Breaking Down the ‘Boys: “Wisdom of the Crowds” can help predict Cowboys’ draft pick
By Jonathan Bales
jonathan@thedctimes.com
11:20 am on April 5, 2013
Jonathan Bales is a special contributor to SportsDayDFW.com. He’s the founder of The DC Times and writes for DallasCowboys.com and the New York Times. He’s also the author of Fantasy Football for Smart People. He can be reached at
jonathan@thedctimes.com.
You can follow him @TheCowboysTimes.
I’ve recently unveiled my big board and second mock draft. While I spend a good amount of time analyzing prospects, I have my own biases and my rankings are certainly not immune to error. I rank a pretty large percentage of players away from consensus opinion; some of those will turn out to be right, but many will also be wrong. There’s a certain level of uncertainty built into forecasting the future of NFL prospects, and as draft analysts, all we’re trying to do is peel away that uncertainty to make accurate predictions.
As I mentioned a few weeks ago, I’m a big believer in “wisdom of the crowds”—the phenomenon by which the collective opinion of individual experts can be more accurate than that of the same expert opinions take in isolation, i.e. as we incorporate more and more expert opinions, we’ll likely gain a clearer depiction of reality. My opinions on one prospect might be vastly different than those of, say, Mel Kiper. If we’re playing the percentages, the prospect’s true talent is most likely to fall somewhere between our separate views on him. As we add in more and more expert opinions, we can get a really great sense of a prospect’s perceived worth.
Up until now, I’ve been creating aggregate big boards my combining the rankings of various draft analysts. The aggregate boards have been based off of their big boards—their personal opinions regarding prospects’ talent—as opposed to mock drafts. There’s a difference between an analyst’s opinion on where a player should get drafted and where he will get drafted, however, and today I’m going to look at the latter.
While I’ve been personally tracking expert rankings in my previous aggregate boards, there’s actually already data available that combines expert opinions on where prospects will get drafted. Play the Draft is a “stock market for the NFL Draft”; experts like Mel Kiper, Lance Zierlein, Matt Miller, Greg Cosell, and so on rank prospects according to where they think they’ll get drafted. As those rankings change, so does the “stock” of each player. You can build your own team, trying to predict future trends to acquire value in much the same way NFL teams do.
Understanding Changing Value
The coolest part about aggregating expert opinions is that you can get a sense of how a player’s stock is changing. In the same way that an actual stock price fluctuates, so does the stock for a draft prospect. Many times, NFL teams can actually acquire value by jumping on a players’ whose stock is down. It’s those prospects—not those whose value has hit its peak—who are most likely to offer value because their perceived value has dropped below their actual worth.
You can see that with a guy like Utah defensive tackle Star Lotulelei. After rumors of a heart condition surfaced, Lotulelei’s stock plummeted.
That’s a case where a player’s perceived worth has probably dropped well below his talent level. Compare that to Florida defensive tackle Sharrif Floyd.
That’s a case where a player’s perceived worth has probably dropped well below his talent level. Compare that to Florida defensive tackle Sharrif Floyd.
After a report or two that a few teams view Floyd as a top talent, his stock soared. Now it appears as though every analyst views Floyd as a top-tier player, but that wasn’t the case just a couple months ago. Whether or not Floyd is an elite talent, it’s very unlikely that his actual value is greater than his current perceived worth, which is through the roof.
A New Aggregate Big Board
Below, I’ve posted the top 25 players as ranked by the experts at Play the Draft. Again, the value of such a board is that it corrects for errors in individual rankings. This “wisdom of the crowds” approach has proven to be more accurate than the majority of individual boards alone. You can click on highlighted names to check out my scouting reports.
1.Luke Joeckel, OT, Texas A&M
2.Sharrif Floyd, DT, Florida
3.Eric Fisher, OT, Central Michigan
4.Geno Smith, QB, West Virginia
5.Dion Jordan, DE/OLB, Oregon
6.Dee Milliner, CB, Alabama
7.Ezekiel Ansah, DE, BYU
8.Star Lotulelei, DT, Utah
9.Lane Johnson, OT, Oklahoma
10.Bjoern Werner, DE, Florida State
11.Jarvis Jones, DE/OLB, Georgia
12.Chance Warmack, G, Alabama
13.Barkevious Mingo, DE, LSU
14.Sheldon Richardson, DT, Missouri
15.Cordarrelle Patterson, WR, Tennessee
16.Jonathan Cooper, G, UNC
17.Xavier Rhodes, CB, Florida State
18.Kenny Vaccaro, S, Texas
19.Tavon Austin, WR, West Virginia
20.Desmond Trufant, CB, Washington
21.Alec Ogletree, LB, Georgia
22.Keenan Allen, WR, Cal
23.Matt Barkley, QB, USC
24.D.J. Fluker, OT, Alabama
25.Datone Jones, DT/DE, UCLA
Again, since this aggregate board is a collection of where draft experts think players will get selected, it should be about as accurate of a representation of draft stock as we can get.
Interesting Notes for Dallas
- The chances of getting a top-tier offensive tackle are almost zero. The three elite tackles are all ranked in the top 10. In all likelihood, the Cowboys won’t even be in position to trade up for one. If the Cowboys have their heart set on an offensive tackle in the first round, it might be Alabama’s D.J. Fluker, currently ranked No. 24 on the aggregate board.
- The elite guards are ranked at No. 12 (Chance Warmack) and No. 16 (Jonathan Cooper). That doesn’t mean that both will be off the board when Dallas is on the clock, however, and I still think there’s a decent probability that one will make it to No. 18. If that happens, I think you’ll see the ‘Boys pounce.
- Look who is rated right at No. 18: Texas safety Kenny Vaccaro. I don’t think Vaccaro is worthy of a first-round pick, but lots of people do. Nonetheless, I’d still say the odds of him making it to No. 18—which are probably a coin flip—are lower than for the guards. That’s because there are two elite guards, increasing the overall probability of one falling even though they’re both ranked higher than Vaccaro.
- If the odds of one of the elite guards falling to the Cowboys are 50 percent and the chances of Vaccaro slipping to No. 18 are the same, there would be just a 25 percent chance that Warmack, Cooper, and Vaccaro are all off of the board when the Cowboys pick.