A new way of building a draft big board

Josh Queipo combs through the data to build a more 'league-accurate' draft board

There is something funky about the draft being one of the league’s marquee events. In the abstract, it’s humans guessing about the future prospects of humans. A stack of cash is involved. No actual action is played. At root, it’s a man stepping to the podium to read names. And yet, the draft remains one of the fastest-growing aspects of the league (and its coverage), monopolizing time, reads, and clicks from other sports where players are, you know, on the field or court.

Why? Draft picks can be anything! And it gives everyone, from analysts to fans, a chance to play armchair GM, which has been the heartbeat of the transaction-over-action analysis that typifies modern leagues. The NFL still dominates Sundays – and Mondays… and Thursdays… and playoff Saturdays. But it also gobbles up the offseason with showcase bowls like the Senior Bowl and Shrine Bowl, where coverage and viewership have grown year over year. In 2024, the Senior Bowl boasted over 900 team personnel and over 1,100 media members in attendance for the week’s events this year. The Combine has also become a primetime event. Is carried live by the NFL Network and covered by multiple other major media outlets across the country.

The draft itself, once a secondary piece of programming aired during daytime hours the last weekend in March, has now become a primetime event with the first round airing across pivotal viewing hours on Thursday night and the second and third rounds occupying similar time slots on Friday evening before the final rounds airing on Saturday. The NFL reported impressive viewership numbers for the draft last year and expects similar, if not better, numbers this year.

As such, a cottage industry of draft evaluation has grown. Cost-controlled talent is one of the foundational pieces of roster management in a league that actively works to create parity through a limited resources system. The fastest way for a moribund franchise to turn things around is through an epic draft that infuses young talented players that will empower the team to funnel financial resources into complimentary pieces.

This cottage industry, whose roots can be traced back to the godfather of the space, Mel Kiper Jr. of ESPN, revolves around assessing each player’s position, physical build, college production, and physical traits to determine an overall evaluation of how likely that player is to help a franchise. With just 250-260 draft picks each year across the league one of the most important aspects of these evaluations is the tiring of players into rounds.

With such tidy dividers, almost all evaluators started to use round grades as one of many ways to differentiate prospects. But, as with most things, such slim margins mean different things to different evaluators. And this isn’t just various evaluators coming to different on prospects. Most evaluators see the process of how they grade differently, even when it comes to breaking down those round grades. Some have worked in the league or taken their lead from a specific team’s formula. Others have charted their own path, relying on historical markers or trends to split prospects between a Day One, Day Two, or Day Three guy.

The simplest way to execute this process would be to line up each prospect on an evaluator’s “Big Board” and divide the round grades as the NFL divides the rounds: The top 32 players get a 1st round grade; the next 32 get a second-round grade. So on, and so it goes. But it’s a struggle for many evaluators. Every year is different, and every analyst has their individual thresholds. For some, the gulf between the first and 32nd player may not be as wide as for another analyst. And yet, when they publish their rankings, both analysts will have players slotted one-to-32, with a corresponding ‘first round’ grade.

To better reflect their evaluations, analysts have started to refine how they dole out grades. For many evaluators, first-round grades are only bestowed upon the top 8-15 players in a draft, while they may give out as many as 50 or more third-round grades. Each analyst has developed their method.

When discussing processes with analysts in the public space, all have unique processes for evaluating individual players and position groups, and how they delineate their “round grades”. None – besides Kiper – have a process that matches rounds to the number of picks that make up each round of the actual NFL Draft.

NFL teams do not look at the draft the same way as those on the outside. One team executive I spoke with said they do not think any team uses round grades as part of their evaluation process. So why should the public at large? With each passing year, coverage and discussion of the league has pushed more and more to eliminate the gap between how fans and the league itself understand and talk about the game.

Major media companies like ESPN, NFL Network, and The 33rd Team employ people who have worked within the league to help translate it to the consumer. Analysis is becoming more minute and nuanced. Thirty years ago, Merril Hodge and Ron Jaworski were about the only game in town breaking down what is now commonly referred to as “All-22”. Now, with increased accessibility to those tools, there is no shortage of deep-dive analysis. Football vernacular is expanded. Understanding of the salary cap and contracts is at an all-time high. It should stand to reason that draft analysis should follow suit.

Using Data To Create A More Centralized Approach

Most players drafted will not amount to much at the NFL level. Of those who are drafted, over 50% will not even start for the majority of games in a single season, according to Pro Football Reference. Looking at the 2011-2018 draft classes shows some striking similarities year-over-year in overarching outcomes from each class.

Within each classification, there are similarities. The average number of players within each draft class who become an All-Pro at least once centered around 5% of the total population. The high was 2017’s class with 6.72% of the 253 draft selections earning the distinction and the low being 2014’s 2.73%.

Similarly, the percentage of players selected to at least one Pro Bowl was relatively stable, around 13%. The same could be said for the number of players who started at least four seasons of their career, three seasons, two seasons, and one season. The final group, those players who started at least one season in their career, makes up just over 50% of the entire drafted population in this sample.

Given the stability of these numbers, I suggest that a more statistically appropriate and potentially uniform approach to tiering players would be similar to the simplest method I noted earlier. Evaluators should arrange their draft boards and differentiate tiers using the following method. In parentheses after the tier would be the number of players that fall into this tier based on a 255-player draft.

Tier 1 - Top 5% - Most likely to become All-Pro’s (13)

Tier 2 - Next 8% - Likely to earn a Pro-Bowl nod (20)

Tier 3 - Next 13% - Likely to start at least 4 years in their career (33)

Tier 4 - Next 6% - Likely to start at least 3 years in their career (15)

Tier 5 - Next 8% - Likely to start at least 2 years in their career (20)

Tier 6 - Next 12% - Likely to start at least 1 year in their career (31)

That would account for 52% of evaluated players, 132 players total. And it would match the data above in terms of expected outcomes. All other players would fall into an amorphous “tier 7”.

If the use of “rounds” has become so ingrained in our vernacular that we cannot fully abandon the use of the term, this tiering system would allow an easy translation.

This process helps to put into context the realities of draft outcomes. It can also help to explain why NFL teams still place such a premium on high draft picks despite ample research showing the relative value of those picks to be lower than previously thought. But that shows the flaws in the evaluation of players more than the valuation of picks in a vacuum.

The definition of round grades at this juncture is outdated from expected outcomes. Teams notoriously have boards that include far fewer than the 250-260 prospects actually selected each year – Bill Belichick routinely entered drafts with a 60-man board. That is because they inherently know that only around half of those players will contribute at a starter level for even a single year. Their goal is to find as many within those first six tiers as they can with the limited number of picks they have.

For the rest of us, it would make sense to shift our own understanding of draft prospects based on the probability of outcomes — and adjust our draft boards and round grades accordingly.