John Elway and his group of talent evaluators at Denver Broncos HQ have watched hours of film, conducted interviews, examined medical reports, clicked stopwatches, talked to college coaches, among other tactics, to prepare for the NFL Draft.
Have the Broncos also used analytics to assist in their decision making? Fans should hope so because there is valuable information to be gleaned by using draft analytics when preparing for that all-important pool of new talent.
In a perfect world, when a team is on the clock, it selects the highest-graded available prospect, especially when the prospect is at a position of need. It seems like a no-brainer. Easy, right?
However, the world isn’t perfect and the draft is even less so. Decisions are always more difficult than they first appear.
In the real world, when a team is picking out that card to send to the podium, there are usually several prospects available who are all graded similarly. Some are at positions of need. The decision isn’t easy. So how can analytics help? Allow me to illustrate.
Most draftniks agree that the Broncos’ top five needs are wide receiver, cornerback, defensive line, interior offensive line and offensive tackle.
The following strategies based on analytics will aid in the decision-making process when the Broncos are faced with the task of choosing among similarly-graded players when they find themselves on the clock on draft day.
Every player drafted since 2011 is among the data set in this analytics study and the metric used is Performance Value (PV). Performance Value is Pro Football Reference’s Car AV, but averaged across every season of a player’s career.
By doing this, PV accounts for seasons lost due to injury. This value can be used to compare players and position groups.
Let’s examine a couple of scenarios to understand how the analytics will help the Broncos in their draft strategy.
Highest PV by Position
With pick number 15 in the first round, the Broncos might feasibly have to decide on selecting South Carolina DL Javon Kinlaw, Alabama WR Henry Ruggs III, Louisville OT Mekhi Becton, or Florida CB CJ Henderson. These players are graded high enough to be considered as reliable options to be selected at picks 13-17, right in the area that the Broncos will be. All four prospects also represent areas of need.
First, the comparison of the average PV of players by position selected in round one starts to shed some light on the decision process. Historically speaking, OT has had the highest PV, followed by DL, WR, and CB respectively as the below graph indicates.
But one cannot stop there in the analytics evaluation. It is important to know the risk each of the positions poses in round one. The following is the success rate of players who were selected in round one (success rate is defined as any player whose PV is greater than the mean for their position).
Success Rate by Position
CB has the highest success rate at 97%. Followed by OT and DL at 89%, then WR at 74%. WR has the lowest success rate and therefore the highest risk for a bust in round one.
Since 2011, 26% of WRs selected in round one have been below the mean (average) for their position. Remember, this is the mean for all players selected in any round, not just players selected in round one. Selecting WR is a very risky first-round proposition.
Next, one must consider how good or bad each position can be. DL has a max PV above the mean at 10.3. Meaning, the best single PV selected in round one of these four positions came from the DL.
WR comes in second place at 8.2, followed by OT at 7.9, and finally CB at 7.6. How about the examples of bad performers? The worst individual Performance Value is at OT at -2.7 below the mean, followed by WR at -1.8, followed by DL at -1.3, and finally CB at -0.6.
In laymen’s terms, CB may not have the highest value in the group, but the risk of a terrible bust is low. In comparison, OT has a similar peak value of CB, but there is a stronger risk of selecting a terrible bust.
These metrics indicate that DL has the highest PV of the four positions and is a relatively low risk.
For visual representation purposes, the below graph illustrates every player selected and their performance against the mean by round.
Knowing When to Draft Certain Positions
The final area of analytics to be examined is the 'drop-off' in Performance Value by round for each position which helps the Broncos deduce whether to wait until the second round to take a certain position. The Broncos do have one second- and three third-round selections to address their top-five needs
The most significant drop in PV going from round one to round two is at DL with -2.37. The smallest drop is at WR with -0.46. The risk of waiting for the second round for a DL is higher, while waiting on WR is lower.
In fact, the PV of a WR in round two is almost the same as round one. The Broncos could wait until round two and find a WR who is likely to perform as well as their first-round counterpart. The opposite is true for DL.
Best Choice in Round One Based on Analytics
If the choices were limited to the four originally listed in this exercise, the analytics would point to South Carolina DL Javon Kinlaw as the best option.
In round two, the Broncos could have to decide between similarly-graded players such as Arizona State WR Brandon Aiyuk, Houston OT Josh Jones, and Alabama CB Trevon Diggs. Using the same analytics process, the Broncos can make an easy decision.
OT has the highest round-two Performance Value followed closely by WR and then CB as a very distant third.
WR has the highest success rate at 79%. CB is at 63% and OT is at 59%. The highest individual peak value also is at the WR position at 11.2.
OT has an individual peak value at 5.6 while CB comes in last at 4.0. The lowest individual peak value is with OT at -2.4 while WR and CB are tied at -1.9.
Round Two Best Option? WR
The WR position has the highest reward with the lowest risk and frankly, a round two WR's value is almost the same as one selected in the first round. The drop-off in Performance Value to round three is nearly the same for all three positions.
The analytics overwhelmingly point to WR in round two. The decision would be Aiyuk in this scenario.
If these two scenarios unfold this way on day one and two of the draft, and the Broncos want to maximize the Performance Value and limit the downside risk, Elway should select a defensive lineman in round one and a wide receiver in round two.
From there, Denver could fill out the final three positions of need (CB, OT, and iOL) in round three with that trio of picks Elway has.