David Does Daily (Fantasy): Week 2 Recap & Learnings

September 19, 2014 - by David Hess

This is the third installment in David’s diary about playing one-day fantasy football for the first time. All other posts are listed in this blog section.

After familiarizing myself with the FanDuel interface in Week 1, I entered a few more tournaments in Week 2. I basically used this week to explore a few free daily fantasy player projection sites, and entered some lineups based on those projections.

Projection Sites I Explored

Here are the sites I checked out this week, all of which were partly chosen because they have projections available that take into account the FanDuel scoring system and player salaries. This allowed me to quickly see which players each site considered the best value picks.

RotoWire — I mentioned this site last week. I like the lineup optimizer they have. It allows for a lot of customization of what players are included/excluded, which makes it easy to quickly come up with alternate lineups designed around specific concepts (e.g. lock in 2 star RB’s then optimize around them, lock in a couple studs from a specific team, then optimize around them, etc.)
FantasyPros — Their projections seem to be based on the consensus of multiple fantasy “experts.” Without diving deeper into historical accuracy, it seems like a “wisdom of the crowd” approach might be a good idea, rather than trusting a single source.
numberFire — Nothing particularly noteworthy about their features, I just wanted some more variety in my projections/entries.

Neither FantasyPros nor numberFire have free automated lineup optimizers, so I tried to “hand-optimize” some lineups using their projections. I didn’t spend too much time on them, so I wouldn’t be surprised if I didn’t find the absolutely optimal choices (according to each site), but I think I probably came pretty close.

Week 2 Results Recap

Bankroll Entering Week 2					$249.60
Day	Contest	Fee	Lineup Notes	Result	Won
Thu	NFL 50/50 League	$2	Rotowire / remove Probable player	80.18 (88 of 100)	--
Thu	NFL 50/50 League	$1	Rotowire / removed TR Pick Unders & ATS fades	107.34 (32 of 100)	$1.80
Thu	NFL 50/50 League	$2	Rotowire / removed TR Picks (remembered about D)	84.34 (76 of 100)	--
Thu	NFL 50/50 League	$2	Fantasy Pros	93.48 (57 of 100)	--
Thu	$20K Thu NFL Squib	$1	Fantasy Pros / only players with price drops	89.64 (12595 of 21753)	--
Thu	$15K Thu NFL Snap #2	$2	Fantasy Pros alternate	105.74 (2726 of 8620)	--
Sun	NFL 50/50 League	$2	NumberFire	108.84 (39 of 100)	$3.60
Sun	$5K Sun NFL Dive #4	$1	Stack on WAS vs JAX (injuries)	58.42 (5624 of 5746)	--
Sun	$5K Sun NFL Dive #4	$1	technical error, couldn't submit real lineup	27.58 (5746 of 5746)	--
Mon	$5K Mon NFL Snap #4 (Mon-Thu)	$2	Stack on PHI & ATL	160.34 (152 of 2873)	$6.00
Mon	$5K Mon NFL Snap #4 (Mon-Thu)	$2	Stack on IND & TB	116.48 (2051 of 2873)	--
Total Net Winnings					-$6.60
Sign Up Bonuses Earned					$0.88
Bankroll After Week 2 (ignoring fees for tournaments in progress)					$243.88

Results this week were not great, but there were some obvious extenuating circumstances for some of those poor showings, which I’ll discuss below.

Also, check out the last two tournaments listed in the table. I noticed on Monday that there are contests on FanDuel where you can only pick players from Monday and Thursday. That’s only two games, so in theory that shouldn’t be a great place to find an edge — the fewer choices there are, the less likely it is for a bad opponent to make a bad decision that you can take advantage of.

However, I thought it might be interesting to try a double-stack strategy here. If I managed to stack on two teams that did well, it seemed like I’d have a good chance of placing in the money. I entered one team stacked on the Eagles and Falcons, and another stacked on the Buccaneers and Colts.

As you might expect, the former did a little better than the latter — good enough to nab me 152nd place out of close to 3,000 entrants. Of course, that’s only a payout of 3x my entry fee, so the net winnings from my two lineups was only +$2. Still, I think I may explore this type of tourney more.

The TeamRankings Private Tournament

I also entered the TeamRankings private tournament using the screen name teamrankings, as opposed to my personal account, tr_david. I finished in 60th place there, in a 100-person 50/50 contest (where the top 50% of entrants double their money). My strategy this week was the same as in Week 1, when I finished 5th, but in retrospect I noticed a possible shortcoming of the lineup.

Because I followed a strict heuristic (i.e. remove players from teams where we had a strong Under pick or where we we strongly favoring their opponent against the spread), I ended up with three Eagles players.

It seems like that degree of “stacking” might be bad for a 50/50 tournament. If the Indy defense had a strong night, that could have ruined a third of my lineup. In a 50/50 contest, I would guess that you don’t want to place so many eggs in one basket.

What I (Think I) Learned In Week 2

As mentioned above, a couple of my negative results have built-in excuses for poor performance. Hopefully I can learn from these:

Injuries (and Suspensions) Suck!

I’m not trying to make excuses, as injuries and suspensions are simply something I’ll have to deal with if I want to continue to play daily fantasy. Just like grocery stores have to plan for a certain percentage of their merchandise to be stolen or to spoil on the shelf, I’m sure daily fantasy players need to be prepared for otherwise great lineups to get submarined occasionally by injuries. Still, here’s a recap of how my lineups were hurt this week:

One of my big-tournament lineups was stacked on Washington in their game against Jacksonville. It featured Robert Griffin and DeSean Jackson, both of whom were injured in the first quarter.
Two of my six Thursday lineups featured Adrian Peterson. Getting zero points from an $8800 player is not ideal.
Another lineup used Jamaal Charles. His injury meant I got only 1.7 points from a $9100 player.

Injuries (and Suspensions) Rule!

I didn’t think to take advantage of this, but I noticed the winner of our TeamRankings private tournament used Matt Asiata in his lineup. Who is Matt Asiata, you ask? He’s Adrian Peterson’s replacement on the Vikings, and he contributed 16.9 points from a $4500 salary. This week, he’s up to $6200. Backups for players who get injured or suspended after the week’s salaries are already set seem like great options to plug in.

Don’t Wait Until The Very Last Second

FanDuel experienced some severe site slowdowns Sunday morning — so severe that I was unable to change a placeholder lineup that I had entered in a tournament. I had reserved a space in a tournament by entering a lineup of Peyton Manning plus a bunch of $4500 scrubs, and planned on changing the lineup before the tournament started.

However, the site glitches prevented me from submitting my new lineup … and that’s how I managed to finish in last place out of 5,746 people. Seriously.

Projections Beat My Theories

My two cashes in Sunday contests this week were from:

A lineup where I simply used numberFire’s projections to get the highest projected score possible
A lineup where I removed players from our Under picks and 3-star point spread pick fades, and then let RotoWire optimize the remaining players

I tried a couple lineups where I came up with new “angles”, and those did not go so well:

In a big tourney, I stacked on Washington, thinking that the Jaguars were worse than their Week 1 score had indicated. This seems to have been true, but the lineup still did terribly, even outside the injuries to Griffin and Jackson.
I thought that perhaps players whose prices dropped from the week before could be good bargains, so I made a FantasyPros-projection-based lineup using only those players (whom I identified using RotoGrinders’ Market Watch tool). This was the worst of my three FantasyPros based lineups.

So the moral here is that I still don’t know what I’m doing, but I’m at least a bit more experienced.