The NFL Draft by Numbers: Moving up by trading down?

One of the key themes in Raf's 'Cowboys Draft '10' series has been and remains the possibility of the Cowboys trading down. Trade down anybody? and Trade Down City are two posts on the topic in addition to Saturday's Money Makes the Mock go 'Round that you should check out if you missed them.

There are a couple of reasons why teams would consider trading down.

Deep draft: This draft may be one of the deeper ones in recent history, so some teams may be looking to acquire more picks to get more value (than they would in 'normal' years) with their selections.

Cost reasons: Even though this year is uncapped, not all team owners are willing to risk big bucks on unproven prospects - and instead see trading down as an option to minimize financial risk. The performance difference between the No. 1 pick and the No. 5 is arguably not that big, but moving down four spots could save you anywhere from $20-30 million that you can then invest elsewhere. It's no surprise that all four top teams in the draft would like to move down.

More picks: Teams with fewer picks - like the Cowboys - may see trading down as an opportunity to fill more gaps with more picks. Other teams that are starved for talent at many positions may try to fix this in one go in this talent rich draft. 

There are many more reasons for trading down, but regardless of what the reasons are, what all of them have in common is that too often you're trying to get a Mercedes on the sly and end up with a bunch of Hyundais.

You can post-rationalize any trading down scenario as much as you like, but at the end of the day, when you trade down, you are exchanging quality for quantity. Or are you?

Today we'll try and see whether historically there have been draft trends that would indicate that trading down could have yield a higher value. To do this, I needed to find a metric that indicated the total value of a player, irrespective of the position played. I found it in the "Career Approximate Value" (CAV) metric developed by Doug Drinen at Pro-football-reference.com (PFR).

In my own words: Approximate Value is an attempt to assign a value to any player at any position for any given year, by weighting position specific metrics (i.e. yards or points scored/allowed) with an indicator for durability (total games played and seasons as their team's primary starter) and quality (Pro Bowl and All Pro nominations) and then normalizing all this at a team level. Read up on it at PFR if you need to know more.

I then used PFR's outstanding draft search app to look at all first through third round picks from 1980 to 2005. Why this time period? Simply because a quarter century sounded like a nice figure, and even the last draftees from the class of 2005 should be coming to the end of their first contracts now. So here it is, another new metric to keep the stat geeks like me happy.

So what do these numbers mean? First off, the numbers by themselves don't mean much, it's in relation to other Career Approximate Values (CAV) that they become meaningful. Take the first line, tackles. Tackles drafted in the first round have an average CAV of 48.3. Tackles drafted in the second round have a CAV of 29.4, significantly less than the first rounders, third rounders even less. Ok, no surprises there. A first rounder is better on average than a second rounder? Duh.

A further breakdown of the tackle position shows a similar decline of key metrics by draft round:

Again, no surprises. First round picks on average have longer, more successful careers, particularly in positions where the Bust Factors are relatively low.

Which brings us back to the original question, can lower round picks yield higher value? In the table above showing all positions, I have highlighted the second round numbers for quarterbacks and centers in bold. You'll have noticed that these numbers are extraordinarily high, both relative to the first round picks and also to the other positions in the second round.

At first glance you could argue that you could get better value from centers drafted in the second round than from centers drafted in the first, and that a second round quarterback might be the highest value pick in terms of bang for the buck. And that would be true, but...

But you need to keep in mind that averaging those numbers may not always represent the most accurate picture when outstanding individual players can dominate a small sample size. Of the 22 second round QBs, the four top QBs in terms of CAV are Brett Favre (155), Randall Cunningham (107), Boomer Esiason (105) and Drew Brees (94). Similarly, of the 24 second round centers, Kevin Mawae (89) and Dermontti Dawson (83) have a significantly higher CAV than the best first rounder, Jeff Hartings (58).

But then again, in 1991, the Seahawks and the Raiders decided to draft QBs Dan McGwire (CAV: 2) and Todd Marinovich (CAV: 3) respectively in the first round ahead of second rounder Brett Favre.

So it is what it is: On average, you got better value in the second round for a center than in the first, and a second round QB has been a pretty good investment.

Whither goest thou, pick no. 27?

This year's 27th pick presents a unique challenge from a draft point of view for the Cowboys. In normal years, there are usually around 20 first round graded players in a draft class. This year, we think the Cowboys have 25 on their board, which leaves some room for one of them to fall to the 27th.

But just like the example of the center, do the stats show more areas where the Cowboys could (in theory) trade down and still get better value in terms of CAV? Of course.

Because the Cowboys are nominally just out of reach of the first round graded players, I've split the draft classes into the Top 10 picks per round (picks 1-10) the Middle 10 (11-20) and the 'Bottom Ten Or So' of the round (21 upwards, depending on the year), on the hypothesis that 'Bottom Ten or So' would contain a lot more players that were actually graded for the next round. Here are the results:

The draft history of the last 25 years suggests that moving out of the bottom of the first to the top of the second can, on average, provide greater value for some positions.

Historically, this has been true for centers, wide receivers and tackles (and QBs too, but that's not important here today). A trade down for defensive backs and linebackers decreased the value of the pick only marginally, and with an additional third or fourth rounder as a result of the trade, might have had a positive overall value.

But then there is also the cautionary tale of also Guard Steve Wisniewski, 8-time Pro Bowler and 2-time All Pro for the Raiders, and the one that got away .

The Cowboys drafted Wisniewski with the first pick of the 2nd round in 1989 (#29, CAV: 90), and immediately traded his rights and the 1989 sixth round pick (#140-Jeff Francis, CAV: 0) to the Raiders for a 1989 second round pick (#39-Daryl Johnston, CAV: 25), a 1989 third round pick (#68-Rhondy Weston, CAV: 2) and a 1989 fifth round pick (#119-Willis Crockett, CAV: 0).

Finally, and this is only anecdotal as I have no data to back up this statement, but it seems most of the stories you hear about trading down ended up badly for the team trading down.

As you think about the draft positions that the Cowboys are reportedly targeting and compare these with the table above, what do you think are the chances the Cowboys will trade down this year (provided of course they find a trade partner)?