Who should be the number one overall pick? A curious question entering the home stretch, the final two weeks of drafting. This week and last, I’ve examined mainly value- first, players who I felt were over-drafted, being taken well above where I think they should fall. Second, I looked at the opposite side of the coin - players who were well under-drafted and offered serious value at their current ADP. But one question I haven’t tackled is the strategy of the top pick. Can you really have “value” at #1 overall?
I’ve often found the best snake draft strategy is simply to take the best player available at any given moment (similar to the real NFL, eh?), and that applies even to the first pick. So who’s the best available player with the whole board in play?
The Weekly Advantage Argument
For some time, I strongly considered the theory of taking Rob Gronkowski at number 1 overall in 2015. I still consider him a top-10 pick and currently have him at seven. But why not one? And why did I think of grabbing him numero uno? The argument is rather simplistic (and, unfortunately, it is in its simplicity where it ultimately falls short).The difference between Gronk and every other TE on the board is more dramatic than the difference between player 1 at any other position and player 2 at said position. Just think of it logically for a second, position by position, looking at the 2014 totals for 1st and 2nd per position:
- QB: 1. Aaron Rodgers 342 2. Andrew Luck 336, difference: 6 points
- RB: 1. DeMarco Murray 282 2. Le’Veon Bell 272, difference: 10 points
- WR: 1. Antonio Brown 251 2. Demaryius Thomas 223, difference: 28 points
- TE: 1. Rob Gronkowski 178 2. Antonio Gates 148 3. Jimmy Graham 137, difference: 30 points
Keep in mind, Gronk was a little banged up early on and also didn’t play week 17. So, Gates played an extra week, had a freak 3-touchdown day in week 2, and still was outscored by five touchdowns. I also included Jimmy Graham’s (the concensus 2015 number 2!) 2014 numbers -- when he was with Drew Brees, remember-- to illustrate the difference. Gates likely wouldn’t come close to repeating his 2014 even if he hadn’t been suspended four games and Graham is on a much more run-oriented offense. Sketchy math indicates he’s good for a minimum of 2 points-per-game better than the next best TE. Only Antonio Brown comes close to replicating that performance over the next best competition, and there’s a solid argument to be made for at least a tiny regression- or the catching up of the WR field.
So, what’s not to love?
The problem comes in when you actually do it. Not taking a WR or RB until pick 20 (or pick 24 in a 12-team league) typically kills the look of your team. Even if you go RB/WR at 20/21, you don’t get to do it again until the 30s. In general, I’ve found my Gronk-at-1 mock draft teams don’t look very thrilling. Missing out on that early opportunity to grab value really hurts.
Another issue: the weekly advantage theory strangely conflicts with mathematics. Personally, I’m really not sold on the idea of Value-based-drafting (VBD) which is for lack of a better description similar in some ways to real-life-WAR (wins above replacement) in that it seeks to create a measurable “generic” player-x per each position that one can then measure other players against. The problem is, every league is different. And replacement-level vs. performance is not really a good way to evaluate decisions in a fantasy setting where every game/point is vital and the season is remarkably short and often matchup-based. (Though perhaps that’s an interesting argument in Gronk’s favor by itself: he’s matchup proof!)
The Universal Draft Coefficient
A much more scientific idea I have come across was developed by the folks over at Pro Football Focus called the Universal Draft Coefficient, or UDC.
Editors note: UDC is very similar to the value column and next value column within our cheat sheets - we may be able to add the UDC this week since it's only one extra mathematical step (RPPG) over our value number and I feel it makes it easier to spot variations (it's a smaller number) over what we already provide.
What I like about UDC is PFF’s development of a three-pronged axis of sorts to measure a player’s impact compared to every player in the league’s placement on that axis: that is, it’s value measured among the performance of their peers without the pesky semi-arbitrary “baseline” VBD concept. As the folks at PFF explain, the three lines of the axis are “Positional Starting Slots” (many people unwisely apply general rankings to leagues with specific starter rules that would affect the value of each player), “Real Points Per Game” (PPG calculated using games where the player doesn’t play or plays only a partial game as a replacement for an injured player really can affect general understanding of value), and “Positional Starter Average” which helps create a more mathematical arena in which to place and understand a player’s performance. It’s a long read, but check out the article, Gronk, even in leagues that allow TE’s in the flex spot, still only checks in at number 14 (in 2013) and number 20 (in 2014) overall. Strangely here, the numbers also tell a different story in regards to Gronkowski’s performance over other tight ends: Graham’s UDC places him much closer than his point differential would suggest: at number 15 overall, only one spot behind, in 2013 and 25 overall in 2014. Interesting.
So, is the monkey-wrench thrown? Are we kaput for this whole crazy notion?
So you may say, well wait, there’s so few good tight ends! Yeah, but most leagues only play one tight end, and there’s the rub: you have to play two running backs. Now, a look back at the 2014 scoring leaders suddenly tells a different story. The number one RB, Murray, put up 282 fantasy points in 2014. But the #20 RB (who was started in 10-team leagues with two RB slots, mind you!) was Andre Ellington, who put up only 122 fantasy points- that’s a whole ten points per game less than Murray. Suddenly, Gronk’s two-per-game over Gates doesn’t look so impressive! But for fairness, let’s compare Gronkowski only to the number 10 scoring tight end, who would have also been started in our theoretical vacuum 10-team league. J. Witten landed there in 2014, scoring a total of 93 points to Gronkowski’s 178, for an 85 point difference, or about a 5.3-fantasy-point-per-game difference, which still pales in comparison to the 10-per-game difference at the RB position.
Gronkowski NOT The #1 Pick
Ultimately, the numbers just don’t bore out: Gronkowski at one makes sense on some surface levels but it’s not a feasible plan. But this numerical exploration does give rise to one interesting notion: forget snake drafts, in auctions, the “Studs and Duds” method is clearly the way to go! Nabbing the top guy on a points-per-game basis and a replacement level (21st or below in RB rankings, 11th or below in TE) would theoretically score higher than two mid-tier options almost regardless of the positional combination you select. Quite curious!
So Who Do I Love?
Honestly it’s a toss-up. I included Eddie Lacy in my overdrafted list, but after the Jordy Nelson injury, I have to elevate all of my remaining healthy Packers. Le’Veon Bell could be docked for his suspension, but he was lights out when he played in 2015. Adrian Peterson? Oh, he’s now playing for a guy who has coached the NFL’s leading rusher 5 times in Norv Turner, and he’ll likely get all the passing down work, too. That’s my top tier, and Jamaal Charles is a very close 4th. I think an argument could be made for any of the top-three but ask yourself: who’s the one you don’t want the other guy to have in week 17, for the Superbowl game? My gut says it’s Bell, so act accordingly.
Best of luck. As always you can tweet me @pkaragianis
— Petros Karagianis (@pkaragianis) August 25, 2015