Writers of Pro Football Prospectus 2008

Most Recent FO Features

MurrayLat13.jpg

» Scramble for the Ball: With All the Fixings

An idiot's (two idiots'?) guide to Thanksgiving football, prepped and primed for the monsters-in-law who only watch these three games in a year.

17 Dec 2014

Scramble for the Ball vs. DYAR Fantasy Football

by Tom Gower and Mike Kurtz

Tom: Earlier this year, I got an email from Ben, who wrote:

Just stumbled upon your site. It's everything the Bleacher Report isn't. My new favorite site. Have you made DYAR accessible for fantasy? It would be fun to compete with other players over a single statistic, DYAR. ESPN isn't including that statistic on their options. The only way that I see doing it is with your premium package. From the description, only DVOA is mentioned, so I'd have to come up with a way to adjust for DYAR. It also lacks the instantaneous feedback that pop fantasy sports offers. This would take me back to the nerdy days of manual fantasy record keeping.

I'm sure this is a big market opportunity that you have given plenty of thought about.

Tom: To which I replied:

Thanks for the question. It's definitely an idea that's come up before.

One problem with DYAR as a fantasy stat is, well, the D part of it changes as we get a different, generally better idea of how good defenses are and how impressive a performance truly was. Picking an example at random, LeSean McCoy had 22 rushing DYAR when we ran Week 2 Quick Reads. As I look at his performance now, based on the full-season adjustment for facing San Diego's defense, he has 8 rushing DYAR for that Week 2 game. If you want to run a DYAR league, do you retroactively adjust old games or just accept that early season results may be skewed? I don't like either option, personally.

Now, you could do YAR, as that doesn't change (McCoy had 16 rushing YAR that week). A while back, somebody calculated a YAR estimator where you could customize your fantasy setting to get something pretty close without actually checking YAR. That was before the current version of DVOA, though, and AFAIK nobody has re-done the calculation. Our position pages do have YAR on them, so you could keep score manually by saving the prior week's page and comparing the YAR difference. As you said, that'll bring you back to the old days of manual scoring, but you could do it.

Practically, the big problem with a YAR league is YAR is dominated by quarterbacks. LeSean McCoy led all running backs last year with 486 YAR. Peyton Manning had almost 2,700. That was also 844 more than second-place Philip Rivers. You'd have to move away from YAR or throttle it down somehow unless you want your league A) determined by who gets the top passing quarterbacks and B) highly sensitive to the week-to-week performance of your quarterback.

(D)YAR fantasy seems like a much better idea than it probably actually is, which is why we keep not doing anything with it even though it keeps coming up.

Mike: That brings up an interesting question. Why has roto fantasy football never carried the cachet of head-to-head?

Tom: Part of it is football, because of its schedule, has a particular rhythm that makes head-to-head more appealing.

Mike: The major downside, as you said, is that numbers are adjusted after the fact. That is a huge problem for head-to-head leagues, but not really a problem for roto leagues.

Tom: You can track the progress of your team through the Thursday, Sunday early, afternoon, and night games, and then into Monday. It feels more competitive. Whereas baseball head-to-head would be a trickle of games coming in over the course of a week or whatever. Baseball head-to-head is boring. Football head-to-head is fun. Most people like competition. I don't think it's much more complicated than that.

Mike: Probably not, as frustrating as head-to-head fantasy is.

Tom: Head-to-head fantasy also guarantees interesting and compelling games for the top teams late in the season. The last time I played fantasy baseball, half the teams in the league were dead by midseason. Dead in the sense of having no realistic chance of winning. A couple owners mostly dropped out and had to have the commissioner adjust their rosters. A couple others, including yours truly, kept plugging along in vain. I haven't seen the same problem in 10-team or even 12-team fantasy football leagues, because the fourth team generally doesn't have to be that good to make the postseason. Heck, the third-place team I beat this past weekend went just 8-6.

Mike: That seems to be more a function of parity in fantasy football

Tom: Parity, another reason head-to-head is more interesting in fantasy football.

Mike: Well, no. It's a weird sort of forced parity. The team with the most "points for" in my league missed the playoffs. That's a good player being punished pretty harshly for random chance.

Tom: It's a weird sort of forced parity because if you got rid of forced parity, fantasy football would be less interesting.

Mike: For someone who isn't a natural fantasy football player, you sure are a hegemon.

Tom: There's no way touchdowns should be worth six points. I've accepted that fantasy football is Not Real Football for good and valid reasons.

Mike: There is a difference between "not real football" and "arbitrary and capricious." And that is why we keep seeing people asking us about DYAR and DVOA for fantasy. They want a fair fight.

Tom: Haven't we had the "let's fix fantasy football standings" conversation before, where we suggested a hybrid model where half the standing is based on head-to-head, and half is based on total points? And, yes, that may be the royal "we."

Mike: We have, actually. We've even come up with some half-brained suggestions to deal with in-game injuries.

Tom: To use my league this year as an example, I had a record of 10-4 and finished second in points. My record in that scenario would be, oh, say 21-5, 10-4 for head-to-head and 11-1 for points ranking.

Mike: But I think the two sentiments are in the same vein. Fantasy football puts a huge emphasis on players on teams that win that particular week and penalizes heavily teams that, through no fault of that particular player, just don't have it. That's the appeal of DYAR. That just because a player didn't catch a touchdown doesn't mean it's a wasted roster spot.

Incidentally, I would say this would lower both actual and apparent parity. RB2s in particular would be dominated by guys who get vultured but consistently convert first downs. That a more savvy player would have knowledge of.

Tom: Or by whichever low-volume back has reasonable efficiency that week. Or gets a couple useful catches, as in the case of Washington fullback Chris Thompson, who finished 13th in DYAR this week among running backs. I guess he did catch a touchdown pass, so maybe he's a bad example. Instead, try San Francisco fullback Bruce Miller, 17th in DYAR among running backs. It's more fun to have Arian Foster (57th in DYAR), C.J. Anderson (58th), and Jonathan Stewart (61st and last) be viable fantasy options, since they all had at least 22 carries.

Mike: More fun, probably. There is also something to be said for the fun of a game that rewards a greater depth of football knowledge.

Tom: I'm in favor of rewarding depth of football knowledge when it comes to fantasy. I'm just not sure how you want to do it, or even if most people want to do it. Fantasy football is popular in part because you don't have to obsess over the NFL to do it.

Mike: Brutally punish anyone who rosters any Detroit running back. That is how.

Tom: I don't know what you have against Detroit running backs, but I suspect it could be cured by forced viewings of Tennessee running backs.

Mike: In that my brain would be a puddle on the floor, yes. A given value of "cured."

Tom: As a Colts fan friend said to me, watching Trent Richardson may be destroying his brain by questioning what really is a hole a running back can be reasonably expected to run through. If only we could turn that into a useful answer, then we could really revolutionize fantasy football the way it needs to be.

Mike: Trent Richardson: an enigma wrapped in a riddle wrapped in two first-round draft picks.

Lessons Learned

Mike: Dear God, the Bears are bad. That's not really a lesson from this week, though.

Tom: You could still make it this week's lesson.

Mike: Nah, that's just piling on. This week I re-learned that I love defensive football.

Tom: Buffalo-Green Bay?

Mike: It has been a treat to watch Detroit's defense this year, but thanks to the NFL's wacky scheduling rotation this past week we had Green Bay-Buffalo on the television, and it was immense fun.

"Adjustments" has to some degree snuck into the fan's parlance as a general term for "I don't like what is happening but don't know enough to explain why," but this was a game full of actual adjustments. Buffalo began the game firing its ends on every down. Aaron Rodgers had two disastrous series, then Green Bay started running up the middle toward the now-isolated tackles, with great success. Buffalo switched to stunts to cover themselves a bit more, which slowed the pass rush and gave Rodgers some time, but at that point he was anticipating the rush and his timing was all off. McCarthy then started doing wacky things with Randall Cobb; end-arounds, straight-up runs, slot screens, Cobb became the focus of the offense as Green Bay attempted to recalibrate the passing game as time wound down.

Tom: I haven't had the chance to watch that game yet, but I've read several times about how Eddie Lacy ripped off big chunks of yardage for one stretch, then didn't get the ball at other times. That's the first explanation I've heard as to why.

Mike: It even sort of worked, as Buffalo's offense stalled (as if it ever started), but by that point Green Bay was a one-dimensional offense against a tremendous front seven, and the game ended with a holy roller safety.

Just incredibly fun to watch. A great performance by the coaching staffs. A great performance by Buffalo's defense and Green Bay's offense. A ... really terrible performance by Kyle Orton. I guess you can't have it all.

Tom: I guess not.

I'm tempted to make a grand statement about Brock Osweiler, finally allowed to attempt a pass in a competitive game. Two attempts is enough to start making definitive conclusions, right?

Mike: You are a card-carrying member of the sports media. Two attempts is enough to project an entire career.

Tom: Instead, I'll look at another rookie quarterback, the one everybody was watching on Sunday. Johnny Manziel was ... not good.

Mike: In so many words.

Tom: He was, maybe, even worse than Brian Hoyer was the week before, when even people who recognized and appreciated Hoyer's strengths could understand why the Browns would make the switch at quarterback.

Mike: Hoyer's strengths? "Has a pulse?"

Tom: Naturally, we'd heard clamors for Manziel ever since people realized the Browns would be starting Brian Hoyer. "Has an idea of how to read defenses and find open receivers, to whom he is willing to (try to) throw the ball, and can get it there enough of the time to be useful," is what I was thinking. This doesn't sound like much, but it ends up being hard to do. As shown by the number of people who try to do it and failed at that.

Mr. Manziel was a fun-to-watch, incredibly effective college quarterback. That's distinct from being a good NFL player. There are exceptions, but for the most part NFL teams play the players who are most effective. That an NFL team is not playing a player, especially one like Manziel in whom the people above the coaches has invested so much, is an act with significant positive informational value. Ignore that at your own peril. Not a new lesson, granted, but one worth repeating.

Loser League Update

Full results for this week's Loser League action and the season to date may be found on the results page. Each week, Scramble for the Ball highlights the top scorers at each position.

Quarterback: Speaking of, here's Johnny! If you rostered Johnny Manziel, you finally got the payoff you'd been hoping for after a string of 15-point penalty weeks. With two turnovers, under 100 yards passing and just 13 yards rushing, he had only 1 point. That was half as many points as this week's runner-up quarterback, and one-third as many as the third-place passer! (Jake Locker and Andy Dalton, for the curious.)

Running backs: Just like at quarterback, the top three running backs in the league had 1, 2, and 3 points. Tre Mason fumbled for his 1, while Alfred Blue got to his 2 more honestly. (Trent Richardson had the 3, and your Scramble writer swears he had not looked at the Loser League results when he brought up Manziel and Richardson.)

Wide receivers: Reggie Wayne's 24 yards on four catches was the third-lowest yardage total of his career on at least four grabs, ahead of the 15 yards he got against the Bengals in Week 5 and the marvelous five catches for 21 yards he had in the season finale against the Bills in 2009, a great feat of stat-padding to get him to 100 catches. With the fumble he lost, he had 0 points. Matching him was Andrew Hawkins, who drew plenty of headlines for the things he did off the field.

Kicker: Did Blair Walsh have the most excusable 0-for-3 on field goals in recent memory? He put it wide right from 53 yards and was short from 68 yards at the end of the game. Nobody in the past 20-odd years had kicked a field goal 68 yards in the NFL -- the list of longest field goal misses shows Greg Zuerlein's 66-yarder against the Dolphins in 2012 is the longest miss that had the distance. His only close attempt, from 26 yards, was blocked. No matter how excusable it was, Loser League shows no mercy in awarding him -4 points.

Awards!

Keep Chopping Wood: As the Aaron Kromer story showed this past week, whatever is going on in Chicago is not just an on-the-field problem. But the Chicago Bears showed on Monday night they had no answers on the field for anything that was going on off the field. The list of mistakes is too long to usefully enumerate, from Kyle Fuller's regression to the multiple bootlegs where Bears defenders left receivers uncovered to go after Drew Brees to whatever Jay Cutler was doing to whatever his offensive linemen were doing at times.

Mike Martz Award: As Dan Steinberg and the Washington Post headline writer put it: The Redskins and Jay Gruden gave up at the end of the Giants game, and it was weird and sad. Most teams fight to the bitter end, or close to it. Washington instead tried just enough to show they were not trying, punting twice in the final five minutes down two possessions but still running plays, even when the Giants muffed a punt to give them the ball with nine seconds left. It was a performance truly deserving of the Mike Martz Award.

Lock of the Week

Tom: We are now both 4-9 on the year. If you were betting against us, either or both of us, you'd be up on the season. We even actually did get some cool weather and rain in Chicago like the long-term forecast indicated we might. They couldn't help the Bears do anything on offense, though, or actually prevent Drew Brees from doing pretty much whatever he wanted. While I did not see the Pittsburgh-Atlanta game, reports I got indicated that Pittsburgh's advantage on the lines was as significant as a look at Atlanta's lines at any point since, oh, about May would suggest. Like the NFC South winner, we will not finish the regular season above .500.

Mike: I've never been happier to be wrong

Tom: I'd like to ask Julia to verify that one.

As a reminder, all lines are courtesy of Pinnacle Sports and were accurate as of time of writing. All picks are made without reference to the FO Premium picks.

Mike: I want to put fake money on Titans-Jaguars, if only for the novelty.

Tom: You should.

Mike: The real problem is that I would have no idea who to pick.

Tom: In that game? Jacksonville. The Titans really are just that bad, and weighted DVOA suggests the Jaguars should be favored by more like -5 than -3.

Mike: Well, the problem is that they are two of the four worst teams in football, even by weighted DVOA, and I'm not sure how much I trust DVOA spreads that far negative.

Tom: The Jaguars defense has kept things semi-respectable much of the season. Beating the Titans by a touchdown or more will be a function of whether their offense can stay out of their own way long enough to string plays together.

Mike: That is not something I am willing to ever bet on.

Tom: Fair enough.

Mike: I will instead note that Tennessee rose a whole two (2!) spots in the DVOA boards last week. And gets the points. Momentum! So there's your "how did we get here" lock of the week. Tennessee Titans +3 at Jacksonville Jaguars.

Tom: There are a number of lines this week that are just attractive enough to get you into trouble, like the Eagles by 7.5 in Washington. The Vikings in Miami maybe should be closer to -5 than -7, but I think of that Dolphins defensive line against Minnesota's offensive line and wonder why it's only -7. Heck, Green Bay in Tampa maybe should be within a touchdown, yet it's too easy for me to see that game ending up 48-7. I'm almost tempted to take the Bears getting a touchdown, except I saw A) Monday night's game and B) the Thanksgiving game.

Instead, I'll go with one of my teams for this season, or more specifically against Case Keenum and/or Thad Lewis. Weighted DVOA suggests this line is already inflated. Weighted DVOA isn't incorporating the downgrade from Ryan Fitzpatrick, and even with Haloti Ngata the Ravens have a fine defensive front. Arian Foster can do great things, but I think Gary Kubiak's team wins another game in Houston. Baltimore Ravens -5 at Houston Texans.

Posted by: Mike Kurtz and Tom Gower on 17 Dec 2014

9 comments, Last at 21 Dec 2014, 1:47pm by Andrew Potter

Comments

1
by Crunchums :: Wed, 12/17/2014 - 5:06pm

Retroactively adjust old games, do something other than head-to-head to make that not weird, and do whatever to handle how the shape of the DYAR bell curve varies dramatically by position. Seems obvious to me.

edit:
I think people approach things from the perspective of "how could you jam DYAR into fantasy football" because fantasy football is what people are used to. All I am looking for is a competition that rewards accurate predictions of DYAR outcomes. There's no reason it has to look anything like fantasy football.

2
by Steve in WI :: Wed, 12/17/2014 - 7:39pm

I've had several ideas for interesting ways to score a fantasy league, although I don't know if anyone besides me would find them appealing:

- a league where for every position, you pick a team and not individual players. By necessity, you'd only get to start one team per position and still it would probably be best if it was limited to 8-10 players. No more worrying about running back by committee!
- a league with no starter/bench divide: you add up the totals for every player on your roster. This would reward drafting a reliable but low-ceiling guy over a potential star who's unlikely to see the field much but has an outside chance at a big season if he can move up the depth chart.
- a league that more closely mimics the real NFL structure and puts a premium on smart drafting rather than the waiver wire: your team at the start of the season is the team you're basically stuck with, and in case of injury or underperformance you can only pick up guys off the street who are generally unlikely to be good. Each player would draft somewhere around 35-40 guys, and in the case of injury, suspension, bye week, whatever, you've got to substitute someone from your own bench - no snagging the hot sleeper of the week from the waiver wire. Obvious downside would be the length of time it would take to draft, plus what do you do about the defense?

I think it goes without saying that one or more of my fantasy teams has gotten burned by teams with 3+ running backs who are viable options to get most of the carries, players I choose to bench who score 20 or more points more than the ones I started, and opponents who've made like 40 waiver wire moves this season.

3
by rageon :: Thu, 12/18/2014 - 12:35am

I came up with a scoring system once that fairly closely correlated to DYAR within position groups, but also normalized players between positions. A couple years of using it mostly just lead to permanent retirements from fantasy football and some friendships ending. It has since been retired.

4
by Lance :: Thu, 12/18/2014 - 7:31am

"Mike: Well, no. It's a weird sort of forced parity. The team with the most "points for" in my league missed the playoffs. That's a good player being punished pretty harshly for random chance."

This is why I eventually stopped playing. I was paying a lot of money each year for something like 7 years straight. Yet, even though I regularly finished in the top 4 in scoring, I only make the play-offs like twice. I once finished #2 in PF but #1 in PA and ended up 5-8 missing the play-offs. Since PA was completely out of my control, I was basically just horribly unlucky. At that point, why play?

5
by Sakic :: Thu, 12/18/2014 - 9:32am

I've had the same problem...in one of my leagues I had the most points scored against in the league for three straight years. Fortunately, the commissioner finally installed a wildcard playoff spot for the top point scorer to ensure that even if you run into a string of bad luck in head-to-head matchups you can avoid the big screw-g.

6
by dbostedo :: Thu, 12/18/2014 - 9:35am

Because it's fun, and the luck will come out in your favor at some point. I always think of the money put in as payment for the fun of playing. Anything I'd win would just be a bonus.

7
by Lance :: Thu, 12/18/2014 - 11:08am

I get that to an extent. And it's why I stayed with it for so long-- I enjoyed the interaction with friends and acquaintances. But at some point there is a competitive aspect to it and when my "luck" was such that I'd have the second highest score of the week but inevitably play the guy with the highest score of the week and thus "lose" the game, it stops being fun.

Perhaps, over time, it would have balanced out? Or perhaps not, because FF isn't random like flipping a coin. I put thought into my picks and starters. It took time to figure out match-ups and whatnot. In the end, it stopped being fun.

8
by serutan :: Thu, 12/18/2014 - 5:54pm

Tom : "... even with Haloti Ngata the Ravens have a fine defensive front."

Methinks you meant "without", since I have a vague memory of Ngata being injured...
______
Was wr

9
by Andrew Potter :: Sun, 12/21/2014 - 1:47pm

He's suspended for the rest of the regular season under the PED policy.