***Official*** 2012 FBG Subscriber Contest Thread (1 Viewer)

[inca911]

Not sure what it means if anything.

I think it means you have a lot of free time.

 

[Stinkin Ref]

big OL ####### THATS WHAT I'M TALKING ABOUT to Dippa..... ....

 

[Stinkin Ref]

one last comment on what Dippa put together....

with this contest, FBG basically kind of gave us an overall cheatsheet/ranking of how they saw these players heading into the season...and how they thought they would perform for redrafters throughout the season...

so while this data is kind of nice to see as we move along for the contest...it is also a snapshot on how good the FBG's are.....the more guys they have staying close to the "zero" difference line as we move along.....the better they are......at least as of beginning of August or so and injuries not withstanding....

does that make sense or have I been drinking too much again....?

 

[Pigskin Fanatic]

does that make sense or have I been drinking too much again....?

both. with the obvious caveat of top dollar guys getting injured and the bottom dollar guys getting opportunities that weren't expected, it's a good point. cedric benson, for example, wasn't on a team yet iirc when the contest was released. or at least wasn't on a team when the fbg crew was probably going over this. there are several players who have wrong team names tagged for that reason.

 

[Stinkin Ref]

does that make sense or have I been drinking too much again....?

both. with the obvious caveat of top dollar guys getting injured and the bottom dollar guys getting opportunities that weren't expected, it's a good point. cedric benson, for example, wasn't on a team yet iirc when the contest was released. or at least wasn't on a team when the fbg crew was probably going over this. there are several players who have wrong team names tagged for that reason.

yeah obviously you can't hold injuries against them and this was at the beginning August....but of the players that stay healthy, it's a snapshot of what they thought in the beginning of August or so....

 

[sbuck143]

On the question of how to reconcile for a QB (or any positions) average output vs their volatiity and trying to equate it all back to 'value': I'm a little surprised nobody has brought up the classical mathematical representation of volatility.

Standard Deviation.

The example of two players who both average 20 pts a game:

Player A, 4 weeks of scores, that go 0,40,0,40 has a standard deviation of 20.

Player B, 4 weeks of scores, that go 20,20,20,20 has a standard deviation of 0.

The player with the lower standard deviation is more reliable, less volatile in their results and it is fairly easy to argue that given two players of same avg value, the one with lower standard deviation will be preferred for long term survival in the contest. But it should also be obvious that if those 4 weeks were weeks 13 thru 16...then Player A outscores Player B over the important wk 14 to wk 16 period.

Std deviation is a simple enough spreadsheet function, and can be quickly generated for a list of players. What you do with it from there is up to you. At the very least, it would provide a basis to compare players of approximately equal average points. You could also do a 1-sigma bell curve for each player (adding and subtracting 1 std dev from their average, to get their 68% confidence range) and be able to compare for example, a 20 pt guy with std dev of 6 with an 18 pt guy with a std dev 4. Hint: they have same floors, but one has a higher ceiling.

With this mindset, it should be a mathematical exercise to optimize an ideal roster- the roster that nets you highest average points with lowest std deviations that fits under $250.

The biggest issue I see is what we all see with fantasy football. Not enough data points. Sample size is very small - 4 data points per player right now. But its better than 1 data point and gets a little more reliable each week.

 

[Tennessee_ATO]

As for LHUCKS' question about the odds of Tannehill contributing over Stafford, granted it's a small sample size, but this year it appears to be 50%.

I have Stafford in 2 of my 3 leagues. I am acutely aware of his underperformance thus far. In this contest, that's the kind of thing that gets you an introduction to the Turk if you don't have an alternative. Even if you think you can get to the final 250, I think it's impossible to compete if you don't get a big score out of your QB position every week. It's a position that lends itself to a variety of "lower tier" guys throwing up something in the 30's almost every week. I would be interested in seeing the numbers for top finishers in recent years. I would bet dollars to doughnuts there aren't any 1-QB rosters in the mix.

 

[beast8812]

One of the biggest problems we have right now is sample size. The thing with the bolded argument above is Newton scored 12 points that week and your saying 9 other Qbs scored 30+. To me that is a huge issue. The rest of your team is now responsible for making up 20+ points that you lost at QB.

The idea is that you have another QB other than Newton. And it isn't so much that 9 other QB's scored 30+ as much as which 9 other QB's scored 30+. Players like Luck, Ponder, Griffin ect are players that are likely to be paired with more "studly" QB's. It should diminish what Flacco gives you that week because a lot of other QB's gave you similar production. So you don't need to make up 20 points at QB, because it's likely that your team benefited from this "backup" qbs.

I would rather have a QB that is a constant 24 points a week for the first 13 weeks then someone who has 35 one week and 12 the next. Those 12 point weeks will kill you, especially if a third of the qb field scored 20 more points then your qb. But in the last 3 weeks you want that qb who can bust out almost a 40 point week because that is how you win in the end. It kind of comes back to are you building a team to make the finals or a team that is more boom or bust, but could win in the finals.

Not sure your point. Flacco had a 16 point week. Sure it's more than 12, but it's not like Flacco is putting up 20-25+ every week either. Volatility is there for every QB.

Me personally I took Stafford and Flacco. A lot of people would argue I have too much invested in my Qb's, $38.

I don't think too many people would argue that. It's only what, $4 more than what Rodgers + Tannehill would have cost you.

But the reason I took the two of them is I expect Stafford to have some huge weeks and Flacco to have a steady average of around 20 - 25 points a week.

Okay, it's a viable strategy. But again I point out that Flacco had a below week (16.6 in week 2).

So far Stafford has been a bust and Flacco has been saving me, so what do I know. I have a small roster where I tried to take safe guys or boom guys. I doubt I will make it thru the first 13 weeks, but if I do I feel like I have enough guys with good upside to do some damage, hopefully the safe guys of the world like Flacco can get me there.

Maybe, I'm missing your point because I don't really see one (not trying to be a jerk, really I just don't see what you're getting at here). All I said was that I don't have a problem with Newton being ranked (in terms of value) ahead of Flacco because a) they both have had 2 good weeks in elimination weeks and 1 sub 20 point week, and b) in Flacco's best week, he wasn't that "valuable" despite his high score, because there were several other low cost alternatives that produced similiarly, and many of them were probably paired with the high cost studs. That's not the case in either of Newtons 2 good weeks. I guess my point is that you took Flacco as your "safe" QB in case of your stud qb faltering. In that particular week, his value isn't very high because alot of "safe" QB's produced equally. The sample size is indeed small, but as of right not it's all we have to work with. Will the results change, absolutely.

At this point we've lost track of the original argument and there are so many variables been thrown in that either one of us can make an argument for our case. You broke my response down like it was a physics term paper, when it was just my opinion on 2 players. My point was just giving you an example of how I looked at selecting qb's and nothing more. I think Newton can be more explosive and have bigger weeks to help you win the whole thing, but I don't believe hes the better bang for the buck. I've never made it to the top 250 so I am definitely not an expert at this, less then 2% make it to the finals, luck is a bigger factor then anything else. The original argument was which players are a better bang for the buck.Newton $28 - 99 points Flacco $11 - 106 points If I can have the same performance over the full season I'll take Flacco and save myself $17. We will just have to agree to disagree.

 

[Ignoratio Elenchi]

On the question of how to reconcile for a QB (or any positions) average output vs their volatiity and trying to equate it all back to 'value': I'm a little surprised nobody has brought up the classical mathematical representation of volatility.

Standard Deviation.

...

The biggest issue I see is what we all see with fantasy football. Not enough data points. Sample size is very small - 4 data points per player right now.

Right, standard deviation has been used before but I think after just four weeks it's not going to tell us much. Really, all of this discussion about who the best players are is something that's better left until the end of the season. In general, I'm more interested in roster construction strategy than specific player performance anyway - e.g. "which is better, 2 QBs or 3 QBs," as opposed to, "who is better, Flacco or Newton?" I think it was Modog last year who used standard deviation to track which teams were more likely to put up scores of 200+ points (which is typically what's required to finish in the big money) and showed that, while small rosters had lower average scores than larger rosters, their inherently greater variance made them more likely to finish with a really high weekly score.

With this mindset, it should be a mathematical exercise to optimize an ideal roster- the roster that nets you highest average points with lowest std deviations that fits under $250.

In theory this is doable, but there are like quintillions of possible legal roster constructions. It's an interesting question how to optimize the process of finding the optimal rooster. I've always wanted to but never got around to really giving it a shot.

 

[Tennessee_ATO]

It seems to me that there is no way to effectively assess the relative value of selecting specfic players at a position, or even player distribution at a specific position. Those decisions impact players and distribution at other positions too. I don't see how you could isolate that one specific variable in any type of analysis. Way too many other variables out there IMO.

 

[Saint]

In theory this is doable, but there are like quintillions of possible legal roster constructions. It's an interesting question how to optimize the process of finding the optimal rooster. I've always wanted to but never got around to really giving it a shot.

I've been trying to find the optimal rooster for decades. Heard it tastes like chicken!

 

[BassNBrew]

In theory this is doable, but there are like quintillions of possible legal roster constructions. It's an interesting question how to optimize the process of finding the optimal rooster. I've always wanted to but never got around to really giving it a shot.

I've been trying to find the optimal rooster for decades. Heard it tastes like chicken!

The solution to the optimal roster is pretty simple. Copy whatever memobrown does, drop his worst player from the lineup, and substitute a better player...collect your cash.

 

[BroadwayG]

that was not a good thursday night :X

 

[Wadsworth]

Danny Amendola can join Hernandez, Fred Jackson, Stephan Hill, Kenny Britt, Dion Lewis, Rashard Mendenhall, Jonathan Dwyer and Justin Blackmon on my team of underachievers. I think this may be my worst contest yet and I am amazed the team is still alive at this point.

 

[QuizGuy66]

I would assume Amendola's injury affect a greater proporion of big-rooster teams than small-rooster teams?

-QG

 

[Captain Quinoa]

I would assume Amendola's injury affect a greater proporion of big-rooster teams than small-rooster teams?-QG

I'm a small (20) roster and he has counted every week for me, 3 times as WR once as flex. My RBs are bad as it is and I have no McFadden this week. My only TEs are Hernandez (gimp) and Pettigrew (bye).

 

[BroadwayG]

I would assume Amendola's injury affect a greater proporion of big-rooster teams than small-rooster teams?-QG

I think he was a pretty popular pick across the board

 

[Ignoratio Elenchi]

I would assume Amendola's injury affect a greater proporion of big-rooster teams than small-rooster teams?-QG

Amendola owners:

SIZE	OWN	ALL	OWN %18	408	2302	17.7%19	213	984	21.6%20	194	792	24.5%21	203	656	30.9%22	181	581	31.2%23	181	505	35.8%24	141	407	34.6%25	143	315	45.4%26	110	242	45.5%27	88	211	41.7%28	63	153	41.2%29	67	137	48.9%30	114	219	52.1%TOT	2106	7504	28.1%

So he's on way more large rosters than small rosters, relatively, though of course the small rosters that do own him will presumably be much more impacted by the injury than the larger rosters.

 

[eaglezzz]

I probably would have took Amendola this year but I took him last year and he got hurt early in the year and burned me. Couldn't make the same mistake two years in a row. That is the same reason why I didn't take McFadden or Hillis this year either.

 

[Pigskin Fanatic]

So he's on way more large rosters than small rosters, relatively, though of course the small rosters that do own him will presumably be much more impacted by the injury than the larger rosters.

true. i have 21 players (6 receivers) and Amendola has counted for me every week. It'll be tough for my team to replace that production, though I pegged him as my fifth best receiver so this may be a bigger testament to other receivers underachieving. On a whim I also picked up Steve Smith, maybe that'll pay off assuming he is reactivated next week in place of Amendola. 153 teams still alive with both Smith and Amendola, interesting.

Survival Rate 	Overall Contest Survival RateTotal teams with these players	201 		Still Alive	153 	76.12 	56.48

 

[HairySasquatch]

My days are numbered with the Amendola injury. This is going to hurt with only 5 total WRs. That leaves only 3 possible starters week 5, 6, 7, and 10.

Is it possible to find out how many teams with Amendola have 4-6 WR's?

 

[Hatch]

Stafford on bye and Locker... Please let me make it through this week!!!!

 

[Saint]

I had a brainstorm this morning as I was thinking through the winning roster combinations from the last several years. Hoping some of you guys with access to the stats and databases can help test the validity of this theory.

Is there any correlation to winning roster and the percentage of points coming from each position? Such as, was there a strong correlation in the winning teams that QBs accounted for 20% - 22.5% of their points each week? Same for RBs, WRs, etc.

I think the theory would work regardless of roster size. I just don't know if that's a common theme or not. If it test to be true, with some correlation, could we apply the same logic to the teams today to possibly forecast rosters that seem to be in the spectrum that could be in the winnings this year?

All I know for certain, is that my team isn't one of them as I have been scraping by each week!

 

[atcdav]

Stafford on bye and Locker... Please let me make it through this week!!!!

3 QBs remember for next year!

 

[Stinkin Ref]

Stafford on bye and Locker... Please let me make it through this week!!!!

3 QBs remember for next year!

to me at least 3 qb's with different bye weeks always seemed to make sense....even if you went big on a stud....kind of a no brainer almost....but you never know I guess

 

[Alberta Clipper]

Is there any correlation to winning roster and the percentage of points coming from each position? Such as, was there a strong correlation in the winning teams that QBs accounted for 20% - 22.5% of their points each week? Same for RBs, WRs, etc.

I don't think there is a winning % or formula. I look at each year as a unique puzzle and at this stage there are still many ways that it can be solved. Every year the top teams end up with some great values that emerge but those are not necessarily the same pos from year to year.Another factor that makes any historical breakdown less important in predicting future is the scoring system tweaks this year.

 

[kevzilla]

Dear Antonio Gates:

Please be Antonio Gates Sunday night. I've been waiting patiently.

Signed,

Guy With Pettigrew on Bye and Who Drafted Lance Kendricks for No Discernible Reason

 

[Mario Kart]

Bush and Amendola are out for me for the upcoming weeks. I only have 21 players on my team so that is going to hurt and may be the end of me. Looks like the larger roster size teams might have advantages. Not sure of my survival.

 

[Saint]

wonder if amendola going down gives any value to steve smith??? Probably not, but i'm hopeful for points.

 

[pizzatyme]

Stafford on bye and Locker... Please let me make it through this week!!!!

3 QBs remember for next year!

to me at least 3 qb's with different bye weeks always seemed to make sense....even if you went big on a stud....kind of a no brainer almost....but you never know I guess

I went Schaub, Luck, Skelton, and Tebow. Looks ugly, but could have 4 starters going within the next few weeks. Who knows?

 

[Das Boot]

Things always change in a hurry in this contest, but right now I'm wishing I had taken a chance on Mendenhall.

With Andre Brown getting dinged and Bradshaw with a fumble, maybe Wilson starts getting more PT - I'm still pretty confident this will happen as the season goes on.

 

[BassNBrew]

Ugly week here. Sitting at 42 points. Top scorer is a kicker, QBs haven't broken 6 pts.

I'll need a cut line around 115 to have a sniff at advancing.

 

[Ignoratio Elenchi]

DAVID WILSON SIGHTING

 

[BassNBrew]

DAVID WILSON SIGHTING

Too bad he won't count for me. Don't need another 10 point guy.

 

[Ignoratio Elenchi]

DAVID WILSON SIGHTING

Too bad he won't count for me. Don't need another 10 point guy.

He's currently counting as my flex. I'm sitting at 128 despite a mostly mediocre day all around. Still have a few players left so I will likely improve on that score, but as it stands now Tony Gonzalez may be single-handedly dragging my team into week 6.

 

[Gawain]

DAVID WILSON SIGHTING

Too bad he won't count for me. Don't need another 10 point guy.

He's currently counting as my flex. I'm sitting at 128 despite a mostly mediocre day all around. Still have a few players left so I will likely improve on that score, but as it stands now Tony Gonzalez may be single-handedly dragging my team into week 6.

I'd love 128, currently at 66.9, but haven't had a QB play yet and have two 0's still to fill at RB and WR. Need some big scoring from the 4pm games.

 

[Ignoratio Elenchi]

DAVID WILSON SIGHTING

Too bad he won't count for me. Don't need another 10 point guy.

He's currently counting as my flex. I'm sitting at 128 despite a mostly mediocre day all around. Still have a few players left so I will likely improve on that score, but as it stands now Tony Gonzalez and Reggie Wayne may be single-handedly dragging my team into week 6.

Updated.

 

[FUBAR]

Last changes were:

1. Luck and Adrian Peterson to Flacco, Reggie Bush and D upgrade from Atlanta to Houston.

2. David Wilson and Earl Bennett out for Mendenhall, Royster, Broyles and grab a 3rd D.

Who were your late changes?

1. Luck and Adrian Peterson to Flacco, Reggie Bush and D upgrade from Atlanta to Houston. :X Although taking Houston D might pay off.

2. David Wilson and Earl Bennett out for Mendenhall, Royster, Broyles and grab a 3rd D.

This seems smart for now.

 

[Mario Kart]

THE CONTEST IS NOW LOCKED.

Useful Links:

LIVE SCORING! - Fellow FBG OrganizedChaos puts this together every year on his own time and on his own dime. He is once again providing this wonderful service for free, so definitely check it out, and if you like it I encourage you to use the donation links on his site to send him a few bucks to show your appreciation.

Ownership stats

Querier

Click here to see your team.

 

[Mario Kart]

I know with the bye weeks and all, scoring tends to go down a little. Is scoring down a lot right now? Not much offense today so far.

 

[RCon]

On the question of how to reconcile for a QB (or any positions) average output vs their volatiity and trying to equate it all back to 'value': I'm a little surprised nobody has brought up the classical mathematical representation of volatility.

Standard Deviation.

...

The biggest issue I see is what we all see with fantasy football. Not enough data points. Sample size is very small - 4 data points per player right now.

Right, standard deviation has been used before but I think after just four weeks it's not going to tell us much. Really, all of this discussion about who the best players are is something that's better left until the end of the season. In general, I'm more interested in roster construction strategy than specific player performance anyway - e.g. "which is better, 2 QBs or 3 QBs," as opposed to, "who is better, Flacco or Newton?" I think it was Modog last year who used standard deviation to track which teams were more likely to put up scores of 200+ points (which is typically what's required to finish in the big money) and showed that, while small rosters had lower average scores than larger rosters, their inherently greater variance made them more likely to finish with a really high weekly score.

With this mindset, it should be a mathematical exercise to optimize an ideal roster- the roster that nets you highest average points with lowest std deviations that fits under $250.

In theory this is doable, but there are like quintillions of possible legal roster constructions. It's an interesting question how to optimize the process of finding the optimal rooster. I've always wanted to but never got around to really giving it a shot.

In a best ball format, I'm not sure low standard deviations are optimal. The player that scores 0 every other week doesn't count those weeks if you have a backup who scores at all.

 

[RCon]

DAVID WILSON SIGHTING

Might be useful here, what with McFadden and LeShoure on bye, and Benson not doing much today. He'll count for me unless Marshawn and one of a few WRs catch him.

 

[Utter Chaos]

I would assume Amendola's injury affect a greater proporion of big-rooster teams than small-rooster teams?

-QG

Amendola owners:

SIZE	OWN	ALL	OWN %18	408	2302	17.7%19	213	984	21.6%20	194	792	24.5%21	203	656	30.9%22	181	581	31.2%23	181	505	35.8%24	141	407	34.6%25	143	315	45.4%26	110	242	45.5%27	88	211	41.7%28	63	153	41.2%29	67	137	48.9%30	114	219	52.1%TOT	2106	7504	28.1%

So he's on way more large rosters than small rosters, relatively, though of course the small rosters that do own him will presumably be much more impacted by the injury than the larger rosters.

That depends on how you look at it. He's on way more large rosters if you're talking about percent ownership but looking at the numbers only there's 1199 rosters between 18-22 players that have him vs 907 rosters between 23-30 players that own him so you could make the point that he's on way more small rosters.

 

[ConstruxBoy]

Ugh, 104. Brutal week by Flacco with Stafford off.

 

[Drunken Cowboy]

I am around 115 with Royal (-7). I am not feeling very good, but I do think the cut will be really low.

 

[Dinsy Ejotuz]

Didn't have any of my seven WRs get into double digits.

138 + Gates-10

Will be lucky to survive I think.

 

[Instinctive]

At 115, need some serious points from Colston, Gates, and Brees (good grief Flacco...) to survive.

I love Dwayne Allen. I think he's my pick of the year, since I was too stupid to take both Mendy and Dwyer.......

 

[placebo]

Yea went with $2 Dwayne Allen as my lone TE behind Graham and it looks like I'll get by with that. Will be using him this week regardless of what Graham does.

 

[DWI]

The live scoring for the Minn-Tenn game seems to have quit. I think im around 150 with Colston and Grahm left.

 

[Deranged Hermit]

Hopefully 159 will be good enough. Reggie Wayne saved my ### this week.