HairySasquatch
Footballguy
day traders for fantasy football....lol That could be an interesting contest. Start with x dollars and whoever earns the most money over the course of the season wins. Price is dictated by performance and add/drops.
Fantasy Football - Footballguys Forums
This is a sample guest message. Register a free account today to become a member! Once signed in, you'll be able to participate on this site by adding your own topics and posts, as well as connect with other members through your own private inbox!
***Official*** 2012 FBG Subscriber Contest Thread (3 Viewers)
- Thread starter Ignoratio Elenchi
- Start date
urbanhack
Fight The Power!
You'll both be out after week 5.
I get what you're doing, I'm just not sure I agree with the methodology. In theory the prices have last years stats and anticipated improvement/decline already built into them. So I would expect that spending $40 on Qbs would net close to the same points regardless of stud/scrub vs 2 mid players when u use last years stats. But the people choosing between say Rodgers + Tannehill and Ryan + Cutler are doing so because they think Ryan and Cutler will outperform their value, and thus, their stats from last year. That in a sense they're more like Stafford last year than they are like Cutler/Ryan from last year. In addition, they get the "security" of a theoretical higher floor each week. What I mean is that Rodgers + Tannehill might produce a higher per week average, but it only takes a poor week on Rodgers bye or on an "off week" for Rodgers to be eliminated from the contest. That of course is debatable, but is a belief that one might use when deciding upon QB strategy.
flc735
Footballguy
no, with this way, you get to decide what kind of production each player will get. if you don't like julio, than use bowe's last years stats
this will vary person to person. i think julio will put up the numbers roddy did last year. if you disagree, pick your own player and find out yourself
it doesn't, and thats what i am showing you. all i can do though is tell you to try it yourself
Chicago Hooligan
Footballguy
Fewer, I think, but still pretty close. I'm in the upper 20s for both years.The new pricing/scoring made me think about some other strategies, like, if you can pick "the right studs," you'll be sitting pretty. But that's the case for every FF league, and in best ball I want more players.Ultimately I'll be counting on the $8 to $15 range guys to make or break me again.
Chicago Hooligan
Footballguy
Of course the next time I log in I'll probably take one look and freak out about my WRs being crappy or something, so who knows.
HairySasquatch
Footballguy
You have sold me at the QB and TE positions where you only start one player....and to a lesser extent at RB where you start two. Where this doesn't work for me is at WR. You have to start three each week. If you choose 3 studs (or even 2), I feel like it leaves you thin numbers wise...either at WR or making the sacrifice at another position.That being said, I'm going to be rolling with a low roster total this year, but I will not be surprised if I'm out of it by week 8....at the same time, I predict that the winner of the whole thing has a roster of 22 players or less. Just like Indiana Jones in the last crusade....you just have to choose wisely. Last year, 4 out of the top 5 finishers had 18 players and that was with higher prices for the studs. I would rather go for the gold with a small roster, than go with a large roster and settle for the bronze.
Last edited by a moderator:
flc735
Footballguy
i mentioned that early on. i said the only position that you could argue is WR because the position is so volatile. i dont agree but i dont disagree with you on thati agree with your last point. 100th place doesn't do anything for me. another factor i have not mentioned is that it will be very difficult to finish in the top 10 over the last 3 weeks with a 26 man roster. you would have to hit on several sleepers, which is less likely than your 4 RBs staying healthy all year
The Kerminator
Footballguy
That's not a bad start. Can you post your week 2 lineup?
northwoods
Footballguy
I'd like to see the week 7 lineup...
kevzilla
Footballguy
Well, Week 7, for sure.
The Kerminator
Footballguy
Got it nailed down except I've got $8 left to spend on 2 more WRs. Which 2 give the best combo?
northwoods
Footballguy
Come on King, if I posted that here then 10,000 other people would put that on their contest entry...On a more serious note, you can't go wrong taking cheap WR's from big passing teams...
Last edited by a moderator:
But then doesn't it just come down to how well you expect players to play compared to last year. What if I don't think what Brees/Rodgers/Stafford/Brady did last year will be duplicated. What if I believe the gap between QB1 and QB7 will be closer than it's ever been?
Also, I'm not sure that Avg Pts per week is the best metric. I think I'd care a lot more about minimum points scored in any given week (not saying that based on your beliefs on players that it'd change anything).
I need to think more about this, but have a bit of head cold so I'm a little slow today.
butcher boy
Footballguy
D Aromashodu MIND Coale DAL
Sake-Bombers
Footballguy
You know what I like? That Julio and Roddy are the exact same price. Julio hype off the charts, but Roddy's been a consistent reception machine. It comes down to owner decision which one to take, if any. Will be interesting to see ramifications, especially in finals.
Mr. Soup Nazi
Footballguy
I'm feeling the most torn at QB. I've gone with about 25 different combos so far with about 8 different players.
Ignoratio Elenchi
Footballguy
I did a very simple analysis of this a couple years ago at the WR position. Compared something like a stud + scrub WR pair vs. every possible combo of two other receivers that cost the same amount of money or less. I seem to recall that it turned out that the "stud+scrub" combos beat the other pairs about 50% of the time. It was basically a coinflip. But that analysis had some significant flaws, and in particular I think that phenomenon is more likely at the WR position anyway. I think at QB, for example, stud + scrub is probably a significantly better option than two mediocre guys.Of course, all of this analysis depends on picking the right players. It doesn't matter if, on average, it would've been better to pick an expensive QB than two cheaper QBs. If you pick the wrong expensive QB, those averages aren't going to keep your team alive.
The Kerminator
Footballguy
PM me.
Yeah, truth be told, I've already got those 2 WRs picked. And my guess is that 10,000 other people already have the same cheap WRs on their roster already.
flc735
Footballguy
come on. lets use some common sense here. if you think brady will thrown for 4,000 yards and 30 td's, then use romo's numbers from last year. stop with the what-if's. if you disagree with the numbers i am using, i have given you the info to do it yourself. if you still don't get it than go prove me wrong with your 28 man rosteri am not using average points per game. i am using samples from last year to simulate the weekly variance in order to avoid exactly what you said.
flc735
Footballguy
i dont think so. for example, i have no idea who you are talking aboutPM me.
The Kerminator
Footballguy
Maybe Northwoods could PM you.i dont think so. for example, i have no idea who you are talking aboutPM me.
LHUCKS
Footballguy
yep, me too...I have switched from 2-3 QBs with varying combinations on a daily basis.
Das Boot
Footballguy
Yep, it's all luck.
I am using common sense. It's based on individual projections. If I think Ryan is going to put up Stafford numbers and Brady is going to put up Romo numbers I'm better off taking Ryan. Now if you want to argue that if I have these expectations I'm best off taking Ryan and Tannehill and spending the remaining money elsewhere that's one thing. But by grabbing someone else who I have high expectations for, say Cutler, I get some insurance in case I'm wrong about Ryan plus the added bonus that I might be right about both of them. If I'm wrong and they only perform as expected, yes I'm probably behind the people that took Rodgers and Tannehill, but it's not going to cripple my team the way it would had I picked Ryan and Tannehill and was wrong about Ryan.'flc735 said:
Ignoratio Elenchi
Footballguy
I haven't been following the thread closely and am not going to go back and figure out exactly what flc is doing, but he's right that simple averages won't capture the variance which is a crucial component to determining the usefulness of a player. If I'm looking at two WRs who are both projected to score 120 points this season, for example, I'd probably rather have the one that will score 20 points 8 times and 0 points 8 times, than the one that will score 10 points 16 times. Just saying that he averages 10 PPG doesn't account for how frequently he might actually crack into your starting lineup and contribute to your total score. When you're looking at actual stats from last year to determine which combos were better than others, you have to account for this feature. For example, say I had Calvin Johnson last year, who was on bye in week 8 (I don't know if that was his actual bye week, let's just pretend that it was). Would you rather pair him with a WR who scores 15 points a week all season long, or a WR who scored 0 points all year long, except blew up for 25 points in week 8? If you're just looking at average scores, you'd think the 15 point-per-week WR was vastly more useful than the 25 point-total WR, but your conclusion (in this isolated scenario) would be wrong.
This is what I did a couple of years ago, I paired up every combination of WRs and figured out their actual (i.e. max) score for each week of the season. As I mentioned earlier, I recall that the conclusion was that it was basically a coinflip whether "expensive + cheap" WR combos outperformed "two middle of the pack" WR combos. But as I also mentioned, that kind of analysis was sufficiently flawed that I wouldn't draw any strong conclusions from it.
I think WR is a position where it doesn't make sense to pay top dollar for the "elite" guys, you can probably approximate their weekly production while reducing your injury/bye risk by taking a greater quantity of cheaper WRs. But I believe the opposite is true at other positions.
flc735
Footballguy
i think julio will average 17 ppg this seasoni go here and see roddy scored about that last year, so i copy his week by week stat line and pretend it is julio. this is the best way that i know of to simulate the natural week by week variation.'Ignoratio Elenchi said:
then i randomize is, move a couple weeks here, move a few over there ect... i never change the numbers, i just reorder them and match the bye weeks
then i do the same with, say a $3, $10 and $17 players.
then i compair julio+$3wr with a $10 and $17 guy in a start 1 per week scenario
then i repeat this process a few times over, changing the weeks, the players, player prices ect...
the results overwhelmingly showed that as far as qb, rb, te, k and d/st went, 1stud+scrub > 2 mid players
it also showed that 1 stud was better at WR, but to a much lesser degree
i don't feel confident in my ability to predict the top kickers and defenses, so i am sticking with a bunch of $3 guys despite the numbers
then i used my super cool excel sheet, which picked out my flex play each week when i entered a whole team. the results still showed the same thing
ignor, it's not always a coin flip when choosing between the two options. half of it depends on the player prices. this is my first year but from this thread, it seems like the difference between studs and mid guys is a lot smaller. if so, that would point towards studs over more mid guys
if, say, the top players were $45, i would be telling everyone to make 30 man rosters.
Last edited by a moderator:
Ignoratio Elenchi
Footballguy
I never said it was. In fact, I don't see where you and I really disagree at all.Regarding your comment about the price differences, here's an illustration comparing the price distributions from 2011 and 2012. I think you're right that the shallower dropoff and lower prices overall at RB and WR this year means that studs are more useful this year than in years past, though I still believe that to a certain extent, quantity > quality at the WR position. Despite the lower prices, it's still not really feasible to construct a team that literally has studs at every starting position that's also strong enough to last the whole year. You have to make some concessions somewhere, and I think WR is probably the place to do it.
The question isn't simply, "Should I take a $28 WR and a $2 WR, or two $15 WRs?" It's actually, "Should I take a $28 WR and a $2 WR, or two $15 WRs, or three $10 WRs, or four $7-8 WRs, or some other combination thereof?" The reason small rosters get slaughtered every year is not because they don't have enough studs, it's because they don't have enough quantity to overcome the inherent variance in weekly scoring, injuries, bye weeks, etc. All it takes is one bad week. Lower prices on the top players doesn't necessarily fix this problem, though it does presumably make it somewhat easier to address.
So you need to not only compare "Julio + $3" vs "$10 + $17", you also need to compare "Julio + $3" vs. "$9 + $9 + $9" vs. "$12 + $5 + $5 + $5" etc. etc. And you can't just look at "start 1" scenarios, since "Julio + $3" may only produce a single startable score per week, while "$10 + $17" or "$9 + $9 + $9" may more consistently produce two or even three startable scores per week (even if they don't match the single max score of the Julio pairing). You're going to be starting 3-4 WRs every single week of the season, so just comparing pairs as you're doing (and as my old flawed analysis did) isn't really answering the important question.
I won't be surprised this season when the overwhelming majority of entries are 18-20 players and packed with studs all over. And I won't be surprised when the overwhelming majority of them are eliminated before the finals. I still believe a winning strategy involves a larger roster, this year it's just easier to pack better players onto a larger roster.
northwoods
Footballguy
I'm re-thinking my large roster bias and will probably opt for something around the mid-20's roster size this year.
As predicted, my latest iteration has 25 players. Locked in barring last minute changes due to injuries, etc...
QB - 2 @ $33
RB - 7 @ $88
WR - 9 @ $87
TE - 2 @ $24
K - 3 @ $9
TD - 2 @ $10
This year, I have 1 less TE, 1 less K and 1 less TD than last year. Going to roll with quality over quantity at those positions and see what happens...week 7 byes will probably decide my contest fate. Need to make it thru to week 8!!
As predicted, my latest iteration has 25 players. Locked in barring last minute changes due to injuries, etc...
QB - 2 @ $33
RB - 7 @ $88
WR - 9 @ $87
TE - 2 @ $24
K - 3 @ $9
TD - 2 @ $10
This year, I have 1 less TE, 1 less K and 1 less TD than last year. Going to roll with quality over quantity at those positions and see what happens...week 7 byes will probably decide my contest fate. Need to make it thru to week 8!!
)
Ignoratio Elenchi
Footballguy
There's a ton of debate about large vs small, and I always fall on the "large" side, but I actually believe 25-27 has been the sweet spot for roster size. I consider that "large", especially considering the sheer volume of entries that are 20 or less.I, too, have been re-thinking my large roster bias due to the pricing changes this year, but when I really think about it here's where I end up:
What does relatively cheaper studs mean for small rosters? It means they can have presumably better players on their team than in previous years. If you were going with 18 players last year, you couldn't literally have the top players at every position, you had to make some kind of sacrifices somewhere. This year I assume you have to make fewer such sacrifices. But as I mentioned in a post above, the problem with small rosters has never been a lack of studs. So lower prices don't especially help the small rosters, because they're still small, and being small has inherent problems that aren't fixed with better players.
What does relatively cheaper studs mean for large rosters? It means they can have presumably better players on their team than in previous years. And this has always been the common knock on large rosters - that they lack the elite firepower necessary to put up big point totals when they're needed. Well now that's less of a problem. You can still have the advantages that come with a greater quantity of players, but without having to eschew so much top-end talent to do so.
So if I thought large rosters were a good idea last year, the change to the pricing structure should only strengthen my conviction that large rosters are a good idea this year. Again, when I say large I don't necessarily mean 30. My own submission will probably once again end up in the ballpark of 26 or so.
HairySasquatch
Footballguy
Last year 4 out 5 (2nd place had a large roster) had the minimum roster size of 18. Maybe it was a fluke last year. Average number of players in the top 30 last year was QB 2.3, RB 5.8, WR 7.1, TE 2.8, K 2.8, and DEF 2.7. So the average size of the top 30 was 23.5 total. My take from this is that a lot of small rosters will be eliminated with poor play on any given week or injury to a main guy. But at the same time, if a small sized full of studs makes it to the finals and all of the guys are healthy, they will do very well.
Ignoratio Elenchi
Footballguy
This has always been the theory pushed by the small-roster proponents. I think it might have some merit; I think a stronger argument relies on the idea that smaller rosters exhibit greater variance, so they are more likely to end up at the top (and bottom) of the final standings.Last year's top five seemed to support the small-roster theory. However, if I recall correctly, in 2009 and 2010 the top of the final leaderboard was dominated by large rosters, so the results have been inconclusive at best. My opinion continues to be that even if small rosters possess some kind of theoretical advantage in the finals, it's not enough of an advantage to outweigh their other disadvantages when building a roster in August. There's also too much reliance on price as an indicator of production in these discussions. It doesn't matter who's a "stud" on August 1 when Dodds sets the prices. It matters who's a stud when week 14 rolls around, and the guys who are studs in week 14 aren't always the guys who were considered studs on August 1. The idea that small rosters are "full of studs" and large rosters aren't is misguided, I think.
I think I might know the 2 QBs, and the 2 TEs as I have all 4 of them on my roster (with 1 extra TE). QB: 2 - $30RB: 9 - $63WR: 8 - $91TE: 3 - $48K: 3 - $9Def: 3 - $9 Probably need to trim the number of players on this one, but I think my scatter-shot RB strategy will stay. I tried this last year and it didn't work well, but I can't seem to let go of it. I can tell myself stories on all these cheap RBs that just 1 injury and they will be studs.
The only problems I am having now is going with 3 QB's, 3 K, and 3 DEF wondering if I should go to just 2. Also I have been having a hard time deciding on TE. I like the idea of going with 5 guys and rotating them rather then taking 1 stud TE. My RB's and WR's are locked in now, along with my top 2 QB's/K/DEF. Just tweaking things now. Really like my team, it has both studs and depth
Dinsy Ejotuz
Footballguy
I was max-roster-size-or-die the last few years, but am consistently coming in with 26-28 this year. Also think this year is much more comfortable in terms of building rosters I like. In previous years it seemed much harder.
Scatter gun RB worked well for me last year (finished 7th or 8th overall with 30 man roster). Bush, Bush, Barber, Jones, Wells. Burleson stepping ob and leaving 10-12 points off the board prevent a top 5 finish.eta: The QBsw aren't hard to figure out but shoot me a pm if you want to try at the TEs.
Last edited by a moderator:
I agree with most of what you said but wonder if you've factored in the fact that while the stud pricing has dropped, the mid tier remained the same. Last year it looked like Dodds used auction pricing, this year it looks linear.The other kick in the pants is that there aren't any $2 defenses and kickers, that's knocked 2-4 spots off many potential large rosters.
HairySasquatch
Footballguy
It shall be interesting 
...I used to be a large roster guy. My experience has been if I had a bad week/injury from one of my main guys during weeks 11-13, there was a good chance that I was going to be eliminated no matter what size of roster I had. It all comes down to choosing the right guys who will score big consistently....either by higher priced guys who have done it before or by the shotgun method.
...I used to be a large roster guy. My experience has been if I had a bad week/injury from one of my main guys during weeks 11-13, there was a good chance that I was going to be eliminated no matter what size of roster I had. It all comes down to choosing the right guys who will score big consistently....either by higher priced guys who have done it before or by the shotgun method.
Ignoratio Elenchi
Footballguy
This is a good point, though I'll link my illustration again showing that prices are down at RB and WR pretty much across the board (except for a moment where they meet around WR40). Studs do appear relatively cheaper this year - that is, their prices dropped off more than the mid-tier players - but the mid-tier players still cost less than they did last year. RBs in the 18-24 range are $4-6 less than last year, RBs in the 36-48 range are $2-3 cheaper than they were last year, WRs in the 24-36 range are about $2 less than last year, then they are equal around WR40 but then you get a $2 discount off of last year's prices pretty much all the way from there on down to WR90 or so. The point being that if I wanted to go with a large roster last year, and could only afford to spend $17 on my top WR, that would get me WR30. This year, $17 gets me WR24. As I said, small rosters can get better players than they used to, but the quality of players was never the real problem for small rosters. They're still going to face the same challenges small rosters faced in past years. Large rosters can get better players than they used to, which was arguably the real problem with taking a large roster. So I'd argue that the large rosters benefit more from the decreased pricing than small rosters.
This is frustrating. I'm having a hard time figuring out how to squeeze 16+ kickers on my roster at these prices without sacrificing too much at all the other positions.
HairySasquatch
Footballguy
Lets talk weekly cut offs.
Last year weeks 1 through 13:
(1) 0.00 (2) 138.90 (3) 142.00 (4) 142.10 (5) 104.90 (6) 112.35 (7) 108.15 (8) 128.55 (9) 140.25 (10) 147.15 (11) 133.90 (12) 154.55 (13)164.25
Any thoughts on what they will look like this year? Just trying to figure out if I have a realistic shot at getting through a few difficult weeks with my current roster. If they are relatively close to last year, I feel comfortable.
Weeks I see that should be on the lower side: Week 6 (Lots of popular choices this week with NO, CAR, and CHI on byes), week 7 (6 teams instead of 4 on bye this week), and week 8 (the two best RBs on bye).
Last year weeks 1 through 13:
(1) 0.00 (2) 138.90 (3) 142.00 (4) 142.10 (5) 104.90 (6) 112.35 (7) 108.15 (8) 128.55 (9) 140.25 (10) 147.15 (11) 133.90 (12) 154.55 (13)164.25
Any thoughts on what they will look like this year? Just trying to figure out if I have a realistic shot at getting through a few difficult weeks with my current roster. If they are relatively close to last year, I feel comfortable.
Weeks I see that should be on the lower side: Week 6 (Lots of popular choices this week with NO, CAR, and CHI on byes), week 7 (6 teams instead of 4 on bye this week), and week 8 (the two best RBs on bye).
LHUCKS
Footballguy
this is what ya'll should be looking at.
It won't be the same this year.Scoring rules changed, bye weeks are different, pricing structure changed.Any way the dd has me basically over 140 every week except wk 5-8 where I'm in the 125 to 130 range.
HairySasquatch
Footballguy
Do you feel safe with having a 140 or above average? I don't know how close to 140 you are but that seems to be cutting it really close weeks 9-13.
