East Coast Offense: Draft Strategy 2.0

This article is part of our East Coast Offense series.

Draft Strategy 2017

I've written extensively about draft strategy in the past, and for the most part what I wrote is still relevant in today's game. I want to augment that piece here with a few under-emphasized points and concepts.

1. Bench Size

The size of your bench has a surprisingly large effect on player valuation and roster construction. If you have a three or four-man bench, the value of suspended players like Doug Martin and even Ezekiel Elliott takes a big hit. While you're waiting for them to return, you'll likely get other injuries, and in Elliott's case, you'll have to deal with bye weeks as well. Moreover, the teams with extra bench space can cycle through the back end of their rosters, adjusting personnel based on in-season developments. You'll be forced to stand pat or use your active roster for experimental pickups, costing you points.

In leagues with large benches like the NFFC (10 players), suspended players have more value as you'll still be able to make speculative pick-ups, and you won't have tough decisions regarding taking a zero or dropping a valuable player.

There are other more subtle considerations like drafting pure backups with little immediate value but major upside, e.g. Derrick Henry, D'Onta Foreman and DeAndre Washington) vs. players with modest upside but serviceable value in the short term, e.g., in PPR Shane Vereen, Theo Riddick and Charles Sims. With a small bench, you'll want to draft the pass catchers who can get you enough points if your top RBs get hurt. You can't afford to use a pure backup in a starting spot, so you'll likely have to drop the Henrys and Foremans early in the year. If DeMarco Murray or Lamar Miller go down in Week 8, it'll likely benefit another team that scooped them from the waiver wire.

In a league with a large bench, you can draft the Henrys and Foremans more aggressively, and if your top backs get hurt, you can always sub in some cheap PPR guys without dropping your highest upside backups.

Finally, if you're investing in a premium kicker, defense QB or TE, it helps for their bye weeks to be later in the year when your league has a shallow bench. It's awfully tough to pick up a second kicker in Week 5, when you have short-term injuries to key players and only a few bench spots with which to work. Fortunately Justin Tucker's bye is Week 10, and Stephen Gostkowski's Week 9, but if you're drafting Dan Bailey, realize you're probably dropping him during Dallas' Week 6 bye if you have a small bench.

2. One RB?

The strategy of going without RBs in the first seven rounds or more was popularized by Shawn Siegele. It capitalizes on the fact that RBs are more likely to get injured than WRs, so not only do you invest your most valuable picks on more stable commodities, but your team actually gets stronger as the season goes on and others' valuable investments crash and burn. Moreover, because your top picks are so WR-heavy, you have more backup RBs than most teams, some of whom can become starters when inevitable injuries strike.

It's an excellent strategy for an environment where people are still largely drafting RBs early because you get discounts on top WR, and you'll get good backup RBs when the RB-heavy drafters fill their squads with WR in the middle rounds. Unfortunately, the market has caught up with this strategy somewhat, and WR prices are getting pushed up in keeping with their lower risk profiles as well as the league's pass-happy tendencies. So if you really go zero-RB, not only will you have more competition for the top WR, but you'll also have many other teams competing for the choice sleeper RBs in the middle rounds. Zero RB can still work, but it's far stronger when no one else is doing it.

A modified version I prefer in 2017 is 1-RB, i.e., you use one premium pick on an RB – either an elite one like Le'Veon Bell or David Johnson at the top of the draft – or one that slips more than he should, e.g., I got Todd Gurley in mid-Round 3 of an NFFC draft in July. Otherwise, you go WR-heavy with your other top four or five picks, and you chase RBs in the middle rounds like a zero-RB drafter would, only you don't need two to hit, but just one. You're still competing with other teams who are doing the same, but your desperation level is far lower, and you can use one PPR specialist as a placeholder, while you speculate on your eventual starter. In the Gurley league, I got Duke Johnson for example in Round 9.

3. Ignore ADP

Peter Schoenke wrote a strong piece articulating the perils of drafting solely by ADP, but I'll take it one step further:

If you're an experienced drafter, you're actually better off not knowing ADP at all.

For the novice, ADP is essential because you'll wind up drafting a QB in the second round or a kicker in the 10th. You need to know how the market values players at their respective positions in order not to overpay egregiously.

But for the knowledgeable drafter, knowing where everyone slots according to the market is more a curse than a blessing. For starters, you'll more or less know who the top-two rounders are. If you're picking fifth and choosing between Odell Beckham and Mike Evans, you don't need to look up their ADPs to know whether the one you don't pick will be there on the way back. Or if you like Beckham and Doug Baldwin, it's pretty obvious which one is likely to be there in Round 2 and which isn't. It's obvious.

In the middle and late rounds, ADP is far less stable – it's hard to say whether John Brown will go in Round 9 or Round 12. And it doesn't matter much as so many of those picks are busts anyway, and the difference in auction dollars between a ninth- and 12th-type player is a buck or two, i.e., you should be taking the player you want at that point anyway and not worrying about ADP.

So the primary benefit to knowing ADP is in Rounds 3-6 or so where the cost difference between players is still somewhat significant, and it's not entirely obvious just from knowing the player pool where particular ones would slot in that range. If you know that, for example, Isaiah Crowell is a third rounder, you might not grab him over Joe Mixon whom you might still nab in the fourth.

But even that knowledge is overrated because while if you like both equally, you might be able to land both by picking them in the right order, you also might get swiped by one pick on Mixon in the fourth anyway. Moreover, if you took Mixon in the third, the non-Crowell player you get in the fourth might turn out to be more useful than either. The value gained by having a chance at drafting both of your preferred players depends on the percentage chance that it actually happens and even if it does the percentage chance you're correct that Crowell is appreciably better than the substitute fourth rounder you get. We're talking about small EV differences.

Compare that to the downside of knowing ADP, which is the influence it has over you. How many times have you waited an extra round on a player you loved, assuming he'd be there for you half a round ahead of his market ADP, and someone else – with more courage or simply less knowledge of the market – reaches for him instead, and you're stuck taking either someone who's even more of a reach or a "value pick" you personally don't even like.

Without your conscious knowledge, ADP sets the boundaries for where you can credibly justify taking a player, and over time erodes your ability to evaluate that player except within the acceptable range. You're no longer allowed to go anywhere your observations take you. You're no longer allowed to make bold picks or major mistakes. In the short term, that could end up mitigating downside risk (though many ADP-friendly teams will fail all the same) but long haul, you need those bold picks and concomitant mistakes to refine your process. Your personal observations only translate to actionable draft wisdom by being tested and improved over time. An ADP-compliant process short-circuits that mechanism and keeps you mired in mediocrity forever. And why are you even playing fantasy football if you're not trusting your personal observations and predictions? Don't you want to test your football knowledge and analytical skills against that of your rivals? What's the fun in drafting someone else's players?

Finally, it's important not only to question the effect of knowing ADP on your process, but also the value of it as a predictive tool at all. One argument in favor of ADP is the average of many informed forecasters is better than any particular single forecaster because the aggregation of forecasts irons out individual biases. This theory is known as "The Wisdom of Crowds.," and it holds true under certain conditions, namely that the forecasters come to their conclusions independently. If 100 fantasy experts did independent research to generate their rankings, it would indeed be difficult to beat them.

But that's not remotely what's happening to generate ADP. Not only are drafters aware of other people's rankings, but they're aware of ADP during their drafts and actually use it to gauge value. I'm drafting player X in Round 3 because so far Player X is a Round 3 player. This isn't the Wisdom of Crowds, it's groupthink! Groupthink is worse than individual forecasts because instead of ironing out biases it reinforces them. (See, for instance, the 2008 stock and real estate market crashes where everyone thought home prices would never fall. That particular instance of groupthink (and greed) destroyed a huge portion of the world's economy.)

Bottom line, if you're an experienced drafter who knows the player pool, you're better off not knowing ADP. It's important to make bold picks as well as your own mistakes from which you can learn and improve.

4. Don't Be Dogmatic

People learn rules of thumb that prove useful like "wait on quarterbacks" and "don't draft a kicker or defense until the last couple rounds." For the novice player, these are good practices, as it prevents unforced errors and makes it more likely you come out of the draft with a competitive team. But as you get more experience it's important not to mistake rules of thumb for immutable laws. I put out a poll on Twitter last year,

If it were public knowledge Justin Tucker would kick 7 55-yard FGs every week, where would you draft him?

— Christopher Liss (@Chris_Liss) September 6, 2016

Of course, these things will never happen, and even if they did, we wouldn't know it in advance, but the example illustrates the point that taking kickers and quarterbacks late is something we've done simply because that's the way the game has played out the last couple decades. There's nothing inherently good or bad about getting your points from RB/WR instead of QB/K. The example is particularly relevant this year because Justin Tucker hit 10 50-yard FGs last year, and is quite likely the best kicker of all time. Moreover, his team knows it and tries far more long FG attempts than the average one. As a result, in formats like the NFFC that award five points for a 50-yarder, Tucker should go well before the final two rounds – I'd be willing to take him as early as the 12th round.

Bottom line, don't be dogmatic about these rules of thumb. It used to be that going WR-WR to start your draft violated a key one, namely that you had to get a top RB in the first two rounds. That's plainly no longer the case in the modern NFL. The best drafters are always looking at new possibilities of which people mired in old-school thinking are not yet aware. Be willing to ditch outdated heuristics when new developments warrant. The earlier you do so, the bigger your edge will be.