I love nerdy spreadsheets. I really do. There's comfort in data, in that numbers don't care about this opinion or that. Used in an intellectually honest manner, numbers allow for reasonable conclusions to be drawn and myths, sometimes popularly held, to be dispelled. As this is NFL Draft season, I've turned my Excel spreadsheet loose on two draft assertions that have become more and more accepted as fact: first round trades are rare, and the NFL Draft Trade Value Chart is obsolete.
It turns out that both are myths.
First, during this time of year, people present all kinds of wild draft possibilities. (I'm more guilty than most.) Some people don't enjoy the speculative nature of the draft trade scenario posts, and have stated as fact that first round trades are rare and/or difficult to pull off. Nonsense. I looked at the last ten drafts, and 117 of 320 (36.6%) first-round draft picks have traded hands. Something that happens about a third of the time (never less than 25% of the time, and at times closer to 50%) cannot be said to be a rare occurrence. Clearly, first round trades aren't difficult to pull off - and there are willing partners to be found. Let's bury this particular myth. Bury it and then bury the shovel.
Second, more and more people are asserting that the trade value chart - developed by Jimmy Johnson during his time in Dallas in the early '90s - is increasingly useless. If that were true, we'd see an increasing deviation from the values found in the chart as the years have rolled on. This, as I will show, is clearly not the case. For better or worse, Johnson's creation is as viable today as it has been for the past decade.
Before I delve into the numbers, feel free to pose your draft trade scenarios with the standard NFL draft chart as the basis for your clever ideas. Then invite anyone who attempts to rain on your chart-based parade of trade ideas, by suggesting that first round trades rarely happen and/or that the chart is worthless, to stop.
I wiped out all trades that involved any significant players. For example, San Diego traded the first overall pick in 2001 (3000) for picks 5 (1700), 67 (255), 2002 48 (420) and Tim Dwight. By the numbers, the Falcons gave up 2425, while the Chargers put up 3000. What value did Tim Dwight have to the Chargers? Did they value him as a late first/early second round pick? I don't have the first clue. As I can't assign a value to Dwight, that represents the value the Chargers (or Atlanta) thought he had; I don't have a real number for what the Falcons gave up.
Compensatory picks were another variable, but one I was able to account for after a fashion. For example, a team might have given up the first pick in the fourth round in a trade. Depending on the year, that pick, taking place after compensatory selections, could range from 97 to 107. I therefore bypassed all compensatory picks in an attempt to get at chart-based numbers. (Compensatory picks can't be traded, and thus have no trade value.) So, the first pick in the fourth round was always counted as selection No. 97 worth 112, pick 186 was worth 17, and so on. I rounded numbers that had decimals either up or down.
I kept track of a particular type of pick: future first rounders that were surrendered during the course of a trade. Other future picks were also tracked, but the most dramatic effect was on the first round values, and the first round was all I looked at anyway.
As an aside, this posting makes no attempt to divine the wisdom of trading up or down. Some teams trade up and down the board quite well. Others, like Buffalo, are dismal. The J.P. Losman trade was particularly bad, with Buffalo getting 780 (22 overall) while giving up 1355 (43, 141 [adjusted due to compensatory picks] and 20 from the next year). If Losman had been the savior of the franchise, I don't think that Buffalo fans would have reason to gripe. As it turned out, he was a bust and cost the Bills dearly.
You will find charts below that contain data for each first round from 2001 through 2010. Here is a bit of explanation of the terms I've used:
- TD picks: Trade Down: The number of picks the team that traded down gave up
- TD value: Trade Down: The value the team that traded down received
- TU picks: Trade Up: The number of picks the team that traded up gave up
- TU picks: Trade Up: The value the team that traded up received
- TD percentage benefit: The percentage of benefit the team that traded down received. A percentage below 100 represents a points benefit for the team trading up, while a percentage over 100 represents a points benefit for the team trading down.
|Year||2001||TD Picks||TD Value||TU Picks||TU Value||TD % Benefit|
|Year||2002||TD Picks||TD Value||TU Picks||TU Value||TD % Benefit|
|Year||2003||TD Picks||TD Value||TU Picks||TU Value||TD % Benefit|
|Year||2004||TD Picks||TD Value||TU PIcks||TU Value||TD % Benefit|
|Year||2005||TD Picks||TD Value||TU PIcks||TU Value||TD % Benefit|
|Year||2006||TD Picks||TD Value||TU PIcks||TU Value||TD % Benefit|
|Year||2007||TD Picks||TD Value||TU PIcks||TU Value||TD % Benefit|
|Year||2008||TD Picks||TD Value||TU PIcks||TU Value||TD % Benefit|
|Year||2009||TD Picks||TD Value||TU PIcks||TU Value||TD % Benefit|
|Year||2010||TD Picks||TD Value||TU PIcks||TU Value||TD % Benefit|
|Condensed||Year||TD Picks||TD Value||TU Picks||TD Value||TD % Benefit|
|Half Decade||Year||TD Picks||TD Value||TU Picks||TU Value||TD % Benefit|
As you can see, for every year but 2002, the team trading down came out ahead - and even in that one case, the team trading down was within 0.6% of breaking even. You notice that in 2001, 2003, 2006, 2008, 2009 and 2010, the team trading down was no more than 5% ahead of the game. In 2004, 2005 and 2007, each featured trades in which a future first rounder was part of the deal. As noted above, those trades disproportionately favor the teams trading down - and 2007 (43.6% net gain for trading down) featured a pair of them.
I ran the numbers a second time and wiped out trades that included a future first rounder. Those numbers (years only - how long did you want this post to be?) are below.
|Year||TD Picks||TD Value||TU Picks||TD Value||TD % Benefit|
|Year||TD Picks||TD Value||TU Picks||TU Value||TD % Benefit|
You can see how the numbers are no longer badly skewed. In this light, you can see that the variance is fairly minor: no more than a 3% gain for teams trading up, and no more than 7.4% overall (with 16.5% for 2007 as an outlier - can you believe Matt Millen actually fleeced anyone?). What this says to me is that, over time, the draft chart does hold up pretty well. Some things stand out:
- It is advantageous, from a points perspective, to move down instead of up. Duh. Draft choices are commodities, and it is absolutely a seller's market - just not outrageously so.
- Getting a team to trade out of the first round is far more expensive than swapping first round picks. Even after factoring out trades involving future first round picks, trades that featured a team dropping out of the first round altogether skewed results a bit for that year. Call it a psychological block that the team trading up has to overcome. Bear that in mind if you're looking at scenarios in which Buffalo trades up to get into the latter stages of the first round. I'd look at a 20% markup on the value of that late first round pick.
- Teams trading down generally give up one pick, while teams trading up in the first almost always give up two or three picks. The one outlier is the five picks Tennessee gave to Houston in 2004 for the No. 27 selection. This goes back to the second point, in that the Titans had to overpay (680-886) to get the Texans to drop out of the first. When looking at your dream trade scenarios, bear in mind that anyone sending picks to Buffalo are exceedingly unlikely to send more than three.
- Snagging a future first rounder is a tremendous boon. Even Dallas, who traded a 22 for a future 22 came out ahead 780 to 1320 in the Brady Quinn trade with Cleveland, thanks to a second rounder. Delayed gratification comes with a heavy price.
- Johnson either hit the nail squarely on the head when he devised the draft value chart, or the league has adopted it. This is crystal clear when the impact of future first round picks are dropped from the equation. Trades average out to within plus or minus 3.1%, with 2005 being the lone outlier at 7.4%. When you consider all the variables at play in draft trades (and remember I dropped trades that included significant players), Johnson's draft chart is astoundingly accurate.
- The draft chart has become more, not less, relevant over the past half decade. From 2001-2005, the teams trading down averaged a 4.2% gain, a number that dropped to 1.0% from 2006-2010. Yeah, that's right. For the past five years, first round trades have been 99% in line with Johnson's creation. It's gotta be the hair.