Last week, I talked a little about the misuse of what is commonly known as average draft position (ADP). The focus was on how three players, Eric Hosmer, Brett Lawrie and Desmond Jennings have all zoomed up the ADP because there is a desire for many enthusiasts to own the next great thing. What happens is drafters look at the most recent ADP, determining an order and ideally assuring drafting a specific player. Meaning to get the coveted player, they need to jump the ADP a few spots. This cycle repeats itself and some player’s ADP continue to rise and rise as a result. While I have no issue with going after your guy, there comes a point where his reasonable value exceeds the expected value of that draft spot, eliminating any possible return on your investment.
Here is what is meant by expected value of each draft spot. An expected value can be determined for each player, based on that player’s anticipated production. This is really just the value used as a guide in auction formats. If you graph out a draft, using these expected values, you can determine a rough idea of what each draft spot is worth. The names of the players may change from year-to-year, but the distribution of value remains basically the same. One quirk of the data is it is not linear. The difference between consecutive players is largest at the beginning of a draft. This is not unlike the NFL rookie draft where there is a value placed on each pick to facilitate trading. If a team trades down three spots in the first round, it costs more than a three spot difference in another round to balance the deal.
Here is an outline of the pick values for a typical 15-team draft. The value of the first and last pick of each round is presented; consider this value to be +/- $1.
ROUND |
FIRST |
LAST |
1 |
$45 |
$31 |
2 |
$30 |
$24 |
3 |
$24 |
$21 |
4 |
$21 |
$18 |
5 |
$18 |
$17 |
6 |
$17 |
$16 |
7 |
$15 |
$14 |
8 |
$14 |
$13 |
9 |
$13 |
$12 |
10 |
$12 |
$11 |
11 |
$11 |
$11 |
12 |
$10 |
$10 |
13 |
$9 |
$9 |
14 |
$8 |
$8 |
15 |
$9 |
$9 |
16 |
$8 |
$8 |
17 |
$7 |
$7 |
18 |
$6 |
$6 |
19 |
$5 |
$5 |
20 |
$4 |
$4 |
21 |
$3 |
$3 |
22 |
$2 |
$2 |
23 |
$1 |
$1 |
There are other implications of this chart that we will broach in future columns, but for today, the idea is you can estimate the expected return of investment per pick according to the above chart. As an example, a third round pick should return in the neighborhood of $22 or $23. Bringing this back to last week, Hosmer, Lawrie and Jennings are being selected in this range, so they should be expected to return between $21 and $24. For those that play in mixed auctions, would you pay that much for an unproven player? That is a pretty good chunk of change. Our current expected dollar value for each of the three is about $15.
But, there is another fallacy of ADP I want to discuss. There are some who feel a successful draft is one in which you select as many players past their ADP as possible. In fact, some sites score your draft using this as the measuring stick, awarding positive points for every draft spot you pick a player after his ADP and negative points for early picks. Here is the problem: the goal of a draft is not to beat the ADP, but to assemble as much talent as you can based on how YOU feel your selected players will do and how they will contribute intrinsic value to your team. Another way to look at it is the objective is not to beat the ADP, but to amass as much talent that has the potential to return more than the expected value as defined by the above chart.
This is not to say ADP is useless as it can be a tool to help you accomplish the above. The ADP gives you an idea of the market value of some players. You may have a set of players you want. The idea is not to pick the ones that beat the ADP but to use the ADP to help you determine the order of your picks to maximize your chance of getting the biggest return on investment from these picks, relative to the value chart above. Sometimes this entails your drafting a player ranked LOWER on your personal draft board if the ADP says you can likely wait a round to pick someone you rank higher. (Those who completely ignore the ADP may just take the higher ranked player first.)
In terms of dollar value, let’s say you expect Player A to return $15. That means he is “worth” a 7^{th} round pick. However, his ADP is 10^{th} round. Now let us say you also like Player B, ranked a few spots lower on your board. Maybe he is supposed to also return $15, but round off puts Player A higher than Player B. If you take Player B first and then take Player A in the eighth, you get $30 worth of potential value. If you took Player A first, then you would be left with Player C in the next round, who could be valued a buck or two less so your combined value is only $28 or $29. If you do this three or four times throughout the course of the draft, you can amass significantly more potential value than what is dictated by the value chart above. Think of it this way, instead of having $260 to spend at your auction, you have $265-$270. It might not seem like much, but in competitive leagues, every little bit helps. Plus, this is a zero-sum economy. If you effectively have $270 to spend, someone else has $250, or at minimum, that $10 is dispersed amongst several players.
As mentioned, there are several other game theory implications of the value chart that we will talk about down the line. But for now, hopefully it opens up your mind a little to not rely on the ADP as a draft list, but to also not completely ignore it either. ADP is a tool and if used properly, it can help you gain an advantage over your opponents. But if used frivolously or ignored, you can miss out on the opportunity to gain said advantage.