College Football DFS Diary – 12 September 2015

Last Week

Tournaments

Finances – entered five contests, risking $0.25 in each one, for a total of $1.25. Placed in three contests, earning $1.75, for a $0.50 profit.

Roster Construction – From a macro-perspective, I did the high-variance thing in all five of my lineups, stacking both quarterbacks with one of their receivers (in the WR slots, and not the flex). Where I went wrong, tactically, was to deploy WRs in only two of my 10 total flex positions. I have seen research before that WRs are the highest variance players, for better or worse, and in a tournament setting, I need more of that. This week I will committ to using only WRs in my flex position.

Stacks – Of the ten stacks I deployed (five contests, two QB/WR stacks in each) the pairs only connected for touchdowns on four occassions. I must hit more of these.

Player Selection – From the Thursday night slate, Ralph Webb and David Ungerer were total duds. On Saturday, I really feel like I made some quality plays, returning tremendous value from little-known assets like Greg Powell (9.6X), Brandon Ross (5.7X) Jenson Stoshak (6.2X) and Ian Sadler (5.1X). All were owned in less than 6% of leagues.

$/pt – when I refer to Ian Sadler and 5.1X, what I mean there is that he returned 5.1 points for every thousand dollars of salary spent. 24.3 points divided by a 4.7k budget equals a multiple of 5.1. I am kind of fixated on this notion because it helps me understand the big picture of what I am trying to score. Every week I have $50k with which to construct a lineup. Last week, to win the tournament, I would have needed to score about 280 points. 280/50 is 5.6, so basically every player on my team needs to score 5.6 times his salary to win. Here is what my multiples looked like. in aggregate, for week 1 tournaments.

QB – 3.75

RB – 3.21

WR – 3.13

FLEX – 3.02

Multipliers

Finances – entered two contests for $1 each, one on Thursday night and one on Saturday night, and failed to place in either. In total, I lost $2 in this format this week.

Roster Construction – Of my four possible stack opportunities, I only utilized three, leaving Auburns Jeremy Johnson to fend for himself. I had one RB-WR pairing and one RB-RB pairing. I think what I am going to try this week for multipliers is to have my flex positions consist of RB-WR pairings, for a little more balance of floor and upside.

Player Selection – With only two lineups, there is not much to analyze. I will say that I am frustrated for spending so much on Marquise Williams on Thursday. Anecdotally, I feel like anytime you take a non-SEC quarterback who is playing an SEC defense in week one, they will more often than not disappoint. I even said that to myself ahead of time, but still pulled the trigger. Elsewhere, I feel pretty stupid for my Algernon Brown play. I dont know if there was enough evidence pointing toward him, and with more production from that spot I think I could have come close to placing.

Stacks – Of the three I played, only Dane Evans – Keevan Lucas connected… and they connected big time.

$/pt results –

QB – 2.33

RB – 1.89

WR – 2.80

FLEX – 1.88

50 50s

Finances – On Thursday I played three games, each for a dollar and lost a total of $3. On Saturday, I played in four games and placed in two, breaking even for the day. In total, I lost $3 in this format for the week.

Roster Construction – Looking back, I am surprised to see that I only stacked my QBs with one of their WRs on 7 of 14 possible chances. I am not sure if that is a good or bad thing for 50-50s, since floor is my main goal rather than volatility. If you have any thoughts or research on this, leave a comment. Of the three contests I finished highest in, all three went RB-RB at the flex. Two went WR-RB. The final two went TE-RB. I know TE isn`t technically a thing, but I put players who will be drafted as NFL TEs in those spots, which I think in retrospect wasn´t exactly a high-floor move.

Player Selection – Of the Thursday slate, I had Jonnu Smith (TE) and David Ungerer in all three of my lineups. They COMBINED for 4.8 points. Silly me. Of the Saturday slate, I feel great about the way I ¨read¨ the BG/Tennessee game and deployed Tenn QB/WR stacks in half the lineups, while rolling with the Tenn RB in the other half. Ian Sadler was also a money play and I`m proud of myself for digging deep into the TTU pass offense to understand the Xs & Os of their offense to find the massively undervalued Sadler.

Stacks – In 3/7 instances where I deployed a stack, the pair connected for a touchdown. By my count, only 8/20 stacks I used connected for a touchdown last week across all formats. No bueno.

$/pt

QB – 3.00

RB – 3.24

WR – 2.89

FLEX – 2.88

Other random thoughts from last week

Toward that goal of hitting 5X returns per player, I think it is probably a trap to play quarterbacks priced over, say, 7.5K. I say this because, look at a guy like Anu Solomon, he had a very strong week last week with 28.66 points. However, because his salary was 9.2K, that performance yielded only a 3.1X return. Basically, if a really good (not epic) quarterback performance in a given week looks like 30-35 points, the only way to 5X that is to spend somewhere in the 6-7k range.

This Week

Finances

From my original $100.25, I am down to $93.75. As detailed in the original post of this series, I´m playing 10% of my budget every week. Technically, that means I should play $9.375 this week, but I´m going to cheat a fraction and play $9.50. That breaks down to
50-50s – $7
Multipliers – $2
Tournaments – $.50

Projection models

SInce we last spoke, and with the addition of new data, I built a sort of rudimentary projection model to get me looking in the right places (hopefully). I am going to keep tinkering with this and, depending on results, start to publish it publicly in the next few weeks.

50-50s

Because I started last week`s lineup creation with tournaments first, I am wondering if doing 50-50s first this week will improve my approach.

Lineup 1

RBs

Filling this out with RBs first, I am going to take Alabama´s Kenyan Drake ($6300). Last week Drake received 5/16 first quarter carries and played deep into the second half when the game was out of hand. Basically, he has the appeal of a pseudo-starter for Alabama, which is expected to score 45 points this week, the second-highest total of the early slate.

Going back to the board, I notice that there are four RBs priced over $8900 and then a drop off all the way to $6800 with Wayne Gallman. Last week, Clemson suffered the loss of WR Mike Williams, and while they have talented options behind him, I could foresee the running game being the biggest beneficiary of his absence. Gallman is the clear-cut lead back in a game where the Tigers are expected to score 38+ points. Also, Clemson has a point-share in this game of 64%, which is the lowest of all the projected high-scoring teams, meaning that the game against App State should be close the whole way. Add in a 30% chance of rain and I can see Gallman being the man for Clemson today.

QBs

Sticking with my theory about high-priced quarterbacks being unlikely to return a 5X performance, I`m going to start my search here with the under $8,000 crowd, which means surpassing Deshaun Watson, Luke Falk, Cardale Jones and Justin Thomas.

Malik Zaire is a tempting name, but witha 70% chance of rain in Charlottesville, I`m going to pass. Brandon Allen is interesting too, but I want to hold out to see if I can work Alex Collins into the Flex spot. Chad Kelly ($7700) from Ole Miss is an interesting option. Team total of 42.5. My reading of the Vegas action is that Fresno should hang for a while, meaning Kelly has a good chance to play a full game. He`s got some dual threat upside. Let´s try him.

Feeling a little tight on my budget, I am looking for a QB with a cheap price tag and a reasonably high team total. Joel Stave, Wisconsin, crosses my mind, but I could see a scenario where that game gets ugly and he throws 11 passes. Minnesota has a 30.5 point team total in a game that should be very close. QB Mitch Leidner ($5700) is imperfect, but as a starting QB with a good chance for 10+ carries and a reasonably strong offense, I think he has a nice floor.

WR

With an ¨average budget¨ remaining of $4700 for my remaining three WRs and two flex, I feel a little strapped for cash. Here I will look to stack with one of my QBs, which will probably be from Ole Miss. Laquon Treadwell is the obvious name, but I don`t have any reason to believe he´s more favored by Kelly and teammate Cody Core is ($4800) comes at an $800 discount. Let`s try him.

At this point I am going to go digging for a deep sleeper for minimum budget.

….scrap that idea.

One name of the under-$5000 crowd that my model really likes is Alabama`s Robert Foster ($4800). Yes, I know I also have his Alabama teammate, which caps my upside, but I also think there will be enough points to go around, so I`ll slot him in there for now.

At this point I am jumping ahead to Flex to see who I can slot in there and keep in the range of $4,600, which I need to average with my final three roster spots. Michigan RB De´Veon Smith ($4800) looks like the lead back on a team that should score 30+ points. To me, he projects to return about 3.5X value and could do even more if the Wolverines are playing this game from ahead, instead of trailing like last week.

$9000 left and two spots.

Iowa´s Matt Vandeberg caught 37% of Iowa´s passes last week and can be had for $4300. Although the Hawkeyes are only expected to score 27 points, the game is supposed to be very close and I think Iowa State´s high powered attack could force this game higher. Let´s take him.

Final spot (Flex) goes to WR Malcolm Mitchell from Georgia ($4700). As the featured receiver on a team that´s expected to go for 35+ points today, I will roll the dice. Allen Lazard was also considered here.

Just kidding! I promised myself for all 5050s this week I was going to go RB-RB in the flex. Patrick Skov ($4400) from Georgia Tech gets the call and gives me a share of an offense that should surpass 40 points today.

Final roster…

cfx week 2 5050 lineup 1

Six more 5050 lineups to construct, so I´m going to whip through them fairly quickly.

Lineup 2

This time through I am starting with Alex Collins, who I think is my favorite RB from the early games, even if his price is substantial ($8900). Immediately this means I´m going to be forced to go even lower budget at QB than I did before, which lands me with a pair of Big Ten QBs in Jake Rudock and CJ Beathard, whose teams are slated for 31 and 27 respectively and who I think can return good bang for their buck. As I already have a share of Iowa´s top receiver, I am skipping a Beathard stack and pairing Rudock with Jake Butt, who appears to be equally prioritized in the offense compared to the higher-priced Amarah Darboh. At RB, I am interested in the Rutgers´ starter Paul James, who should see plenty of action on a rainy day in NJ where his team is slated for 32+ points against Wazzu. Another RB I really like, and who drew rave reviews in week 1 is Minnesota´s Rodney Smith. Through two lineups I have no shares of Ohio State´s offense, so I will slide Braxton Miller in here. I think that offense will be dominant again this week and it wouldn´t surprise me if Miller contributed in the run, pass and receiving game. Sticking with a theme from last lineup, I am getting some exposure to the GT offense in the form of Marcus Marshall. Last roster spot. $3500 left. Hunter Henry from Arkansas gets the call. Two TEs in one lineup, that might be dumb. Also, I have two pieces of the Arkansas offense now.

For time sake, I am just going to insert the rest of my 5050 lineups here.

cfx week 2 5050 lineup 7 cfx week 2 5050 lineups 2-3 cfx week 2 5050 lineups 4-6

Tournaments

To the opposite side of the coin, I want to build these lineups to be as high variance as possible and will be filling both Flex positions with wideouts.

Having figured out that there aren´t many dirt cheap WR options that I like, I know I need to keep my QB and RB spending in check. Right off the bat, that knocks out the top options at each position. Same as before, I really like QB Chad Kelly and RB Wayne Gallman as best-of-tier-2 options. So they are in first. News just broke that Corey Clement is out today, so I am rolling with Taiwan Deal, who might get the Kenyan Drake treatment in the sense that he could be the 1b RB early, but then get extended action through the last three quarters. For my second QB, it comes down to Brandon Allen and Jake Rudock as two options I like in a similar projection range, however because Allen is $1900 more expensive, I go with Rudock.

To the pass catchers and I immediately spring for Treadwell and Darboh, as I think they are the highest-upside pass catchers for both QBs. As I mentioned before, I think Alabama WR Robert Foster could have a big day, so he gets selected too. With $9200 to split between my final two Flex spots, and knowing I don´t have a minimum salary player left to use, I opt for OSU´s Michael Thomas for $5200. Again, I think the Buckeyes will score plenty of points and it wouldn´t surprise me if Thomas was the forgotten man by the defense. Finally, of the $4000 and under WR crowd, I am drawn to Quenton Bundrage going against Iowa. Hunter Henry got consideration here too.

My second tournament lineup is going to be a spin off of the first, swapping in a few players I mentioned and switching a stack.

cfx week 2 tourney lineups

Multipliers

Many of the same names you have seen before. The goal here though was to make sure I had WR-RB in the flex in both lineups.

cfx week 2 multiplier lineups

Total Shares of Players

No time for this right now. Will follow up next week.

Knee jerk reaction is that maybe I have too much of Chad Kelly or Wayne Gallman.

Alright, so now here is the fun part… you get to tell me where you agree or disagree with my thought process, player selection, etc., so go ahead and leave a comment.


Jon Moore is a contributor at RotoViz and a cohost of Rotoviz Radio – A Fantasy Football Podcast.  Continue this conversation with him on TwitterGoogle+ or Facebook.

Advertisements