Given that it’s NFL draft season, we’re going to start off with a series of posts on the NFL draft. This post will cover the idea behind the draft value chart, including the presentation of my very own draft chart. In the coming weeks, I’ll be posting pieces that answer some interesting (well, at least interesting to me) draft-related questions, including:
- Why it’s a waste to draft a kicker or punter (no matter how good a rookie season he has or doesn’t have);
- Positional value and variation in the NFL draft;
- Which teams are better at drafting others (hint: Matt Millen is a much better broadcaster than GM); and
- Comparing the value of past drafts relative to other drafts (we always hear from the pundits, “this is a strong or weak draft” – we’re going to check if they were right).
The work presented on this site is intended to be a work in progress so while I’m posting my own initial draft chart here, I’ll be making adjustments to it over the next month.
Draft Value Chart
For those who don’t know, a draft value chart compares the value of one draft pick relative to other draft picks. The Cowboys in the 1990s made it famous as they used it as a tool to assess whether to trade up or trade down during the draft. The chart was named after Jimmy Johnson, the head coach of the Cowboys during the early 1990s, who led the team to two Super Bowl victories, thanks to the team’s success in the draft, including this. Below is the Jimmy Johnson draft value chart.
Now, I understand that having the first overall pick offers a team the chance to take a bona fide stud like JaMarcus Russell, but early second round picks have shown to have the most surplus value of any pick in the draft. Teams could easily land multiple day-one starters with those picks. Think about the two recent trades the Inglewood Rams have made involving the top two picks in the draft. Take away the fact that RGIII has been a relative bust, do you think they would make that trade again? I think so. Do you think they would trade up for Jared Goff (the man with one of the most responsible irresponsible first purchases) again? Even though the jury is still out on Goff, I doubt it. Currently, all teams use some form of the chart to quickly assess the fairness of trades involving just draft picks. However, as several people have noted before me, the chart overvalues early first round picks relative to other picks in the draft. For example, according to the Jimmy Johnson chart, a team could trade the first overall pick for the first five picks of the second round and the 70th overall pick!
Given the well-documented issues with the Jimmy Johnson chart, there have been numerous attempts to create a draft value chart that more accurately reflects the value that a team can expect to receive from the draft. The best two that I have seen so far (if you see any others, please send them my way!) are from Ethan Young and Chase Stuart.
I like Ethan’s piece. He identifies some of the core issues with Approximate Value or AV (a metric developed by Pro-Football Reference to approximate the value of every NFL player relative to other current and past NFL players) and makes a lot of adjustments to create his Modified Approximate Value Evaluation Model (MAVEM) to estimate the expected value a team derives from each pick. However, there is one crucial piece of his analysis with which I disagree. Ethan looks at the sum of the total value of a player over his entire career. Looking at the full career of players means we can’t look at recent draft results given that some players’ careers are not complete (the 199th pick in the 2000 draft just won’t retire…).
In addition, given that the average career length for an NFL player is 3.3 years (it’s not as low as 2.6 years; there are a lot of issues with the recent Wall Street Journal piece on this), I hypothesize that most teams don’t draft players expecting them to spend their entire careers with that team. For example, of the 32 players chosen in the first round of the 2012 draft, only 13 remain with the team that drafted them. That is, after five years, more than half of the players selected in the first round are either no longer in the league (Trent Richardson, Justin Blackmon, Quinton Coples, potentially RGIII) or have moved on to another team. Some of the players moved on to sign a big contract in free agency because their drafted team either didn’t want to or couldn’t pay them the big bucks (Stephon Gilmore, Kevin Zeitler) or their drafted team discarded them for various other reasons before the end of their rookie contract (Michael Floyd, Shea McClellin). Given the constant roster turnover, it’s unlikely that teams draft players thinking about the long term.
Instead, teams want to maximize the value they receive from the players for the years in which they have control over that player. Under the current Collective Bargaining Agreement, that is the first four years of a player’s career with a team option for the fifth year for first round picks.
Unlike Ethan, Chase only measures the production of a player over the first five years of a player’s career. Chase looks at the marginal value of a player compared to that of a replacement level player on a yearly basis. Players only get credit for the AV produced above two points of AV in each season given the small amount of output needed to produce two points of AV. I like this approach, which has been adopted and used in the mainstream media. While looking at the short term gains has some downside (Aaron Rodgers couldn’t become a starter until his fourth year?!), it makes a lot more sense to understand the value of each pick by valuing players over this length of time.
My Draft Value Chart
My draft value chart is largely based on Chase’s, except for a few key differences:
- Chase looks at drafts that occurred from 1980 to 2007 (he produced his chart in 2012, which means that the latest draft he could analyze was 2007).
- Chase sums the weighted average ‘career’ AV of the first five years for all drafted players.
Unlike Chase, I look at draft results from 2002 to 2012. Why stop at 2002? The Texans’ first draft was 2002, which means that 2002 is the first draft with all of the current teams in the NFL (and 32 picks per round before compensatory selections) and with the current divisional alignment. Going back to 1994, the first year of the seven round draft would make sense, except that it wouldn’t take into account the extra picks available by adding another team.
In addition, I sum the weighted average ‘career’ AV for the first four years for all players drafted in the second round and later and the first five years for first round picks. Why? Since teams have the option to extend first round picks’ contract to a fifth year, first round picks are more valuable relative to players chosen in later rounds. Teams can retain these players for an additional year and delay them from going into free agency an additional year. Teams have a financial motive to choose players in the first round as opposed to later rounds. In my view, this is something that should be accounted for and something that increases the inherent value of a first round pick.
Finally, I include compensatory picks in the analysis. The round in which a pick falls does not matter. At the end of the day, the player chosen with the 135th pick is the 135th player selected, whether that falls in the fourth round or the fifth round. This is important as most draft capital analyses did not include compensatory picks since they could not be traded. However, starting with this year’s draft, compensatory picks can now be traded.
Under my draft value chart, the value of each pick looks like this:
And like this:
So what does my chart tell us? Teams tend to overvalue early picks, somewhat due to the pervasiveness of the Jimmy Johnson draft chart. As I stated earlier, under the Jimmy Johnson chart, the first overall pick was worth the first five picks of the second round and the 70th overall pick. Per my chart, the first overall pick is worth 34.4 points compared to 68.8 points for the first five picks of the second round and the 70th overall pick. Now, the first overall pick is worth the first two picks of the second round and the 54th pick.
What does it mean that the first pick is worth 34.5 points of AV compared to 30.1 for the second pick? By itself, nothing. This chart and any draft value chart has no value in terms of determining the number of wins a team can acquire in the draft. Its value is its ability to compare certain players and picks relative to other picks. We can ascertain the value of a certain pick compared to other picks.
Teams may still overvalue those early picks, but at their own peril. Even at a vital position like QB, it may not always make sense to trade up or keep the first pick to grab a guy like David Carr, even though guys like his biggest fan ever (and a much better player) was available at the start of the second round and taken 36th overall.
Also, later picks (5th to 7th round) hold value. Teams can still receive above replacement-level value for later picks in the draft. I like to think about later picks as a shot in the dark. Most picks won’t work out – some won’t make the team, some will only contribute on special teams (an impact we don’t currently measure other than kick and punt return touchdowns), but every once in a while you hit gold with the 199th pick or even the 252nd (though I would settle for silver or bronze with these picks).
Overall, I think this chart produces a more reliable picture of the expected value of each pick than the Jimmy Johnson chart or Chase’s chart. That said, I’ll be including some of the inherent weaknesses of the methodology behind this chart in my future posts with the goal of producing an updated chart that you can use while watching the draft this year.
I want to thank you for taking the time to visit this site. A lot of hard work has come into putting this together and I hope you enjoy the details and insights from this page. This site is intended to be a forum for discussion on any sports analytics topics, but it will start with a series of posts on the NFL. I am very open to guest writers on this site and would love to hear any constructive feedback/discussion on my posts (side note: I’m okay if you call me an idiot, but at least tell me why). Any suggestions for improving this chart are more than welcome! Thanks for reading!
 Though both are now Super Bowl champs