Home > FNQB, NFL > FNQB: How Much $ is a Draft Pick Worth? Part II

FNQB: How Much $ is a Draft Pick Worth? Part II

[picapp align=”center” wrap=”false” link=”term=gerald+mccoy&iid=8605576″ src=”http://view4.picapp.com/pictures.photo/image/8605576/2010-nfl-draft-round/2010-nfl-draft-round.jpg?size=500&imageId=8605576″ width=”500″ height=”332″ /]

Last Friday, I was able to draw up a pretty good estimate of the market premium spent in free agency to overbid other teams for elite players versus what the same player would make either resigning with his current team, or forcing a trade in which another contender pays draft compensation for his exclusive negotiating rights for a year (i.e. the franchise tender).  I estimated that number at about $8 million in “first three years” money.

What percentage of the total guaranteed money that is really depends on the size of the total contract, which depends on the quality of the player.  Giants DL Chris Canty, who was a target of the Redskins prior to landing Haynesworth, got $22 million of his $42 million in the first three years of his contract, which is a very player friendly deal for a guy who would be best described as a second-tier NFL star.  If there had been no bidding by any of those teams, but rather a trade by the Redskins for the Cowboys’ end (work with me) for a 2nd and a 5th rounder, Canty isn’t getting $22 million in the first three years.  He’s probably getting $14 to $16 million.  His asking price ups from $5 million a season to $8 million a year on the open market with multiple suitors.

One way or another, it’s those valuable draft picks that allow teams to save that money.  The NFL salary cap long prevented franchises from both holding on to their draft picks and using them, and spending mega bucks to outbid the wealthiest teams for the best players.  The Redskins have long tried to leverage their draft choices into cheaper contracts for veteran help, and Haynesworth was a very different route when the team was trying to protect it’s picks after trading for Jason Taylor.

Today, I’m trying to determine the true value of those picks.  It’s established that picks, which all have an inherent expected value in the players they return, can be traded to save money on other contracts.  The tricky part is that the teams that save the most money are the ones that accrue the most picks.  With the exception of the top 5 or 6 picks in the NFL draft, the cost of paying a player’s first contract via the draft is far less expensive than paying a comparable veteran to take that roster spot and lineup spot.  Even in the top five, the expected value of the pick is greater than the expected value of the contract, although the success stories are few and far between, and teams have to be smart — pitfalls in the first five picks of the draft are both well documented and incredibly damaging to the long term welfare of a franchise.

Those are two ways that draft picks provide value.  The third way is through the variance of the players picked: teams that can consistently outdraft the expected value have more value in draft picks, those that are consistently outdrafted by the average have less value.  This analysis will ignore this variable.  By taking the monetary value difference between the contract for a draft pick and a contract for a player who provides the same expected value, and adding the amount saved by trading the pick for a veteran player, while subtracting the difference between those two uses — as not to double count the value — we’ll come to an acceptable “value of a draft pick” figure.  Part III of this analysis will attempt to differentiate value by location in the draft.  Right now, we will concern ourselves with the value of a generic, first round draft choice.[picapp align=”center” wrap=”false” link=”term=chris+canty&iid=7329715″ src=”http://view1.picapp.com/pictures.photo/image/7329715/dallas-cowboys-marion/dallas-cowboys-marion.jpg?size=500&imageId=7329715″ width=”380″ height=”274″ /]

Note: in the absence of freely available contract information, I am going to substitute money intended for the first three years of a rookie contract with total guaranteed money prorated to 3 years.  In theory, this will not throw off anything I am trying to do, but since all guaranteed money is paid in the first three years of the deal, this is going to undervalue the rookie contracts by a couple million dollars per player per first three years.

In 2009, the average guarantee for a first round rookie was $14.4 million, which I’ll round up to $15 million to both simplify this analysis and make it easy to account for the non-guaranteed base salaries that become practical guarantees because they are covered by the release costs associated with dumping a player within his first two years.  Teams that keep their first round pick expect to pay, on average, $15 million — or $5 per year based on the three year principle — in a range between $40 million and $6 million based on how they perform the prior season.  But the expected value for a generic first round pick is a lot more than $15 million.  Win shares analysis in football is still in it’s infancy, but it’s generally accepted that a replacement level team and a playoff team are about 7 wins apart in a season, while the very, very best players in the league at premium positions are worth about 4 to 5 wins in a season.

There’s a lot of variance in those first round picks, but the general expectation is that a team that uses a first round pick will get a player who develops into more than a win-per-year-player.  Not all players drafted in the first round reach that level of play, in fact, more 1st round picks will never reach that level than those who will.  For every no. 3  overall pick that becomes a 3 win/year player — such as Larry Fitzgerald or Andre Johnson — there’s three Joey Harringtons (or Andre Wadsworth, or Akili Smith if you wish), and a Braylon Edwards, Vince Young, or Gerrard Warren type every other year.  It’s not uncommon to end up with Jim Everett or Garrison Hearst when you spend that premium first round pick on a position player, and such results are merely the expectation, not a failure.

Trade Value of a First Round Pick

Teams who have dealt their first round picks before draft day have done so to receive the following pieces:

  • The Raiders dealt an advanced first rounder for the last year of Richard Seymour’s contract/prime, and then his negotiating rights in 2010 (franchise tender).
  • The Bears gave up first rounders in 2009 and 2010 for QB Jay Cutler.
  • The Vikings gave up a first (as well as a third) for DE Jared Allen.
  • The Falcons traded a first round pick for DE John Abraham after trading down from 15th to 29th and acquiring two midround picks in 2006.
  • The Raiders were able to turn Doug Jolley and Phillip Buchanon into a 2005 first round pick from the Jets, by first getting a second rounder for Buchanon from Houston, and then trading that pick and a couple of 6th rounders for the Jets first rounder and a 7th.
  • The Giants traded an advanced first round pick (along with a 2004 third rounder and the rights to QB Philip Rivers) for the rights to QB Eli Manning.
  • The Redskins signed Laverneus Coles to a free agent contract in 2003 as a restricted free agent, causing them to forfeit their first round draft choice.
  • Atlanta gave up their first round pick in 2003 for WR Peerless Price.
  • The Bills gave the Patriots a first round pick forwarded one year (2003) for QB Drew Bledsoe in 2002.
  • Miami gave up first round picks in 2002 and 2003 for Ricky Williams of the Saints.
  • Prior to the 2002 NFL expansion draft, the Tampa Bay Buccaneers sent first round picks in 2002 and 2003, as well as a second round pick in 2002 and 2004 to buy out and acquire the rights to Oakland head coach Jon Gruden.
  • Dallas traded first round picks in 2000 and 2001 to Seattle to acquire WR Joey Galloway.
  • Kansas City traded a first round pick (plus a second round pick) in 2001 for QB Trent Green of St. Louis, and the Rams 5th round pick that year.

The Gruden trade and the Manning-Rivers deals were both highly speculative and useless for these purposes, leaving 11 examples of a first round pick being dealt for player value in the last 10 years.  Now we’ll need a way to establish what was received for their troubles.  Since I’ve been using three-year contract value as a measure of how contracts are valued, it only makes sense to look at player performance for three years after the trade.  This means I’ll have to do some sort of projection for Richard Seymour, Jay Cutler, and Jared Allen.

My metric of choice will be Win Percentage Added, and I’ll use the Advanced NFL Stats blog as my resource.  WPA is a purely descriptive statistic, which means that while it’s quantifying exactly WHAT happened, it’s not really evaluating performance.  Defensive players can’t really record negative WPA, so all players on that side of the ball “add value” and there’s no way to separate a useless player who plays a lot from one who does not play any.  This analysis doesn’t really care, because when you deal a draft choice for a veteran, there’s not going to be bias against playing time, if anything, coaches are biased towards playing the acquisition.

This is the win/year figures for the first three years of the players acquired in the trade:

  1. Trent Green 1.85 WPA/year
  2. Jared Allen 1.73 WPA/year
  3. Jay Cutler 1.66 WPA/year
  4. John Abraham 1.33 WPA/year
  5. Richard Seymour 0.87 WPA/year
  6. Joey Galloway 0.77 WPA/year
  7. Lav Coles 0.12 WPA/year
  8. Drew Bledsoe 0.00 WPA/year
  9. Peerless Price -0.08 WPA/year
  10. Ricky Williams -0.42 WPA/year

The average player acquired in a trade for a first round pick in the last decade has produced 0.78 wins per season above average.  Three of the players were acquired for two first round picks, and five of the first round picks were of an advanced nature (year after upcoming draft).  If we discount those picks at half, a first round pick gets 0.75 wins per season in historical return.  If we discount them at 3/4ths, a first round pick returns 0.67 wins/season for three years.

However, that figure is weighed down significantly by foolish trades that were made more than a decade ago, when first round trades were more common.  Trades made within the last five years for first round picks have been a lot more even, and have returned 1.1 WPA/year over three years.  You probably know that no RFA with a first round tender has been signed since Coles seven years ago, but can you imagine getting a first round pick for Eddie Royal, Dwayne Bowe, or Robert Meachem in a trade today?  That’s essentially what Buffalo did in trading Peerless Price.  That would never happen today.

It’s my belief that teams are so careful with their first round picks that you’re never going to get one in return —  today, at least — without giving up a player who is better than a win/year, and beyond that level, it’s going to cost additional compensation.  This means that first round picks in the upcoming draft are valued at about the level of a win by most teams.  Knowing that, how can we adjust that into dollars?

Trading a first round pick saves contract dollars: Jared Allen might have required money in excess of Haynesworth if he had become a free agent.  He came up about $11 million short in guaranteed money, though he’s likely to get more out of his Vikings deal than Haynesworth will get from the Redskins by virtue of playing for the team deeper into the contract.  That pales in comparison to what Chicago saved on Cutler by acquiring him with three years left on his rookie deal, because when they extended him they agreed on a raise to $10 million a year, which is close to David Garrard and Matt Cassel money.  It’s just a fraction less than Tony Romo got for his extension in 2007.  Ben Roethlisberger, Philip Rivers, Eli Manning, Peyton Manning, and Tom Brady lead the NFL in annual QB salary, most getting between $13 and $18 million a year.  Keeping Cutler out of that category until at least 2013 saves the Bears more than $15 million.

But if a team can expect to have to guarantee an average of $14 million dollars to the rookie contract of it’s first round draft pick, that’s also money we have to consider saved when a team opts to trade the pick.  The problem is that it’s really not money saved.  $25 million in guaranteed money is what a team can pocket when it opts to deal a first round pick for elite talent, but depending on how elite the player is, it’s expected that the money spent to extend the contract (again first three years) will exceed that figure.  So then, in effect, trading the pick would be opting to spend money.

Because the expected win value for trading the pick is about 0.2 WPA/year greater than keeping it — assuming present trends hold — 0.85 EV/yr for spending a first round pick compared to 1.05 WPA/year for trading it for talent, we need only to calculate the money saved figure in the trade against the value added of the pick to get our answer.  My best estimate of a marginal win in football is roughly $9.6 million per, or about twice the rate as in baseball (fewer games, but a much higher replacement level).

Graph Courtesy Advanced NFL Stats, Research by Massey/Thaler

This chart demonstrates the average performance of draft choices in terms of millions of dollars of performance, which is a bit outdated (data from 2000-02), but demonstrates my point.  Teams who trade their first round pick are costing themselves a yearly value of draft surplus of (more recently) between $1.2 and $0.4 million a year, though that number might very well be negative in the rare case of the first overall pick.  Pretty much all draft picks fall somewhere in that range, however, up until the fifth round.  So if that’s the monetary value of a first round pick, $0.8 million/year when splitting the difference for a first round pick

Conclusion

Using the $8 million dollar proration found in Part I, and subtracting the value of three years of that surplus, the actual monetary value of a first round draft choice, independent of the player selected with it, is estimated, per this analysis, at $5.6 million.  To that, we need only to add the value of the expected wins added by a draft choice over the life of the contract, no matter how it’s used.  So 5.1 (average contract length) times average WPA provided annually (0.85) is equivalent to 4.32 expected wins, which is another $41.5 million for the player it is used to select, and then from that, we must subtract the value of the expected signing bonus paid to a first round draft choice, $14.4 million

  • $41.5+$5.6-$14.4 = $33.7 million = the total open market value of a first round pick.

In other words, that number is the price a team should charge another team to purchase it’s first round pick in the next draft.  Yeah, you can see why teams are holding onto those early draft choices.

Next week, I will try to scale this analysis to other parts of the draft to see what non-first round selections might be worth, according to my process.

****  ****  ****  ****

Formula

(To recap):  Expected three year non-prorated market premium, minus the expected financial surplus from the performance of a draft pick against a comparable veteran talent.  In finance terms: market premium – risk premium.  (the draft represents an inherent risk, but taking that risk creates expected return in dollars).  That’s the speculative market component of value, which your franchise has up until the point where you no longer have the pick (be it used or traded).

Then you have the actual player value in wins, which is equal to all the wins you would expect your money to buy.  On the open market, it costs $9.6 million dollars to buy a one win player for one year.  A draft pick in the first round offers 0.85 wins per year, and an average contract length of 5.09 years, both per 2009.  That’s where $41.5 million comes from.  When you trade a first round pick, you forfeit that money because you no longer have the pick.

But you save the average signing bonus paid to first round choices, $14.4 million — the cost of using (signing) the pick.  So that’s subtracted.  In practice, this number is a lot higher than $14.4 million, that’s the average guarantee, but no player actually gets cut on a hunch that he’s so bad as to waste money paid to him before he plays a down.  I’ll try to narrow that figure down at a future date.

All figures are accurate only as of 2009.

Advertisements
  1. June 14, 2011 at 9:30 pm

    Just a FYI, but my study here:

    http://wp.me/p1m41i-7v

    tends to correlate with the kinds of ideas suggested by the Massey-Thaler study.

    Note that while total draft picks correlate with winning, first round draft picks do not.

    I need to turn these ideas into a post.

    David (yea, at it again) .. http://codeandfootball.wordpress.com

  1. June 26, 2010 at 7:39 pm
  2. July 16, 2010 at 11:59 pm
  3. April 5, 2011 at 2:48 pm
  4. June 15, 2011 at 9:40 am

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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 )

Google+ photo

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

Connecting to %s

%d bloggers like this: