Analyzing NFL Drafts by Position

And why the Raiders are a huge team to watch this April

“Gruden, how many spots do you want to move up in the draft?”Credit CBS Sports

Every year mock drafts emerge that are completely wrong. It’s natural, it’s what happens- picks get traded, one team rated someone higher than they should’ve, someone gets caught in a scandal just before draft night. It makes mock drafts inherently inconsistent.

Mocks have their purpose: for me, it used to be how I learned about potential picks for the Eagles. I figured out who some of the better players were, and got disappointed when another team drafted those players. It made draft night make sense.

Later, they kept me from losing my mind after the Eagles were done for the year. It stifled my boredom. Eventually, I started making my own. I predicted that the Eagles would take Nelson Agholor (though not in the first round), and I had a good idea they wanted Eric Rowe (didn’t think they could get him until they drafted Agholor early). Two years ago I had Derrick Barnett locked in well before the draft. I thought they’d grab Sidney Jones at 14 until he got hurt, and they still got him in the second round.

So, I’ve had success. But there was still such randomness. I don’t like randomness, I like to know what’s going to happen before it happens. That makes everyday life all the more surprising, but it can make things like the draft challenging.

This year I decided to try to rise to that challenge, and figure out the analytics of the draft, or at least the first round of it.

And so, I ran the numbers to figure out not which players would get picked when, but which positions would get picked when. I thought that if I could match up a big board with the positional draft analytics I could more accurately predict the events of the draft this year.

I took data from the last 17 drafts, or since the NFL went to its current 32 team status. The league has changed in the last 17 years, so I compared my data of the last 17 years to data of the last 5 years, and the data was accurate 66% of the time, so I see this as being pretty consistent data. Drew Brees has a 67% completion percentage throughout his career, so see this data is as consistent as him.

So, the data:

*Note- I lumped some groups together to show a consistency of the type of player chosen. For example, some rounds the pick was predominantly DTs and DEs, so I put Defensive Line. For others it was predominantly DEs and pass-rushing LBs, so I put EDGE.

Over the last 17 years, these are the top two positions drafted at each pick of the first round:

PickPosition 1, % pickedPosition 2, % picked
1QB 71%EDGE 18%
2QB 24%OT 24%
3EDGE 29%QB 29%
4OT 29%RB 24%
5Defensive Back (DB) 35%EDGE 18%
6Offensive Lineman (OL) 29%Defensive Line (DL) 24%
7WR 35%DB 18%
8DB 29%EDGE 18%
9LB 24%WR/OL 18%
10QB 24%CB 18%
11DB 35%EDGE 24%
12DL 41%RB 18%
13DL 47%EDGE 24%
14DL 47%DB 29%
15DL 24%LB 24%
16DB 29%EDGE 24%
17LB 29%DB 24%
18DB 35%EDGE 35%
19OL 35% (29% OT)DB 18%
20DE 24%WR 24% (and OL 24%)
21Offensive Playmaker (WR RB TE) 35%OL 29%
22QB 29%EDGE/WR 24%
23OL 35%OP 29%
24DB 35%OP 24%
25DB 41%LB 18%
26EDGE 24%OL/WR 18%
27DB 41%RB 24%
28DL 35%OL/OP 24%
29OL 24%WR 24%
30OP 47%DB 24%
31DB 24%LB/RB 18%
32OL 24%DL 24%

As expected, the deeper into the draft, the more unpredictable the picks become. The first ten picks, however, show which positions are valued the most in the NFL. QBs, defensive ends and rushing OLBs, and defensive backs are the foundational pieces to franchises.

This much we knew.

What we don’t know quite yet is what the picks will be this season. With the Cardinals, 49ers, and Jets pretty set at QB, that leaves a question for the first overall pick. The Cardinals have to be looking at Nick Bosa or another defensive lineman for the first pick, and there’s a firm chance that they’ll take Bosa or someone else (it’d be Bosa).

Historically, however, there is still that 71% chance that it’s a QB. The Cardinals drafted Josh Rosen last season, and won’t select a QB anywhere near the first 100 picks this year. One should really consider the possibility of the Cardinals trading the first overall pick. As a team on the rebuild that has needs at almost every position, you’d have to see the value of stockpiling more picks (like the Browns did) when the franchise QB is already in place (unlike what the Browns did).

Josh Rosen is the key to the first few selections of this year’s draft. Credit to ESPN

If the Cardinals “moneyball” this draft, they could fill more needs with instant-starters and contributors from the first round.

The catch is that this would require a team that needs a QB and has an abundance of valuable picks in the draft. Which teams are in a position like that?

Only one, the Oakland Raiders.

Do they really need a QB? Well, not yet. But Jon Gruden was “smitten” with Drew Lock, and met with other QBs quite a bit. A year removed from trading the Raider’s best pass rusher, having three picks, and having a terrible season, I can’t imagine a coach would try to cover their tracks by drafting another pass rusher. Gruden is “the QB guru,” and coaches have HUGE egos. He’ll say that he didn’t have his guy, something will happen to Derek Carr, and the Raiders will have the ammo to trade up in the draft—solidifying their spot over teams who may also be trading up to get their QB (the Jags and Giants)—and restarting the franchise.

Could the Cardinals pass up getting 3 first round picks to move back 3 spots in the draft? They might miss out on a pass rusher (Bosa) but they could still get a top 5 talent in the draft and fill other needs with first round picks.

Because of the data showing that there’s essentially a 71% chance a team trades up for the first round pick, it begins to stir up who will have the other picks in the draft. Although Jonah Williams and Greg Little are excellent prospects to be picked, it seems more likely that another QB will be taken second overall instead of an OT. This goes against the idea that this draft will be one defensive player after another if the first two selections are QBs.

Based on the data I would predict with confidence that picks 3-6 will all be defensive linemen, perhaps with the exception of one offensive tackle being selected.

DT Quinnen Williams will be a top-5 pick this year. Credit to Touchdown Alabama

Here are more findings from this data:

63% of the time the position picked over the last 17 years has been on defense, and this year that percentage could be even higher.

Josh Jacobs—who is widely considered the best RB in the draft (and Underdog’s highest ranked RB as well)—seems to have a hot spot of being drafted between picks 4 and 12.

The top linebackers in the draft, Devin White and Devin Bush, will likely both be taken off the board by pick 20.

In a year that is not considered particularly strong for QBs, one can still expect about 4 signal-callers to be selected in the first round. These 4 are likely going to be between Drew Lock, Dwayne Haskins, Kyler Murray, Daniel Jones, and Will Grier.

Greedy Williams seems to be a likely top-10 selection, as defensive backs are a high choice at the top of the draft.

In a year loaded with DL talent, this data aligns well with prospects available.

Following the Super Bowl, we will be providing a variety of mock drafts based on our predictions from the data, predictions based on team needs, mocks with trades, and mocks without trades.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s