) and the actual consideration received, both expressed as selections in a Value Group. The minimum trade values are presented for each of the five metrics.

The table in Appendix 2 shows that the consideration paid often has more to do with how far a team moves up than it does with the Value Group. DRAFTMETRICS believes the Value Group approach has more validity than going strictly by the number of positions moved. Although there is a certain intuitive logic that supports the case that each successive draft position is worth a little more than the one preceding it, DRAFTMETRICS believes it has demonstrated in prior articles that this is not the case. Rather, “clumps” of draft choices have had the same historical outcome and this is reflected in the Value Groups.

While acknowledging that the sample size is small, the review of the 20 trades may shed at least a little light on the proper strategy. The following table summarizes the 20 trades and includes:
• The number of trades in each transaction type (e.g., Value Group 2 to Value Group 1)
• The most relevant three metrics (in order) for each transaction type
• Whether the team trading up or down has the advantage as measured by the value of draft selections exchange
• Comments on the transaction type

In the 20 transactions studied, teams trading up generally paid less than the relative value charts would indicate. This would tend to indicate that teams trading had the advantage in those trades.

Appendix 1: Relative Values by Valuable Group for the Five Metrics

Notes:
(1) As an example of how to use the tables, the “Five Year Career” table indicates that a Value Group 4 selection plus a Value Group 6 selection were equivalent to a Value Group 1 selection.
(2) Two “x”s indicates that two selections in that Value Group must be received

Appendix 2: Listing of Trades Where a Team Moved to another Value Group

Notes:
(1) This chart is read as follows. The first transaction listed represents a team trading up 8 spots from Value Group 2 to Value Groups 1. The team that moved up paid a Value Group 2 selection and a Value Group 3 selection to move up. Using the probability of a 5-year career as the metric, the relative value for a VG1 selection is a Value Group 4 pick and a Value Group 6 pick. At the other end of the spectrum, if the 1-Time All Pro is used as the metric, the team moving up would have to send a Value Group 2, Value Group 3, Value Group 4 and Value Group 5 selection to the team moving down.

Tony is the founder of DRAFTMETRICS.COM can be e-mailed at draftmetrics@gmail.com and followed on Twitter @draftmetrics

Upcoming Games