Do Bookends Rule? A Study on the Value of Draft Slots/Positions in a Snake Draft

Building a Fantasy Football Draft Value Chart is the first step in a number of discussions and evaluations. Very early on when I built a prototype of the Ideal Draft Value Chart the question arose as to which draft pick position in a snake draft had the highest fantasy point value. What is the best fantasy football draft position in a Brothers Keeper draft? Assuming you need fantasy points in order to win more frequently (which isn't an assumption at all really) it made sense to ask that if an ideal draft occurred where each fantasy player picked the best available fantasy point scorer for the upcoming year with perfect efficiency one should know which draft pick position was the most valuable. Logically people assume the earliest picks have the most advantage. Writers have even tested this sort of thing out (cf. Lab Test: The Snake Draft by FantasySharks.com; see also their 2011 Draft Slots Analysis).

In ESPN's The Magazine's July 26, 2010 NFL Fantasy Special, Christopher Harris published a 13 point article under the title "So Crazy It's Genius." The 4th point's heading is "Bookends Rule." Here Harris uses fantasy research to determine which draft positions produce winners:

4. BOOKENDS RULE
Does the No. 1 pick really give you an edge? Is the last pick really the worst?
     Yes. And no.
     We studied every fantasy draft from 1999 to 2009 and posed this question: What would happen if we redrafted each year based on average draft position? For instance, before the 2002 season, the top five in ADP were Marshall Faulk, Ahman Green, Shaun Alexander, Kurt Warner and Ricky Williams. So in our '02 redraft, those were the first five players selected. After our redrafts, we calculated how each of the resulting fantasy teams in each of those 10 years stacked up in final points. We did this twice. First we drafted six rounds each year and looked at the actual fantasy points each team accrued. Then, for comparison, we drafted only two rounds per year. The biggest winners were the teams with the first- and second- overall picks. But surprisingly, Team 10 went for the third-highest average point total. In other words, bookends ruled.

Then Harris lists the data for the two experiments in two charts:


A few things to consider when you think of Harris's results. First, the data might be skewed by the fact that ADP is not perfect market efficiency. Second, actual results may not account for savvy trading by those participants or some easy schedule of opponents in the fantasy league. Third, why not use perfect market efficiency instead to calculate the highest fantasy point teams?

So what would be the best draft position according to the Brothers Keeper Fantasy Football Draft Value Chart throughout an entire 15 round draft (using FPOB instead of straight fantasy points)?

2006
Rank   Pos   Total FPOB
1         1       472.93
2         2       388.48
3         3       372.13
4         4       335.1
5         5       332.98
6         6       332.9
7         7       325.07
8         8       318.2
9         9       316.19
10       10     307.85

2007
Rank   Pos   Total FPOB
1         1       419.2
2         2       358.76
3         3       343.65
4         4       323.58
5         6       282.18
6         5       280.31
7         7       278.35
8         8       275.94
9         10     262.97
10       9       262.94

2008
Rank   Pos   Total FPOB
1         1       289.94
2         2       287.61
3         3       265.31
4         4       265.05
5         9       261.09
6         8       260.57
7         7       257.73
8         6       256.97
9         10     256.76
10       5       256.7

2009
Rank   Pos   Total FPOB
1         1       357.88
2         2       301.27
3         3       293.22
4         4       282.01
5         5       276.85
6         6       252.25
7         7       239.85
8         8       236.17
9         10     228.94
10       9       222.63

4 Year Avg
Rank   Pos   Total FPOB
1         1       384.99
2         2       334.03
3         3       318.58
4         4       301.44
5         5       286.71
6         6       281.08
7         7       275.25
8         8       272.72
9         9       265.71
10       10     264.13

If we were to eliminate the negative FPOB, namely those picks after round 9, what would the results look like at the end of the first 9 rounds of a Brothers Keeper draft?

2006
Rank   Pos   9 Rds of FPOB
1         1       555.44
2         2       470.17
3         3       453.28
4         4       414.84
5         6       412.95
6         5       412.13
7         7       402.9
8         9       397.09
9         8       396.82
10       10     390.76

2007
Rank   Pos   9 Rds of FPOB
1        1        514.23
2        2        452.08
3        3        437.95
4        4        417
5        5        371.96
6        6        371.68
7        7        369.13
8        8        364.81
9        10      352.82
10      9        351.79

2008
Rank   Pos   9 Rds of FPOB
1         1       376.1
2         2       373
3         3       352.71
4         4       351.23
5         9       346.12
6         8       345.97
7         5       343.53
8         10     343.24
9         7       342.7
10       6       342.55

2009
Rank   Pos   9 Rds of FPOB
1         1       474.34
2         2       416.76
3         3       408.53
4         4       398.26
5         5       392.75
6         6       367.22
7         7       355.83
8         8       351.42
9         10     343.68
10       9       339.65

4 Year Avg
Rank   Pos   9 Rds of FPOB
1         1       480.03
2         2       428
3         3       413.12
4         4       395.33
5         5       380.09
6         6       373.6
7         7       367.64
8         8       364.76
9         9       358.66
10       10     357.63

The results challenge Harris's research. Obviously this is only theory. Perfect market efficiency is not reality. Savvy fantasy football players are in every draft position in leagues all throughout the internet. A more talented player can be in any seemingly poor draft position and do much better than his/her peers. But in a perfect world, with perfect market efficiency it makes most sense to see that in a 15 round draft with Brothers Keeper's scoring the first position in a draft is the most valuable. This makes sense if you consider the fact that in an odd number of rounds (15 or 9), the 1st draft position team chooses the first pick in a round more times than the team in the 10th draft slot.

For the 10th draft position to win a league's championship that player must overcome greater odds than those who pick earlier. Bookends do not rule in a perfect world. The 10th draft position is not at an advantage as Harris suggests.

This seems to correlate with the history of Brothers Keeper as a result of the current Brothers Keeper draft order rule (see Section II. E: The Keeper League Draft Order). Brothers Keeper has not had a repeat champion. Plus, Brothers Keeper has had a team go from worst to 1st. The draft is both helpful and detrimental to the correct teams from the previous year's final standings and so far the draft order regulations has leveled the playing field for the league in successive years.