The Rundown: FantasyScore Week 5

Photo: USA Today Sports

The past two weeks, I've detailed how using the zero running back (0RB) strategy in Fantasy Score's Draft-N-Go's (DNGs) affects your projected score in 5-person and 8-person games. Today, I'm going to experiment with another popular re-draft strategy: waiting until the end to select your starting defense.

In 2015, the game-theory reason for saving defenses for last should be common knowledge, but I'll lay out the argument nevertheless. Most importantly, hitting a home run with a defense translates to a far smaller week-to-week advantage than hitting a home run with the backup running backs and wide receivers you select in the middle rounds instead. In addition, the inherent variance in projecting season totals for defenses means that the one you pick in the middle rounds is more likely to underperform the value of the pick in which you selected it. This same two-fold argument applies to kickers as well, especially with respect to the second point.

In DNGs, however, there are no backup running backs and receivers. You're only drafting a starting lineup, and so it stands to reason that the accepted benefit of saving your defense for last may be mitigated by this difference in game structure. Perhaps, it could even be a net negative!

week 5 draft lists 

As always, before I present the results of my experiment, I first need to give you the Week 5 draft lists:

And once again, the assumptions underlying my experiment are as follows:

1. Everyone's using the same draft list.
2. Everyone's using a wide receiver as their flex starter.
3. Everyone else selects the best player available unless they've already filled that player's position in their lineup.
4. Except us. We're using the "defense last" strategy, so we're not taking a defense until the 9th round regardless of "best player available" in our first eight rounds.

the defense-last experiment for 5-person dngs

It turns out that waiting until the last round to select your defense in a 5-person DNG hardly affects projections at all -- at least this week. Because it seems to be so inconsquential, I'm not going to bother detailing the actual draft results. Rather, here's the summary table showing the number of projected points added or subtracted compared to the "baseline" draft when you wait until Round 9 to select your defense:

As was the cast last week, the first row displays the projected point total when everyone drafts perfectly using our draft list, and no one uses the defense-last strategy. From there, the table indicates the increase or decrease in projected points for each team based on which team is the one employing the defense-last strategy. For instance, when Pick 4 selects their defense last, Pick 4 has a projected score that's 0.1 points less than their score when no one goes defense-last, Pick 3 scores 0.1 points more than baseline. Meanwhile, Picks 1, 2, and 5 are projected to score exactly the same as baseline.

From the green table cells, you can see that, when we pick a defense last from our pick, our projected point total hardly changes compared to baseline when we have Picks 1-4. If we have Pick 5, however, we end up sabotaging our own projection. From the yellow cells, we see that doing so also gives Pick 1 an even larger projected victory than they would have otherwise enjoyed.

And in case you're wondering, the main reason why the above table is chock-full of zeroes, it's because the baseline draft for this week includes Pick 1 selecting a defense in the 8th-9th wrap-around, while Picks 2 and 3 both select defenses in the 9th round.

the defense-last plus 0rb experiment

Because that was pretty underwhelming, let's take things a step further and combine the defense-last strategy with the 0RB strategy. That is, let's see how drafting a defense in the last round and drafting four wide receivers in the first four rounds affects our projected point total compared to baseline. Once again, here's the summary table, which you'll notice has far fewer zeroes this time:

Like before, when we're the person selecting a defense last, there are no draft slots in which we increase our projected point total. Unlike before, layering 0RB on top of defense-last makes us worse off at every every draft slot. In addition, and this is unique among all experiments I've presented so far, Pick 1 relinquishes its baseline advantage entirely by using this strategy: When Pick 1 goes defense-last and 0RB, Pick 2 becomes the projected winner.

Simiarly, when you're Pick 5 and you go defense-last and 0RB, not only do you relinquish your baseline advantage over Pick 4, you nearly make Pick 4 projected to cash (i.e., finish in the Top 2). That's because, at least this week, you end up handing them Rob Gronkowski and Tom Brady by selecting Larry Fitzgerald in the second round and Calvin Johnson in the fourth. Furthermore, you also gift them the second-best defense (Baltimore) in the eighth round and give yourself the worst defense (Arizona) by not selecting the Ravens with your 7th-8th wrap-around picks.

One other interesting result worth noting is that Pick 2's projection changes in all instances where anyone goes defense-last and 0RB.

bottom line

Here are the rules of thumb (at least for this week) if you're considering waiting until the end of a 5-person DNG to select your defense and/or considering selecting four wide receivers in your first four picks:

• If you're assigned Pick 5, do not go defense-las and definitely don't go defense-last and 0RB. The former will decrease your already-low chances of winning. The latter will 
• If you're assigned Pick 1, definitely do not go defense-last and 0RB. It wil completely eliminate your inherent advantage.
• If you're assigned Pick 1-4, it doesn't matter if you go defense-last or not.
• If you're assigned Pick 4, pray to your god that Pick 5 doesn't heed the above advice.