After reading the post on All Green Bay Packers about Ted Thompson vs the Trade Chart, my interest on trading for or trading away future draft picks was piqued. Specifically, what is the relative value that teams are placing on future draft picks versus current draft picks, and how could this potentially apply to the Packers? This is actually not very straight forward, as we can't simply look at the actual picks eventually acquired, because at the time of the trade this isn't known. I'll look at a couple of different ways teams could attach a value to these picks and see if any method returns consistent relative values.
For the purpose of this post, I've looked at trades that happened in the 2009, 2010, and 2011 drafts, which are as follows (Overall pick number or future round number in parenthesis):
Seattle (37) to Denver(1st); San Fransisco (43),(111) to Carolina (1st); New England (73) to Jacksonville (2nd); New England (89) to Tennessee (2nd)
New England (89) to Carolina (2nd); Philadelphia (146) to San Diego (159), (5th); Jacksonville (158) to New Orleans (4th); Detroit (220) to Philadelphia (6th); Tampa Bay (225) to Denver (232),(5th)
Cleveland (6) to Atlanta (27),(59),(1st),(4th); New England (28) to New Orleans (56),(1st); New England (92),(125) to Oakland (2nd),(219), Philadelphia (104) to Tampa Bay (116),(4th)
From here on out when I'm posting trade values, instead of listing each trade again, I'll just list the year and use the same order as I've written them above. Cool? Cool.
So at the time of these trades, teams already knew the exact value of some picks (i.e. the ones in the current draft), but there is an implied value to the future picks based on the value gap between the picks swapped in the current draft. These value gaps, based on the NFL draft-pick Value Chart are as follows:
2009: 530, 542, 222.8, 145
2010: 145, 4.2, 29.2, 4.4, 2.9
2011: 562, 320, 174.2, 24
Those 2010 values seem low, but outside of the first trade, all of the trades involved 4th round picks or lower.
So the first (and most conservative way) to assign values to these picks is to assume that the team you are trading with will win the Super Bowl next year, and thus pick last every round. This is the worst-case scenario for the team trading for future draft picks. Using this approach, future picks have values as follows:
2009: 590, 590, 270, 270
2010: 270, 28.4, 44, 15.6, 28.4
2011: 634, 590, 270, 44
Most of these are significantly higher than the value gap, and all of them are greater value than the value gap, which you would expect. Future picks are going to be discounted since you can't use them until next year, so there's an opportunity cost applied to these picks. Kind of a bird in the hand versus two in the bush situation. So, what kind of discount are we talking about for each of these trades? I'm glad you asked hypothetical reader!
2009: 10.2%, 8.1%, 17.5%, 46.3%
2010: 46.3%, 85.2%, 33.6%, 71.5%, 89.8%
2011: 11.4%, 45.8%, 35.5%, 45.5%
I think we need to look at these results a little bit to understand what's happening. 2009 involved two future firsts and two future seconds, and besides the fourth trade (where New England, of course, got close to a 50% discount on a second round pick), these picks have relatively little discount applied to them. Moving on to 2010, we have much bigger discounts, but as I had mentioned before, only that first trade involved anything higher than fourth round future pick, and these relatively small pick values may amplify the discount rate artificially. (Interestingly enough, in both 2009 and 2010, the 89th pick of the draft was traded for a future second rounder by the New England Patriots, but I digress).
However, once we move to 2011, we see a pretty big jump in the discount applied to three out of four future picks. My theory for this is because of the rookie wage scale, teams are more reluctant to trade away future picks because present picks have become more valuable since players are more affordable, thus teams really need to "sweeten the pot" to get suitors to accept future draft picks. While this (and really all of this) data has sample size issues, I would expect this trend to continue into 2012. The outlier is the trade that many people thought was a drastic overpay, which was the Falcons trading away a 2011 first, second, and fourth and a 2012 first and fourth, which is somewhat fascinating to me, as I never would have thought that the present vs. future value would be that close.
I played around with two other methods of valuing future picks. The first one (which, if I were a GM, I probably wouldn't use), which I vaguely recall reading in discussions of why the Rams accepted the Redskins trade for the number two pick, is to take the median pick value for the round (i.e. pick 16). This method is decidedly less conservative than the previous method, but perhaps has the advantage of accuracy as you will only be 16 picks off at most rather than 32. Using this method, we get the following future values and discounts (discounts in parentheses):
2009: 1000 (47%), 1000 (45.8%), 420 (47%), 420 (65.5%)
2010: 420 (65.5%), 34 (87.7%), 70 (58.3%), 22 (80%), 34 (91.5%)!
2011: 1070 (47.5%), 1000 (68%), 420 (58.5%), 70 (65.7%)
Now for my personal favorite method of determining value: take the pick number you would be trading for if it was for the current draft and add 10 to it. If this number is greater than the last pick in the round, use the last pick in the round. For example: For the Seattle (37) to Denver (1st) trade, Denver had pick 12. For this method, Seattle would place the value of pick 22 to the future first rounder. Alternatively, for the Atlanta/Cleveland Julio Jones trade, Cleveland would just assume the future first rounder would be pick 32, since the Falcons picked 27 in 2011. I personally like this method because it still has some conservatism, but it doesn't take it to the extreme and assume that a team that finished in the bottom half will win the Super Bowl next year. Using this method we get the following future values and discounts:
2009: 780 (32.1%%) 590 (8.1%), 400 (44.3%), 270 (46.3%)
2010: 310 (53.2%), 28.4 (85.2%), 44 (33.6%), 15.6 (71.8%), 31.8 (90.9%)
2011: 634 (11.4%), 590 (45.8%), 310 (43.8%), 46 (47.8%)
Based these results, the most consistent method (by standard deviation of the discount rate), is surprisingly (to me) the second method, with the other two method virtually identical.
So how does this apply to the Packers? Let's say Ted Thompson thinks we are two players away from a 19-0 season. And he thinks those two players are Fletcher Cox and Shea McClellin (or, alternatively, Melvin Ingram and Kendall Reyes, which still supports my narrative). The problem with this is that you're talking about a player who will almost undoubtedly be gone in the first half of the first round, and a second player who is likely to be taken in the first half of the second round. The only way to get both of these players is to include at least one future pick.
Since I'm already through the looking glass, let's pose this hypothetical: Cox (or Ingram) has fallen to pick 14, and Thompson decides he wants to pull the trigger. Pick 14 has a value of 1100 and the Packers' pick 28 has a value of 660. That's a value gap of 440 points. Thompson knows he can't give up his second rounder, because he needs two guys, or this just isn't going to work. So he clears his throat, swallows hard, and makes the call: "We'll give you pick 28, pick 124, and NEXT YEARS' FIRST ROUNDER!" Add up those points, and Ted is trading away his 2013 first rounder at a 60.8% discount using future value method two (which is a discount pretty in line with what we saw last year). He then packages his 2012 second and third rounders (total value 432) to move up in the second round to around pick 47 and nabs his second player.
Insane? Probably. But just when we think that we have Ted Thompson figured out, he throws us another curve ball.
EDIT: Just to clear up, my purpose here wasn't to argue that the Packers SHOULD trade a future first round pick in this draft. It was more of an exercise to see what kind of value that future first round pick would have for other teams based on historical precedent, and then posit a within the realm of possibility scenario where it could happen. I think there's almost zero chance that Thompson would do something like this. Almost.