The Man With No Name
Footballguy
For you database guys, is there a way to tell what teams still alive has the lowest and highest comulative scores through 8 weeks?
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*** 2011 FBG Subscriber Contest Thread (2 Viewers)
- Thread starter Ignoratio Elenchi
- Start date
And, again.Finished with 138.15.ETA: Had I made it through (out in week 3), I would have used 27 of my 30 players. The only guys I would not have used yet are Danny Amendola, Evan Moore and Harry Douglas.
Last edited by a moderator:
HALtheCOW
Footballguy
I have always agreed with the large roster theory. I guess I was just stunned with how many more small rosters there were that entered the contest. I am just curious if the sheer number of entries will have counteracted the bye/injury analysis. I love this contest for this and many other reasons. And I am lucky to be alive. Should be gone soon.
petenice15
Footballguy
I'm out by .1. That is the worst way to go out. 1 yard...amazing.
Does anyone know if stat corrections every happen in this contest?
Does anyone know if stat corrections every happen in this contest?
Touchdown There
Footballguy
No stat corrections for this contest.
Organized Chaos
Footballguy
Team ID - Scoretop 5 teams
Code:
100511 1531.25108488 1515.25108423 1505.00103346 1479.05107444 1475.65
Code:
101339 1035.25110300 1035.55104654 1056.95103315 1058.55110752 1063.85
QuizGuy66
Footballguy
I'm at 1360.05. Not too bad.-QG
apalmer
Footballguy
Right behind you at 1354.05
Last edited by a moderator:
Ehh 1356.55 seems pretty decent.
Tennessee_ATO
Footballguy
I tend to disagree with the assessment once you get through the byes. Unquestionably the smaller rosters expose owners to heightened injury risks, but a properly constructed roster shouldn't be terribly problematic for byes. Assuming a high percentage don't pay proper attention to byes (always good assumption for some reason [see NYJ/GB guy]), however, a disproportionate number of those rosters will be eliminated. I don't think those with 25-30 man rosters necessarily pay any more attention to byes as a whole, they just have a better chance to dumb-luck their way into have contributors every week.All of that said, the high percentage of eliminated 18-19 man rosters doesn't come as a surprise to me. I would, however, be surprised to see the remaining 18-19 man rosters not fare disproportionally better than their 28-30 man counterparts after byes. This is due to the fact that the more expensive players typically outperform the cheaper players. Not always, of course, but presumably many of the underperforming expensive guys won't be on many of the final 250 rosters. At the end of the day, the guy with 3 stud RBs and/or 4 stud WRs is going to throw-up consistently big numbers week-in and week-out. The guy with a bunch of less reliable players may be able to throw-up an equally big number in a given week, but will be less likely to do it consistently.That's just my 2 cents, which probably isn't worth that much. This is the first year I've really been serious about this contest (although I've done it every year in a half-hearted manner), so I'm not some grizzled vet. That said, I played with different scenarios and kept coming back to the short roster built around top-flight WRs as the best option in my view.currently18-20 man rosters 22.4% still alive28-30 man rosters 44.2% still alive
SeniorVBDStudent
Footballguy
We agree in theory on short roster, but I would argue that you need to pick from the right qbs, rbs and tes as there is a bigger pool of potentially higher scoring wrs, pks and defs to choose from randomly. This is similar to a "stud rb" argument. It will vary from year to year depending on a million variables, but in general the best scoring qbs, rbs and tes are more limited resource pools than wrs, pks and defs.Which is why I have 6 wrs, 3 pks and 3 defs to complement Brady, Adrian Peterson and Aaron Hernandez. In hindsight, I would have added a Pettigrew or a Daniels to my TE corps, but cest la vie.
SeniorVBDStudent
Footballguy
To carry this a step further, someone other than me(!) could analyze the entire history of the contest and compare the points per dollar of all the players in each position, and my expectation would be that more "value" would exist in the middle and bottom of the WR, PK and DEF pools than in the QB, RB and TE pools. Perhaps I'm mistaken, but thats my perception.
radiohead417
Footballguy
1109.55....Clearly I'm going to win this whole thing when Blaine Gabbert, Chris Johnson, Chad Ochocinco and Steve Smith (PHI) blow up in weeks 15 and 16.
Meh...1108.70 doesn't appear to be too good. I may not be around much longer.
Tennessee_ATO
Footballguy
In the past I would have said that RBs are the key, but I'm just not sure about that any more. The salaries of the RBs I viewed as "studs" were so much higher than the corresponding "stud" WR salaries that I viewed the WRs as a better investment. Also, in today's NFL it seems that certain WRs are more consistent than the vast majority of RBs. Plus, 3 WRs count every week so the position has 50% more guaranteed scorers each week. I've been mulling over a new philosophy on the relative value of QBs in current fantasy scoring, which I applied here: there are a billion guys out there who are within a very narrow band. As I said elsewhere, Tebow can be straight-up atrocious but still score in the top 12? That tells me that scoring for the QB position is off. I think I underestimated the impact of this contest's TD pass scoring (6 pts per) though, which I should have weighted appropriately. As I posted before, I think QB will be the position that ultimately is my undoing (Rivers, Cutler). On TEs, I think you're absolutely right on that limited resource. I have Daniels and Celek -- 2 guys I thought would exceed expectations and return value. I still think that's true on both, but in terms of absolute value I should have gone more "elite" for 1 of those 2 slots.
Stinkin Ref
IBL Representative
the value at TE has been crazy this year with guys like Graham ($15) Hernandez ($10) Gronk, etc producing WR1/RB1 type numbers at a discounted price....those larger roster guys that kind of planned on snagging some of these guys and almost committing to having their Flex score come from the TE position, might be able to hang on a little with the the smaller roster guys that make it through the bye weeks...some of the smaller roster guys may have some of these guys as well, but they may be using them as the TE only and having to find that flex score from somewhere else....could narrow the advantage/gap a little....
I took what I thought was a somewhat interesting route on the QB position. I only wanted to have 2 to save money for other positions. I took a guy that I thought would be really good this year but had a question mark, Stafford, and a guy that I thought had a pretty high floor so that even if Stafford got hurt I wouldn't be putting up zeroes. My second QB is Eli and while he's had a much better season than I would have guessed so far, he's been invaluable to my team scoring 3 time and putting up a respectable score last week. Eli carried a really small price tag for his consistency, especially to pair with Stafford. Even counting only Eli's scores I'd still be in this thing because he has a high floor.
Tennessee_ATO
Footballguy
I kinda already have with my 18 man roster. I realize that last year -- the 1 and only year with rosters that large -- saw the 18-20 man rosters not fare as well, proportionally, as the 28-30 man rosters. But last year the 18-19 man rosters were hardly down to less than 20% of the 250 finalists (they were over 32% of the finalists). It's not like you've been through it a couple of years now, really only 1. Larger rosters allow for more "Forrest Gumping" for sure, on that there is no debate. 1 year is hardly a sufficient sample size to draw any meaningful conclusions, statistically speaking anyway.
Last edited by a moderator:
BassNBrew
Footballguy
They were 80-90% of the original entries.The reality is there there isn't much difference in some 18 man and 30 man rosters. Build an 18 man roster and then dump one $30 dollar player for 13 $2-$3 players. In best ball 13 > 1.
Last edited by a moderator:
inequality fixed.
Tennessee_ATO
Footballguy
According to the thread from last year's contest, 18-19 man rosters were "almost 46% of the total entries", not 80-90%.That 13 > 1 formula certainly depends on the 13 and the 1, doesn't it? I can get 12 of my buddies, several of whom are scratch golfers, and I'll take Luke Donald for 3 consecutive holes. And that analogy doesn't really factor in value of costs and position play, which is what undermines it. Sure, knowing what we know now, we may be able to select 13 players that can amass the same stats in a best ball format as a $30 player. But 13 at the same position? That a tougher proposition. And doing it before the season starts? Tougher still.
There is no doubt that looking back the "best" players to roster will not all be the "studs". I doubt any 18 man roster is comprised of purely "studs" anyway. I'll wager that there is a higher percentage of "studs" who are at the top of the heap than lower-priced players.
I don't pretend to have a crystal ball into this, but I had a hard time identifying 3, let alone 13, players I felt would be as productive as Jennings for his salary in a best ball. Only 575 agreed with me. Ditto Wallace @ $23. 2017 agreed with me on that pick. Only 90 agreed with both of those decisions on my part. If I'm right on both of those calls, I'm in great shape (assuming, of course, I did my homework properly and positioned myself to survive Wallace's bye). If I'm not right, I'm not in good shape. It's not terribly different from anyone else's decisions regardless of roster size.
I wrote a program earlier this season to build my team, seen here:
http://subscribers.footballguys.com/contest/2011/100535.php
Based on what I observed team size wasn't as big of a factor as balancing your bye weeks. Balancing those goes beyond just making sure you have enough players in, you also need to make sure the ones you have playing have favorable match-ups. Even taking that into account you'll notice the tendency is to avoid studs and go for second tier players with good season coverage. The thing about studs is one bad week for a handful of them will put you out of the contest any given week. By spreading out to several second tier players you improve your odds of at least one coming through week-to-week. Plus the obvious decimation bye weeks can cause.
This contest is all about making it to the final 250 and hoping, since by the end of the season who knows who will be hot and who knows who will have good weeks. You can bet any team at that point has a shot because to make it that far they need solid teams, so it doesn't matter who you got at what cost the only thing that matters is did you build a team that can get that far.
As an aside, given the scoring rules the obvious place to invest is in QBs in such a way that you have a high chance of getting 20+ points a week from that position. So far this season the worst I've gotten from Stafford/Rivers is 15% of the cutoff from that position, several weeks they amount to 25% of the cutoff. You need solid numbers from that position from as few people as possible to make sure they pad you enough to get the rest of your team through.
http://subscribers.footballguys.com/contest/2011/100535.php
Based on what I observed team size wasn't as big of a factor as balancing your bye weeks. Balancing those goes beyond just making sure you have enough players in, you also need to make sure the ones you have playing have favorable match-ups. Even taking that into account you'll notice the tendency is to avoid studs and go for second tier players with good season coverage. The thing about studs is one bad week for a handful of them will put you out of the contest any given week. By spreading out to several second tier players you improve your odds of at least one coming through week-to-week. Plus the obvious decimation bye weeks can cause.
This contest is all about making it to the final 250 and hoping, since by the end of the season who knows who will be hot and who knows who will have good weeks. You can bet any team at that point has a shot because to make it that far they need solid teams, so it doesn't matter who you got at what cost the only thing that matters is did you build a team that can get that far.
As an aside, given the scoring rules the obvious place to invest is in QBs in such a way that you have a high chance of getting 20+ points a week from that position. So far this season the worst I've gotten from Stafford/Rivers is 15% of the cutoff from that position, several weeks they amount to 25% of the cutoff. You need solid numbers from that position from as few people as possible to make sure they pad you enough to get the rest of your team through.
Touchdown There
Footballguy
It is not about "balancing" your bye weeks. You actually want to unbalance them to be heavy on weeks when the cut line will be lower. The cut line in week 5 was 104.9.Overall, I think your method is too much computer work, too little gut.
Touchdown There
Footballguy
And to expand on that, this means you have to figure out the popular selections from the posts in this thread before the contest lock date. If you know which selections will be heavy on the entries, you can do a pretty good job of estimating which weeks will have lower cut lines.
But, you aren't looking for 3 or 13 players you feel would be as productive as Jennings for his salary in a best ball. You are looking for a group of 3 (or 13) players that you feel, on any given given week, will net you one score that is greater than Jennings. You aren't looking for 13 individual players that will outscore Jennings, individually, at the end of the year. The argument against the 18 studs isn't necessarily that the 18 studs won't score the most fantasy points on an individual basis. It's that they are likely to have dud games or get injured and that you have a good chance to cover those dud games and injuries and even outscore the stud as a group over the course of the year.
At least, that was my rationale anyway for going with 30.
No doubt. I'm not trying to say my method is perfect, but then no method of prediction is.As for this being "too much computer work and not enough gut", I'm more interested in the programming aspect of picking a good team and the challenges that presents.
Most of the talk in here is about proper balance between team sizes and player spend anyway, so I imagine there is some interest on what a little creative computer programming can reveal.It is a good point that predicting the cut-offs is needed to really balance a team, this year I only optimized to get the highest score possible each week with later weeks being weighted higher than earlier weeks in terms of overall team value (due to the percentage of teams cut, and the importance of scoring well in the playoffs). Next year I might try factoring in player predictions against value and bye weeks to guess a possible cut-line.Unfortunately I doubt I'll ever be able to predict who will get injured. :(Edit: Also, you don't need to figure out possible cut-lines by thread posts. It's pretty easy to predict high ownership players as they normally have high predicted-points/cost ratios. You can spot them pretty easily in excel, and most people who play this game pick the value players quite a bit.
Last edited by a moderator:
WisWolvrns
Footballguy
You also have to take into account the pricing structure. Since a team can not be filled with 18 studs. Albeit, then it brings back last years discussion to try and define what a stud is...
Overall it's a price point decision. Do I take Ray Rice for $35 or take Felix Jones for $15, Hightower for $15 and Tate for $5? At the start of the season, you don't know which is best. If Ray Rice got hurt, Foster never overcame his hammy issue and Hightower kept his top RB spot in Washington, many different teams would be left right now.
In my view, if you pick any $30+ cost players, you are putting 12%+ of your total salary cap into one player. That player better be scoring 20+ points a week, else you will be in a hole any given week (which is an auto hole during their bye week) that you have to dig out of with other players scoring more than usual.
The problem I see with some rosters are people banking on $3 or $6 players to be one of their 10 starters each week. Those rosters are poorly devised I feel. Since those players are typically all or nothings and shouldn't be planned on as week to week point scorers. Teaming one of those with a Ray Rice makes both score around 10 pts/week, where as the Jones/Hightower/Tate person could get 15 pts from Jones and Hightower and if Tate kicks in, it's a bonus flex player. Which would outscore a 'stud' and 'dud' in a given week. Since it's best ball.. one dud week is all it takes to knock out the best looking team on paper if it doesn't have depth.
So needless to say, I'm on the 30 roster size, trying to fill my top 10 starter spots with players of known quantity, but not necessarily studs, focus in on TE's for my flex due to this year's price/points value they had, and throw in 3 PKs/3 Defs for as cheap as possible... albeit that strategy has worked well mostly due to the cheap Buffalo defense scoring big most weeks. But I picked all 3 of the $2 defenses, figuring most teams will pick at least one of them as a backup, so if I picked all 3, I would cover all bases for any week that one of the 3 scored well so as to not lose ground to anyone else with a $2 defense...
Overall it's a price point decision. Do I take Ray Rice for $35 or take Felix Jones for $15, Hightower for $15 and Tate for $5? At the start of the season, you don't know which is best. If Ray Rice got hurt, Foster never overcame his hammy issue and Hightower kept his top RB spot in Washington, many different teams would be left right now.
In my view, if you pick any $30+ cost players, you are putting 12%+ of your total salary cap into one player. That player better be scoring 20+ points a week, else you will be in a hole any given week (which is an auto hole during their bye week) that you have to dig out of with other players scoring more than usual.
The problem I see with some rosters are people banking on $3 or $6 players to be one of their 10 starters each week. Those rosters are poorly devised I feel. Since those players are typically all or nothings and shouldn't be planned on as week to week point scorers. Teaming one of those with a Ray Rice makes both score around 10 pts/week, where as the Jones/Hightower/Tate person could get 15 pts from Jones and Hightower and if Tate kicks in, it's a bonus flex player. Which would outscore a 'stud' and 'dud' in a given week. Since it's best ball.. one dud week is all it takes to knock out the best looking team on paper if it doesn't have depth.
So needless to say, I'm on the 30 roster size, trying to fill my top 10 starter spots with players of known quantity, but not necessarily studs, focus in on TE's for my flex due to this year's price/points value they had, and throw in 3 PKs/3 Defs for as cheap as possible... albeit that strategy has worked well mostly due to the cheap Buffalo defense scoring big most weeks. But I picked all 3 of the $2 defenses, figuring most teams will pick at least one of them as a backup, so if I picked all 3, I would cover all bases for any week that one of the 3 scored well so as to not lose ground to anyone else with a $2 defense...
Last edited by a moderator:
Tennessee_ATO
Footballguy
That is precisely what you are doing. It's a 1 vs. 13 (or 3) in a best-ball format. I never said anything about any player having the point total of Jennings at the end of the season. I've also said all along that byes represent the biggest hurdle for the smaller rosters, and that I would think the smaller rosters would be disproportionally thinned by byes. I think in any given week 1 of the 13 (or 3) can outscore Jennings (and certainly will in Jennings' bye). Once you get through those and into the finals, however, that's when I believe the smaller rosters (properly constructed of course) have the potential to outperform the larger rosters.
If Jennings scores 72 in weeks 14-16, your best ball from your 13 (or 3) each week has to total 72 or more to "win". I've never said that can't be done, only that predicting it at the start of the season largely requires dumb-lucking into the right recipe (if it exists at all -- after all Jennings could score so well those 3 weeks that no combination of players with salaries that are </= his would outscore him in the aggreggate).
FWIW, I never thought Jennings would be the highest total point scorer amongst WRs. I thought (and still think) he represents the most likely to post a consistently "helpful" score on a weekly basis.
Tennessee_ATO
Footballguy
I agree with the concept, but simply executed a little differently (although I think I had a flawed plan at the TE position as I said before). No player on my team cost $30, and I questioned the worth of players priced that high. I wrestled mightily with Jennings @ $27, but ultimately concluded he was worth it. I went primarily in the $12-$23 range for my team aside from Def and PK. Tate ($3), Crabtree ($10), and Celek ($7) are the only position players with salaries less than $12 on my team.
BassNBrew
Footballguy
Here's a hole or a hurdle to overcome with your theory. The 13 v. 1 wouldn't all be at the WR position. I'd grab 2 cheap defenses and 2 cheap kickers with four of those 13 players. Now I don't have to rely just on the WRs to catch Jennings, I'm getting opportunites at two other positions to attack your stud's talent advantage.
So we now have 17 identical players. I've got your defense matched plus 1-2 more shots at a better score plus your kickers matched plus 1-2 more shots at a better score. You get Jennings and I get 9 longshots. I think I win more often than not. Even if I don't win, one of 30 player buddies will certainly have you covered.
So we now have 17 identical players. I've got your defense matched plus 1-2 more shots at a better score plus your kickers matched plus 1-2 more shots at a better score. You get Jennings and I get 9 longshots. I think I win more often than not. Even if I don't win, one of 30 player buddies will certainly have you covered.
Last edited by a moderator:
bicycle_seat_sniffer
Smells like chicken
not sure how my pile of 30 chumps is still alive:
BSS
This entry is still alive.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
---------------------------------------------------------------------------------------------------------------------------------------------
Matt Schaub $20 15.00 23.70 35.65 13.00 30.80 17.30 26.80 23.45 bye
Matthew Stafford $18 32.25 37.80 30.80 22.90 22.05 27.55 15.15 33.45 bye
---------------------------------------------------------------------------------------------------------------------------------------------
Felix Jones $23 14.10 3.50 17.00 10.10 0.00 4.30 0.00 0.00
Chris Wells $16 18.20 15.30 0.00 31.80 12.00 0.00 4.20 14.30 out? out? out?
Tim Hightower $14 17.20 11.10 16.50 3.30 0.00 0.00 8.80 0.00 out? out? out? out? out? out? out? out?
Willis McGahee $11 6.00 17.10 12.90 12.30 12.50 0.00 7.60 0.00 out?
Pierre Thomas $7 8.80 6.50 4.80 9.90 11.70 2.10 15.00 11.40 bye
Jerome Harrison $5 2.70 0.60 1.60 0.00 0.00 0.00 0.00 0.00 bye
Ben Tate $3 17.60 15.50 9.50 2.00 0.00 4.10 10.40 4.20 bye
Derrick Ward $3 9.90 0.00 0.00 0.00 0.00 0.00 0.70 0.00 bye
---------------------------------------------------------------------------------------------------------------------------------------------
Steve Johnson $19 16.60 23.60 23.40 9.80 6.90 14.90 0.00 11.70
Roy Williams $11 9.50 0.00 0.00 2.50 2.10 8.00 15.90 0.00
Plaxico Burress $10 17.20 0.00 14.50 6.30 7.20 2.60 24.50 0.00
Lee Evans $10 0.00 6.50 0.00 0.00 0.00 0.00 0.00 0.00
Nate Burleson $8 13.00 16.30 3.20 3.60 2.00 13.40 2.10 7.20 bye
Danny Amendola $8 9.50 0.00 0.00 0.00 0.00 0.00 0.00 0.00 out? out? out? out? out? out? out? out?
Jordy Nelson $7 19.70 15.40 7.00 20.10 4.70 18.40 9.20 0.00
Jacoby Jones $6 7.30 7.80 1.00 0.00 1.90 17.60 4.30 8.90 bye
Bernard Berrian $5 0.00 2.70 0.00 3.00 0.00 10.40 0.00 0.00 bye bye bye bye bye bye bye bye
Denarius Moore $3 0.00 28.10 15.70 10.90 0.00 1.60 1.40 0.00
Dexter McCluster $2 11.70 8.90 11.20 6.80 2.80 0.00 4.50 5.80
---------------------------------------------------------------------------------------------------------------------------------------------
Greg Olsen $9 13.80 4.90 22.20 18.50 12.60 11.70 1.90 19.30 bye
Todd Heap $8 7.00 2.30 15.10 10.10 0.00 0.00 0.00 0.00
Lance Kendricks $6 3.30 4.10 4.70 9.30 0.00 13.10 0.00 1.70
---------------------------------------------------------------------------------------------------------------------------------------------
Shaun Suisham $3 1.00 6.00 16.00 4.00 8.00 5.00 17.00 12.00 bye
Jay Feely $2 4.00 3.00 6.00 9.00 7.00 0.00 2.00 12.00
Phil Dawson $2 5.00 9.00 6.00 12.00 0.00 7.00 12.00 7.00
---------------------------------------------------------------------------------------------------------------------------------------------
San Diego Chargers $5 4.00 2.00 3.00 7.00 2.00 0.00 12.00 11.00
Cleveland Browns $3 4.00 6.00 9.00 2.00 0.00 4.00 7.00 1.00
St. Louis Rams $3 5.00 5.00 4.00 4.00 0.00 3.00 4.00 16.00
---------------------------------------------------------------------------------------------------------------------------------------------
TOTAL 162.55 173.70 185.05 155.40 108.40 124.35 141.50 140.05 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
CUTOFF 0.00 138.90 142.00 142.10 104.90 112.35 108.15 128.55 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
BSS
This entry is still alive.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
---------------------------------------------------------------------------------------------------------------------------------------------
Matt Schaub $20 15.00 23.70 35.65 13.00 30.80 17.30 26.80 23.45 bye
Matthew Stafford $18 32.25 37.80 30.80 22.90 22.05 27.55 15.15 33.45 bye
---------------------------------------------------------------------------------------------------------------------------------------------
Felix Jones $23 14.10 3.50 17.00 10.10 0.00 4.30 0.00 0.00
Chris Wells $16 18.20 15.30 0.00 31.80 12.00 0.00 4.20 14.30 out? out? out?
Tim Hightower $14 17.20 11.10 16.50 3.30 0.00 0.00 8.80 0.00 out? out? out? out? out? out? out? out?
Willis McGahee $11 6.00 17.10 12.90 12.30 12.50 0.00 7.60 0.00 out?
Pierre Thomas $7 8.80 6.50 4.80 9.90 11.70 2.10 15.00 11.40 bye
Jerome Harrison $5 2.70 0.60 1.60 0.00 0.00 0.00 0.00 0.00 bye
Ben Tate $3 17.60 15.50 9.50 2.00 0.00 4.10 10.40 4.20 bye
Derrick Ward $3 9.90 0.00 0.00 0.00 0.00 0.00 0.70 0.00 bye
---------------------------------------------------------------------------------------------------------------------------------------------
Steve Johnson $19 16.60 23.60 23.40 9.80 6.90 14.90 0.00 11.70
Roy Williams $11 9.50 0.00 0.00 2.50 2.10 8.00 15.90 0.00
Plaxico Burress $10 17.20 0.00 14.50 6.30 7.20 2.60 24.50 0.00
Lee Evans $10 0.00 6.50 0.00 0.00 0.00 0.00 0.00 0.00
Nate Burleson $8 13.00 16.30 3.20 3.60 2.00 13.40 2.10 7.20 bye
Danny Amendola $8 9.50 0.00 0.00 0.00 0.00 0.00 0.00 0.00 out? out? out? out? out? out? out? out?
Jordy Nelson $7 19.70 15.40 7.00 20.10 4.70 18.40 9.20 0.00
Jacoby Jones $6 7.30 7.80 1.00 0.00 1.90 17.60 4.30 8.90 bye
Bernard Berrian $5 0.00 2.70 0.00 3.00 0.00 10.40 0.00 0.00 bye bye bye bye bye bye bye bye
Denarius Moore $3 0.00 28.10 15.70 10.90 0.00 1.60 1.40 0.00
Dexter McCluster $2 11.70 8.90 11.20 6.80 2.80 0.00 4.50 5.80
---------------------------------------------------------------------------------------------------------------------------------------------
Greg Olsen $9 13.80 4.90 22.20 18.50 12.60 11.70 1.90 19.30 bye
Todd Heap $8 7.00 2.30 15.10 10.10 0.00 0.00 0.00 0.00
Lance Kendricks $6 3.30 4.10 4.70 9.30 0.00 13.10 0.00 1.70
---------------------------------------------------------------------------------------------------------------------------------------------
Shaun Suisham $3 1.00 6.00 16.00 4.00 8.00 5.00 17.00 12.00 bye
Jay Feely $2 4.00 3.00 6.00 9.00 7.00 0.00 2.00 12.00
Phil Dawson $2 5.00 9.00 6.00 12.00 0.00 7.00 12.00 7.00
---------------------------------------------------------------------------------------------------------------------------------------------
San Diego Chargers $5 4.00 2.00 3.00 7.00 2.00 0.00 12.00 11.00
Cleveland Browns $3 4.00 6.00 9.00 2.00 0.00 4.00 7.00 1.00
St. Louis Rams $3 5.00 5.00 4.00 4.00 0.00 3.00 4.00 16.00
---------------------------------------------------------------------------------------------------------------------------------------------
TOTAL 162.55 173.70 185.05 155.40 108.40 124.35 141.50 140.05 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
CUTOFF 0.00 138.90 142.00 142.10 104.90 112.35 108.15 128.55 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Tennessee_ATO
Footballguy
And that's where we differ philosophically. I don't think your 9 $2-3 longshots can score as much as Jennings over a 3-week stretch, because by definition those are little more than random guesses which may or may not hit for you. Could you dumb luck your way into the proper combination? Absolutely. Like I've said repeatedly, I don't pretend to have some inside track on any of this. I just couldn't identify any combination of WRs with Jennings' (to continue to use him as the example) salary (or less) that I liked to consistently produce a scorer as good or better than Jennings.As for the other positions (Defense and PK), you may well be able to add a couple to each position who may well add to your total in the final 3 weeks. I don't consider that much of a "hole" though, since I think that's not much different than throwing $2 on a couple more numbers on the roulette wheel. We all have to make 1 or 2 bets on that (for both Def and PK), but throwing more money on additional numbers doesn't decrease the house edge.
Saint
Footballguy
Based on the conversations, figured I would add this based on the last few years winners (this is covered back on page 5). These are the average number of players per team in the Top 5 of the contest standings...
2008
[*]Average roster count for top 5: 22 players
[*]QB: 3
[*]RB: 6
[*]WR: 7
[*]TE: 2
[*]PK: 2
[*]TD: 2
2009
[*]Average roster count for top 5: 24 players (Note: Rounding errors for #s of players, primarily the extra is built into RB/WR/TE, they were all 0.6 of a player)
[*]QB: 2
[*]RB: 7
[*]WR: 8
[*]TE: 3
[*]PK: 3
[*]TD: 2
2010
[*]Average roster count for top 5: 26 players
[*]QB: 2
[*]RB: 7
[*]WR: 8
[*]TE: 3
[*]PK: 3
[*]TD: 3
Needless to say.... the average was larger rosters, not smaller.
2008
[*]Average roster count for top 5: 22 players
[*]QB: 3
[*]RB: 6
[*]WR: 7
[*]TE: 2
[*]PK: 2
[*]TD: 2
2009
[*]Average roster count for top 5: 24 players (Note: Rounding errors for #s of players, primarily the extra is built into RB/WR/TE, they were all 0.6 of a player)
[*]QB: 2
[*]RB: 7
[*]WR: 8
[*]TE: 3
[*]PK: 3
[*]TD: 2
2010
[*]Average roster count for top 5: 26 players
[*]QB: 2
[*]RB: 7
[*]WR: 8
[*]TE: 3
[*]PK: 3
[*]TD: 3
Needless to say.... the average was larger rosters, not smaller.
kardplayer
Footballguy
Let's look at Jennings (continuing him as the example) for weeks 5-7 (he was on a bye week 8):Week 5 4 for 82 (1 TD) = 18.2Week 6 6 for 82 = 14.2Week 7 7 for 147 (1 TD) = 27.7total over 3 weeks = 60.1In a non-PPR format (only leagues I can sort easily, but let's assume the rankings are close for now), he was 15th, 28th, and 2nd those weeks, and #4 overall for the total.Sub $10 players by week that beat/tied Jennings ($27)Week 5James Jones $9Jason Hill $7Steve Breaston $7Malcolm Floyd $8Darrius Heyward-Bey $3Week 6Devin Hester $7Jordy Nelson $7Jacoby Jones $6Jerome Simpson $8James Jones $9Nate Burleson $8Danario Alexander $6Mohamed MassaquoiJason Hill $7Donald Driver $8Torrey Smith $8Darrius Heyward-Bey $3Week 7NoneSo, 2 out of 3 weeks, there were lots of opportunities to "get lucky" and have a player that beat Jennings, at 1/3 the price or less. 13->1 may not have worked, but 3->1 seems like a virtual lock. You could have a lucky combo of the players, or you could just have James Jones, Jason Hill, or Darrius Heyward-Bey. A lock, no. But definitely not crazy to think he's beatable by a squad of best ball WR's.
Tennessee_ATO
Footballguy
But that's not the query that matters for the finals. Remember, I've readily acknowledged that at least through the byes the short rosters will be eliminated at a higher rate than long rosters. The query that matters for the finals would be: Is there some combination of players whose combined salaries are =/< $27 from which the sum of the top weekly score each week would be > Jennings' score? So, in your example, perhaps James Jones and some player who scored close to Jennings in week 7 would have combined to tally more points than Jennings for that 3 week stretch. There probably was. The next part of the query is, and this is the important part: how many teams had that combination in the finals? Right now, Jones is owned by 121 owners. Hill and Heyward-Bey (the other players who beat Jennings in the 1st 2 weeks) are owned by 32 and 138 respectively. Say there are 15 unique cominbations of players who would fulfill the 1st part of the query. It doesn't matter much if none of the 250 finalists have one of those combinations.As an aside, I think your example doesn't much help your argument someone for the survivor portion of the contest. You and I have a different definition of "lots of opportunities". 5 isn't "lots" in my book. I would say, "lost of opportunities 1 out of 3 weeks, very limited opportunities in 1 out of 3 weeks, and no opportunities in 1 out of 3 weeks". I understand your point and considered it when I was trying to figure out what to do.
CalBear
Footballguy
OK, so here's one way to look at it. I took the top 4 most-owned $3 and $2 WRs (so, 8 total), and looked at their scores so far this year. For $20, here's what you get:
Code:
Greg Jennings $27 14.9 11.5 11.9 16.3 14.2 8.2 20.7 0 Scrubs $20 9.5 23.1 11.7 11.5 15.9 8.2 17.1 12.7
Last edited by a moderator:
Touchdown There
Footballguy
Or try using Jamaal Charles as your stud. Large rosters have lower risk and volatility.
Tennessee_ATO
Footballguy
Velveeta22,
No doubt that larger rosters are less volatile during the survivor portion of the contest. In the finals there will be a handful of players who score big. Either you have them or you don't. I believe that a higher percentage of those "big scorers" will be $20+ players than <$10 players.
CalBear,
What percentage of teams have that combination of 8 WRs though? Broaden the universe to only those WRs who produced a scoring week out of that set of 8. I'll bet it's darn close to "0%" in either scenario.
At the end of the day, it's about maximizing your own chance to win the title. The argument some have espoused has been: "Someone with a big roster will have rostered the right combination of guys, therefore that makes it the right choice for me to maximize my chances." That's a variation on the gambler's fallacy. Maybe you guys have more confidence in your skills at predicting success of $2 and $3 players than I do. From my perspective, taking a bunch of cheap guys was little different than picking numbers on the roulette wheel and hoping for the best.
No doubt that larger rosters are less volatile during the survivor portion of the contest. In the finals there will be a handful of players who score big. Either you have them or you don't. I believe that a higher percentage of those "big scorers" will be $20+ players than <$10 players.
CalBear,
What percentage of teams have that combination of 8 WRs though? Broaden the universe to only those WRs who produced a scoring week out of that set of 8. I'll bet it's darn close to "0%" in either scenario.
At the end of the day, it's about maximizing your own chance to win the title. The argument some have espoused has been: "Someone with a big roster will have rostered the right combination of guys, therefore that makes it the right choice for me to maximize my chances." That's a variation on the gambler's fallacy. Maybe you guys have more confidence in your skills at predicting success of $2 and $3 players than I do. From my perspective, taking a bunch of cheap guys was little different than picking numbers on the roulette wheel and hoping for the best.
CalBear
Footballguy
The point is, it doesn't matter which eight scrubs you choose; as a group, eight reasonable scrubs will almost certainly outperform a single player in this contest. Three of those eight have contributed nothing (Crayton, Parrish, Cobb).
It's got to be better than taking a known sub-optimal strategy and hoping you're successful with it. That's little different than hitting 17 and hoping you draw a 4.
Ignoratio Elenchi
Footballguy
Possibly. But you can also cram more <$10 players onto your roster, somewhat mitigating this. Every year, some of those <$10 players turn into studs. In August, no one knows which one(s) it will be, but whoever they are will be found on some large rosters in the finals. Every roster that makes it to the finals, small or large, has good players on it (judged by their actual performance, and not by their preseason valuation).
I don't see how that's a variation on the gambler's fallacy; gambler's fallacy would be something like, "My players have sucked all year, therefore they're due to do well during the finals." Anyway, I think we agree that larger rosters give you a better chance of surviving the first 13 weeks. The theory is that, by contrast, small rosters give you a better chance of the highest cumulative score in weeks 14-16. I'm not convinced that's the case, but even if it is, the small roster strategy amounts to taking a significant, known disadvantage over the course of the first 13 weeks for a perceived advantage in the final 3 weeks.I've said before, to me, the final 250 is a lottery. When you're building a roster in August, there is absolutely no way to predict with any meaningful accuracy which players will score the most points four months down the road. The best way to win a lottery is not to try to predict which numbers will come up, it's to buy more tickets.
BassNBrew
Footballguy
Greg Jennings $27 14.9 11.5 11.9 16.3 14.2 8.2 20.7 0
James Jones $9 1.10 2.50 6.40 13.80 25.00 10.50 10.30 0.00
Jordy Nelson $7 19.70 15.40 7.00 20.10 4.70 18.40 9.2 0
Here's an example that I was debating while building my roster. Jennings v. Jones/Nelson
wk 1: J/N +4.8
wk 2: J/N +3.9
wk 3: Jennings +4.9
wk 4: J/N +3.8
wk 5: J/N +11.2
wk 6: J/N +10.2
wk 7: Jennings +10.4
Net 18.6 in favor of Jones/Nelson plus $11 to spend on add'l points at other positions.
There are 27 teams remaining with Jones/Nelson. Not only are you going to have to have Jennings outperform Jones/Nelson over the last three weeks, he have to beat them by enough to cover the $11 in upgrades they have elsewhere.
6 of those guys have Hernadez. 3 have Olsen. 13 have Antonio Brown. 7 have Burress. 4 have Fred Davis. 5 have McGahee. 5 have Pettigrew. 2 have Sproles. 3 have Sanders.
You'll need a ton of luck plus have exhibit a huge amount of skill with your other selections to dig out of this hole
James Jones $9 1.10 2.50 6.40 13.80 25.00 10.50 10.30 0.00
Jordy Nelson $7 19.70 15.40 7.00 20.10 4.70 18.40 9.2 0
Here's an example that I was debating while building my roster. Jennings v. Jones/Nelson
wk 1: J/N +4.8
wk 2: J/N +3.9
wk 3: Jennings +4.9
wk 4: J/N +3.8
wk 5: J/N +11.2
wk 6: J/N +10.2
wk 7: Jennings +10.4
Net 18.6 in favor of Jones/Nelson plus $11 to spend on add'l points at other positions.
There are 27 teams remaining with Jones/Nelson. Not only are you going to have to have Jennings outperform Jones/Nelson over the last three weeks, he have to beat them by enough to cover the $11 in upgrades they have elsewhere.
6 of those guys have Hernadez. 3 have Olsen. 13 have Antonio Brown. 7 have Burress. 4 have Fred Davis. 5 have McGahee. 5 have Pettigrew. 2 have Sproles. 3 have Sanders.
You'll need a ton of luck plus have exhibit a huge amount of skill with your other selections to dig out of this hole
Tennessee_ATO
Footballguy
I personally felt I could manage the byes with a short roster through planning. I think I have done that pretty well. That's certainly not to say that my team couldn't go "full #######" in a given week, but I have 1 player on bye this week, none next week, and 3 in week 11. Week 11 is my most nerve-wracking from a bye standpoint, but one of those 3 is Tate (unfortunately I've had to rely on him more frequently than I hoped since I'm part of the groupthink with Felix, Hightower, Beanie, and Tate, but that's another issue entirely). Things I could control, such as the number/type/position of players on bye from week to week, I tried to control. That's why I've felt all along that my roster of 18 stood a far better chance than the "average" 18-man roster of surviving the byes.Maybe blindly (for me anyway) buying more tickets is the best thing to do, I don't know. I tried to buy the tickets I thought had the best chance to succeed, and controled as many things as I could.'Ignoratio Elenchi said:Anyway, I think we agree that larger rosters give you a better chance of surviving the first 13 weeks. The theory is that, by contrast, small rosters give you a better chance of the highest cumulative score in weeks 14-16. I'm not convinced that's the case, but even if it is, the small roster strategy amounts to taking a significant, known disadvantage over the course of the first 13 weeks for a perceived advantage in the final 3 weeks.I've said before, to me, the final 250 is a lottery. When you're building a roster in August, there is absolutely no way to predict with any meaningful accuracy which players will score the most points four months down the road. The best way to win a lottery is not to try to predict which numbers will come up, it's to buy more tickets.
I engaged in PRECISELY that same debate with Jones/Nelson/Driver (I thought Driver had to be included in the analysis since he's consistently been pretty involved in the offense, and it was ultimately Driver's potential late-season involvement that scared me away from going the Jones+Nelson route) in trying to decide what to do about Jennings. I reached a different conclusion, but the exercise was the same and it was a close call on my part. As an aside, I actually encouraged my brother to draft Jones and Nelson in a PPC league we're in. I snagged Jennings in the 2nd round. He was whining about not knowing which would hit in a given week. I told him to start them both and they would collective amass more than most every other WR2+WR3 combo anyone else had. Worked like a charm for him so far.I think that's different than simply saying a big roster is better. This represents the type of analysis that I felt comfortable with, and did with every player on my team. I was not, and am not, married to "18 man roster or bust". The Jones/Nelson excercise is a perfect example, if I liked that combo more than Jennings, I would have chosen them without a problem. It's the $2 and $3 guys you HAVE to have to round-out a 28-30 man squad that seemed like nothing more than a random shot for me. Like I said, maybe I'm just crappy at projecting cheap guys.Finally, Jennings isn't the only WR I have on my team. Obviously if Jennings is strong the rest of the way, that's good for me. But I also have some other guys who could score for me. It's not like Jennings HAS to count for me each week. He's been my top-scoring WR 3 times and my 4th-scoring WR twice. According to the query, no one has my WR combination -- I don't think anyone has any 3 of my WRs rostered. Not sure if that's good or bad on my part, but that's the way it is.'BassNBrew said:
I am still alive with the ever-popular 28-man roster. But I think my roster represents everything that is good and bad about a large roster. I have depth at every position 3QBs, 6RBs, 9WRs, 3TEs, 4ks, 3Ds That has allowed me to absorb injuries and byes fairly easily. I also have a lot of good value picks.
But, I don't have the consistency that you would expect from higher-priced players. So, a guy like Denarious Moore had a good few weeks, but nothing the other weeks. Even Antonio Brown has been hit/miss. I have used 6 WRs at least twice, and a 7th once. (Only blanks are from Bennett and Evans).
So, I have the depth to make it this far, but I don't know that I will have the consistency to score well down the stretch. Its hard to predict when some of these guys will have a good game, and it is rare that they have a "great" game. My best hope is for Gore to continue to play well (I did handcuff him with Hunter, so it would not be a total loss), and for guys like Bennett and Evans to come back from injury and have a random good game. I also would not mind seeing Bradford step up his game down the stretch.
But, I don't have the consistency that you would expect from higher-priced players. So, a guy like Denarious Moore had a good few weeks, but nothing the other weeks. Even Antonio Brown has been hit/miss. I have used 6 WRs at least twice, and a 7th once. (Only blanks are from Bennett and Evans).
So, I have the depth to make it this far, but I don't know that I will have the consistency to score well down the stretch. Its hard to predict when some of these guys will have a good game, and it is rare that they have a "great" game. My best hope is for Gore to continue to play well (I did handcuff him with Hunter, so it would not be a total loss), and for guys like Bennett and Evans to come back from injury and have a random good game. I also would not mind seeing Bradford step up his game down the stretch.
The Real Hipster Doofus
Footballguy
Definitely going to be out this week. From my original roster of 25, I am short this week....
Stafford (bye)
Kolb (injury) my only two QB's
Felix Jones (injury)
Mark Ingram (injury)
Tim Hightower (injury)
Deji Karim (bye)
Calvin Johnson (bye)
Nate Burleson (bye)
Danny Amendola (injury)
Brandon Pettigrew (bye)
So yea, I can definitely see why some people prefer larger rosters.
Stafford (bye)
Kolb (injury) my only two QB's
Felix Jones (injury)
Mark Ingram (injury)
Tim Hightower (injury)
Deji Karim (bye)
Calvin Johnson (bye)
Nate Burleson (bye)
Danny Amendola (injury)
Brandon Pettigrew (bye)
So yea, I can definitely see why some people prefer larger rosters.
