Let's start off by seeing how my value probabilities fared last week for quarterbacks and running backs. (In case you're wondering, I'm saving the accuracy analyses related to wide receiver, tight end, and defense for the columns that'll appear during the NFL playoffs. That allows me to use full-season data for "more random" positions that need as much sample size as possible.)

The Top 3 cash value probabilities at quarterback in Week 15 were for T.J. Yates, Matthew Stafford and Philip Rivers. Yates needed 13.5 points to achieve value, but he left early with an injury after having only scored 3.1 points. Stafford needed 19.2 points, and he scored 24.6, while Rivers needed 18.0 and scored 25.6.  For tourunaments, the Top 3 value probabilities were for Yates, Matt Hasselbeck, and A.J. McCarron. Unfortantely, none of the three achieved value, but there's an easy explanation for them all. Again, Yates got injured. Meanwhile, both Hasselbeck (41.6%) and McCarron (33.9%) had probabilities that were below even money, which means it was actually more likely that they wouldn't achieve value.

The Bottom 3 value probabilities in both cash games and tournaments were for Ben Roethlisberger, Aaron Rodgers, and Carson Palmer. As expected, Rodgers and Palmer didn't even 3x value, let alone 4x. Roethlisberger, however, unexpectedly scored 28.7 points against Denver's stout defense, which was higher than what he needed for cash games (25.8), but lower than what he needed for tournaments (33.6).

My system's worst Week 15 performance -- by far -- related to high-value running backs, which included the same five players for both cash games and tournaments: Melvin Gordon, Fozzy Whittaker, Denard Robinson, Chris Polk, and Matt Jones. Robinson achieved achieved value in both cash games and tournaments, but the other four failed in both formats. The sad fate of this quartet in Week 15 represents a perfect example of why making predictions for non-quarterbacks is really hard: Gordon and Jones suffered injuries during their games, while Whittaker and Polk ended up as little-used backups despite being widely reported as starters prior to their games. In-game injuries and reduced roles -- especially for more mysterious reasons (a la Whittaker and Polk) -- are things we can't possibly know or foresee in advance. This kind of randomness is everywhere in football, and so projection failures such as these are inevitable. What matters most for evaluating a system is that its errors are due more to these random factors than factors we can actually account for in advance. 

Ironically, while my probabilities went 1-for-5 for the Top 5, they went 6-for-6 with respect to running backs in the Bottom 5. Alfred Blue came closest to beating the odds, scoring 10.7 when he needed 13.4 in cash games and 19.2 in tournaments. The trio of Karlos Williams, Alfred Morris, and C.J. Anderson didn't come close in either format, as expected. Neither did Ryan Mathews in cash games, nor Spencer Ware in tournaments. To be fair, however, Ware went the way of Whittaker and Polk, riding the pine (albeit due to injury precaution) even though he was expected to play.

Overall, then, the system was much better picking failures than successes last week, but most of the failures could be explained away to randomness. Still, here's to hoping it does better on the high-value side in Week 16. Speaking of which...

quarterbacks

Below are the quarterbacks with the highest (and lowest) probabilities of achieving value in cash games and GPPs:

The Chiefs-Browns game is forecast to be played in 20-plus mph winds. Therefore, despite being good values, I want nothing to do with Smith or Manziel. The latter, in particular, is off my radar because of the wind, as he relies more heavily on deep and intermediate passes. With those two out of the picture, Bridgewater is the best value of the week for my money. He's coming off back-to-back 20-point games. Furthermore, opposing quarterbacks have achieved cash value (3x) in six of the Giants' past seven games, five of whom also achieved tournament value (4x).

running backs

Below are the running backs with the highest (and lowest) probabilities of achieving value in cash games and GPPs:

With LeSean McCoy out, Williams is a must-play against a Cowboys defense that ranks 27th in both run defense DVOA and FantasyScore points allowed to opposing running backs. Among the other four running backs offering Top 5 value this week, Bell has the most factors working in his favor. Unlike Robinson, Bell's entering his game with a full week of practice. Unlike Spiller, Bell hasn't burned his bridges with DFS players worldwide. And unlike Gaskins, Bell's on the home, heavily favored side of Lions-49ers. Speaking of which, San Francisco's home-road splits on defense are a sight to behold. On the road, they allow an average of 7.56 yards per play and 3.57 touchdowns per game, versus allowing 6.06 yards per play and 2.00 touchdowns per game at home. That divergence is particularly pronounced in run defense, where they allow 58.8 more yards per game and 0.71 more touchdowns per game on the road.

With respect to the bottom half of the table, I'll say this. While value considerations suggest staying away from all five running backs, I think there's a case to be made for throwing Lacy into one of your tournament lineups on the off chance that losing Tyrann Mathieu, the straw that stirs their defensive drink, renders Arizona's defense far less stout than it has been all year. People talk a lot about Mathieu's coverage skills, but what's flown under the radar is that he was a vital cog in the Cardinals' run defense. To wit, in addition to his 5 interceptions and 4 passes defensed, his 11 tackles for loss actually led all defensive backs in the NFL and -- get this! -- was tied for the lead among all Cardinals' defenders, regardless of position. Am I saying Lacy's going to score the 24.4 points required to achieve tournament value? No. But, given how important Mathieu was for Arizona's overall defensive scheme, it wouldn't surprise me if Green Bay's able to move the ball much better than expected. And, if Lacy's healthy, that development makes Lacy an intriguing contrarian play.

wide receivers

Below are the wide receivers with the highest (and lowest) probabilities of achieving value in cash games and GPPs:

Among high-value wideouts, Royal and Butler are clearly the best two choices. For Royal, Alshon Jeffery's absence means more targets against a a Buccaneers defense that ranks 24th in both pass defense DVOA and FantasyScore points allowed to opposing wide receivers. For Butler, he's established himself as the clear No. 2 wide receiver in Baltimore, and the Ravens are going to have to pass a ton if they have any hope of keeping up with Pittsburgh's high-octane scoring machine.

Almost all of the receivers on the low-value side of the table appear because they're prohibitively expensive. So which of these high-priced targets do I think is most likely to beat the odds and achieve value? The answer: I'll take a chance on Allen Robinson, especially in cash games. He needs 23.7 points for 3x, and he's scored at least that many three times so far this season. Most importantly of all, however, is that he'll be running the majority of his routes against Brandon Browner, arguably the worst cornerback in the league (both of those assertions are per Pro Football Focus).

tight ends

Below are the tight ends with the highest (and lowest) probabilities of achieving value in cash games and GPPs:

Although FantasyScore finally increased his salary from the minimum ($2,100), Will Tye still will not be denied as the weekly Fooftballguys Value Guy. If that salary scares you off, however, Williams is a good value play. Being that he's still a relatively unknown quantity from a production standpoint, this is especially the case in tournaments. He could score far more than the required 8.4 points in a shootout with Pittsburgh, or the combination of his own inexperience and Baltimore's lack of quarterback talent could add up to Williams scoring in the low single-digits. That's variance, and tournaments are where we don't mind variance.

defenses

Below are the defenses with the highest (and lowest) probabilities of achieving value in cash games and GPPs:

Since Joe Flacco got hurt, opposing defenses have averaged just under 15 points per game against Baltimore, so Pittsburgh's the play here, especially in cash games. If you want to go even cheaper in tournaments, Miami's worth a gamble because opposing defenses have averaged 18.3 points per game against Indianapolis the past three weeks. And we're not talking about facing the '85 Bears three times here. Rather, those points have been scored by the mediocre-to-bad defenses of Houston, Jacksonville, and Pittsburgh.

week 16 draft lists

Finally, to supplement the statistics-based strategies I recommended earlier in the season, here are the VBD draft lists you should use for FantasyScore's Draft-N-Go (DNG) games:

Follow @IR_DannyT