What's new
Fantasy Football - Footballguys Forums

Welcome to Our Forums. Once you've registered and logged in, you're primed to talk football, among other topics, with the sharpest and most experienced fantasy players on the internet.

***Official*** 2012 FBG Subscriber Contest Thread (2 Viewers)

i don't know why i am getting so much debate here.
That's what this thread is all about. Don't get all sensitive about it. We've been doing this for years)
:goodposting: I frequently play devils advocate to people even if I agree with them. Its more thought provoking and way more useful than just agreeing with everyone. For years ive took up defense of the smaller rosters even though all my teams were in the 25 player range. Besides this contest has so many nuances and is changing every year that I highly doubt the ideal strategy, if there even is one, is simple. We forge new ideas by challenging old ones.
feelings still hurt. flowers would help
 
Teaser: I made a tweak to my team's lineup today that will probably make or break my contest. It's a risk/reward positional bye week strategy that hasn't been discussed on the board yet. In it to win it. Tune in next week for details...
Rolling with Stafford and no backup? That's what I've been thinking of doing.
If your backup choice is between Tannehill (4) and no one, I would pick no one. You would only use Tannehill on Stafford's week 5 bye week. With a well constructed team, throwing up a zero in the qb position in week five shouldn't make or break you. Is it risky....yes....but I would say more for weeks 6-13. If Stafford gets hurt, you are likely done...this doesn't change if Tannehill is on your team.A better backup would be someone like Flacco, Luck, or Fitzy. They aren't supper expensive, but have a better chance to contribute on a week to week basis. And if Stafford goes down, they might be able to keep the ship afloat until Stafford gets back. But in selecting these guys, you are also giving up in other areas. If Stafford doesn't get hurt, you will probably be wasting the 11 dollars for most of the season.Then you could choose to back him up with a higher priced qb like Ryan, Cutler, or Manning. Those guys can put up big numbers on any given week. My argument against this is that I don't feel like they are going to make a big enough difference in total points scored from Stafford to justify spending this much money at the QB position. There are a lot of RB, WR, and TE you are giving up if you choose a backup in this group. I'm looking to spend between $35-$40 at the QB position this year. Now do I pick a top five guy and have around 10 dollars for my back up or do I select a couple players from the $18-$20 range? Going with only one stud is very risky, but something I have looked at as well.
 
Last edited by a moderator:
OK...after several hundred iterations, I think I'm settled in w/this (subject to change several dozen times :unsure: )

QB = 2 $$ 30.00

RB = 7 $$ 64.00

WR = 8 $$ 93.00

TE = 3 $$ 48.00

PK = 3 $$ 9.00

D = 2 $$ 6.00

25 $$ 250.00

With the change to .5 ppr for RB, it made more sense to carry a couple of extra bodies here. Plus, the RB category was my downfall in 2010/2011 (bounced week 8 in each) after having made the final 250 in 2008 and 2009.

I used to be a strong advocate of the 20 - 22 player optimal roster size but am trying a bit more expansion on that to overcome the injury bug that has struck my entry hard the last couple times out.

The only hard rule I employ is a minimum of 8 WR. Having to start 3 every week, + the flex option, you cannot succeed w/out this threshold IMO. I usually also like to carry at least 4 PK to play the lottery angle, but I just couldn't make it work without sacrificing something that held more value to me.

I really would have preferred to carry 3 QB but it would have been too damaging to the RB/WR/TE core.

 
Last edited by a moderator:
Am I the only one going with 3 QBs? Is that a sucker's play now? :unsure:
Other than Vick, Did any other big name QB's miss time last year? It may be lulling people into a false sense of security. If I'm taking Vick, I would look at going with three. If not, I'm probably only taking two.
 
Am I the only one going with 3 QBs? Is that a sucker's play now? :unsure:
Pretty much.1. There will basically be teams that have you QBs covered and applied that money elsewhere.2. If you have a stud, the other two won't score for you enough to matter. 3. If your going with three cheap guys you still won't get Rodgers like production.4. In it to win it.
 
Agreed. I think your analysis works well in retrospect. Knowing (or assuming) you know the relative amount of points scored and distribution of those points over a season, you can come up with what WOULD have been the best roster make up for that specific season. The problem I have with it is that we don't know the amount of points scored or their distributions. Sure we have best guess estimates, but so much of the results of your analysis comes from the distrubtion of the points over the year. And shifting just a few points around could result in much different results.
+100 There is a difference between ex-post probability and ex-ante probability.

 
Agreed. I think your analysis works well in retrospect. Knowing (or assuming) you know the relative amount of points scored and distribution of those points over a season, you can come up with what WOULD have been the best roster make up for that specific season. The problem I have with it is that we don't know the amount of points scored or their distributions. Sure we have best guess estimates, but so much of the results of your analysis comes from the distrubtion of the points over the year. And shifting just a few points around could result in much different results.
+100 There is a difference between ex-post probability and ex-ante probability.
And shifting just a few points around could result in much different results.so what would you say i do? vary iterations for each player 10, 20, 30 times?

...thats what i did. all of that has been accounted for

 
Last edited by a moderator:
Agreed. I think your analysis works well in retrospect. Knowing (or assuming) you know the relative amount of points scored and distribution of those points over a season, you can come up with what WOULD have been the best roster make up for that specific season. The problem I have with it is that we don't know the amount of points scored or their distributions. Sure we have best guess estimates, but so much of the results of your analysis comes from the distrubtion of the points over the year. And shifting just a few points around could result in much different results.
+100 There is a difference between ex-post probability and ex-ante probability.
And shifting just a few points around could result in much different results.so what would you say i do? vary iterations for each player 10, 20, 30 times?

...thats what i did. all of that has been accounted for
You can estimate it, but who knows how the season will break. If your projections come true, then you know who you should grab. Projections are not 100% accurate, so it is hard to make conclusive statements like Stud + Scrub is definitely better than 2 middle level players. Precisely because the middle level players could break out or the Stud could get hurt or under-perform. The bottom line is that you could have a winning team, or a team that goes out in week 5. Same is true of my team and probably everyone posting in this thread. We just don't know what the winning strategy is yet since we don't know if players will live up to projections + what injuries will occur.

 
Am I the only one going with 3 QBs? Is that a sucker's play now? :unsure:
Pretty much.1. There will basically be teams that have you QBs covered and applied that money elsewhere.2. If you have a stud, the other two won't score for you enough to matter. 3. If your going with three cheap guys you still won't get Rodgers like production.4. In it to win it.
Good points. Have one top tier guy and one middle tier guy. Added a flyer cheap. I guess I'd rather not lose if QB1 goes down for the week of QB2's Bye. Maybe the flyer gets me 5-8 points and that's enough. Should at least rethink it though.
 
Agreed. I think your analysis works well in retrospect. Knowing (or assuming) you know the relative amount of points scored and distribution of those points over a season, you can come up with what WOULD have been the best roster make up for that specific season. The problem I have with it is that we don't know the amount of points scored or their distributions. Sure we have best guess estimates, but so much of the results of your analysis comes from the distrubtion of the points over the year. And shifting just a few points around could result in much different results.
+100 There is a difference between ex-post probability and ex-ante probability.
And shifting just a few points around could result in much different results.so what would you say i do? vary iterations for each player 10, 20, 30 times?

...thats what i did. all of that has been accounted for
You can estimate it, but who knows how the season will break. If your projections come true, then you know who you should grab. Projections are not 100% accurate, so it is hard to make conclusive statements like Stud + Scrub is definitely better than 2 middle level players. Precisely because the middle level players could break out or the Stud could get hurt or under-perform. The bottom line is that you could have a winning team, or a team that goes out in week 5. Same is true of my team and probably everyone posting in this thread. We just don't know what the winning strategy is yet since we don't know if players will live up to projections + what injuries will occur.
there are too many factors to solve this tournament in an equation? is that your point?
 
Agreed. I think your analysis works well in retrospect. Knowing (or assuming) you know the relative amount of points scored and distribution of those points over a season, you can come up with what WOULD have been the best roster make up for that specific season. The problem I have with it is that we don't know the amount of points scored or their distributions. Sure we have best guess estimates, but so much of the results of your analysis comes from the distrubtion of the points over the year. And shifting just a few points around could result in much different results.
+100 There is a difference between ex-post probability and ex-ante probability.
And shifting just a few points around could result in much different results.so what would you say i do? vary iterations for each player 10, 20, 30 times?

...thats what i did. all of that has been accounted for
Ok, and what is your conclusion? That on average a stud+cheap guy is better? I'm still waiting for what your definition of better is. But regardless, it makes no difference what's better on average. By selecting a Stud+cheap guy or a couple of middle QB's, I get a specific pair. So all i care about it selecting the best pair. So for one person specific projections, it COULD make sense to go with a couple middle cost QB's over a "Stud" and Cheap player. It looks like for YOUR specific projections, it's always (or almost) better to go with Stud + Cheap.
 
Agreed. I think your analysis works well in retrospect. Knowing (or assuming) you know the relative amount of points scored and distribution of those points over a season, you can come up with what WOULD have been the best roster make up for that specific season. The problem I have with it is that we don't know the amount of points scored or their distributions. Sure we have best guess estimates, but so much of the results of your analysis comes from the distrubtion of the points over the year. And shifting just a few points around could result in much different results.
+100 There is a difference between ex-post probability and ex-ante probability.
And shifting just a few points around could result in much different results.so what would you say i do? vary iterations for each player 10, 20, 30 times?

...thats what i did. all of that has been accounted for
You can estimate it, but who knows how the season will break. If your projections come true, then you know who you should grab. Projections are not 100% accurate, so it is hard to make conclusive statements like Stud + Scrub is definitely better than 2 middle level players. Precisely because the middle level players could break out or the Stud could get hurt or under-perform. The bottom line is that you could have a winning team, or a team that goes out in week 5. Same is true of my team and probably everyone posting in this thread. We just don't know what the winning strategy is yet since we don't know if players will live up to projections + what injuries will occur.
there are too many factors to solve this tournament in an equation? is that your point?
I think the point is that there are many ways to get to the winners circle, not just a single right way. For the first years of this contest people thought there was no way several small "stud" line up would ever do well. They said, yeah, maybe one or two will finish high but it'll mostly be the larger rosters that dominate. Then last year happened.

 
Agreed. I think your analysis works well in retrospect. Knowing (or assuming) you know the relative amount of points scored and distribution of those points over a season, you can come up with what WOULD have been the best roster make up for that specific season. The problem I have with it is that we don't know the amount of points scored or their distributions. Sure we have best guess estimates, but so much of the results of your analysis comes from the distrubtion of the points over the year. And shifting just a few points around could result in much different results.
+100 There is a difference between ex-post probability and ex-ante probability.
That's what she said...
 
Agreed. I think your analysis works well in retrospect. Knowing (or assuming) you know the relative amount of points scored and distribution of those points over a season, you can come up with what WOULD have been the best roster make up for that specific season. The problem I have with it is that we don't know the amount of points scored or their distributions. Sure we have best guess estimates, but so much of the results of your analysis comes from the distrubtion of the points over the year. And shifting just a few points around could result in much different results.
+100 There is a difference between ex-post probability and ex-ante probability.
And shifting just a few points around could result in much different results.so what would you say i do? vary iterations for each player 10, 20, 30 times?

...thats what i did. all of that has been accounted for
Ok, and what is your conclusion? That on average a stud+cheap guy is better?
YES :wall:
 
Last edited by a moderator:
Agreed. I think your analysis works well in retrospect. Knowing (or assuming) you know the relative amount of points scored and distribution of those points over a season, you can come up with what WOULD have been the best roster make up for that specific season. The problem I have with it is that we don't know the amount of points scored or their distributions. Sure we have best guess estimates, but so much of the results of your analysis comes from the distrubtion of the points over the year. And shifting just a few points around could result in much different results.
+100 There is a difference between ex-post probability and ex-ante probability.
And shifting just a few points around could result in much different results.so what would you say i do? vary iterations for each player 10, 20, 30 times?

...thats what i did. all of that has been accounted for
You can estimate it, but who knows how the season will break. If your projections come true, then you know who you should grab. Projections are not 100% accurate, so it is hard to make conclusive statements like Stud + Scrub is definitely better than 2 middle level players. Precisely because the middle level players could break out or the Stud could get hurt or under-perform. The bottom line is that you could have a winning team, or a team that goes out in week 5. Same is true of my team and probably everyone posting in this thread. We just don't know what the winning strategy is yet since we don't know if players will live up to projections + what injuries will occur.
there are too many factors to solve this tournament in an equation? is that your point?
I think the point is that there are many ways to get to the winners circle, not just a single right way. For the first years of this contest people thought there was no way several small "stud" line up would ever do well. They said, yeah, maybe one or two will finish high but it'll mostly be the larger rosters that dominate. Then last year happened.
Keep in mind that small rosters were 10x of the population at the start.
 
Agreed. I think your analysis works well in retrospect. Knowing (or assuming) you know the relative amount of points scored and distribution of those points over a season, you can come up with what WOULD have been the best roster make up for that specific season. The problem I have with it is that we don't know the amount of points scored or their distributions. Sure we have best guess estimates, but so much of the results of your analysis comes from the distrubtion of the points over the year. And shifting just a few points around could result in much different results.
+100 There is a difference between ex-post probability and ex-ante probability.
And shifting just a few points around could result in much different results.so what would you say i do? vary iterations for each player 10, 20, 30 times?

...thats what i did. all of that has been accounted for
Ok, and what is your conclusion? That on average a stud+cheap guy is better?
YES :wall:
Ok now address the rest of my post please.

 
Agreed. I think your analysis works well in retrospect. Knowing (or assuming) you know the relative amount of points scored and distribution of those points over a season, you can come up with what WOULD have been the best roster make up for that specific season. The problem I have with it is that we don't know the amount of points scored or their distributions. Sure we have best guess estimates, but so much of the results of your analysis comes from the distrubtion of the points over the year. And shifting just a few points around could result in much different results.
+100 There is a difference between ex-post probability and ex-ante probability.
And shifting just a few points around could result in much different results.so what would you say i do? vary iterations for each player 10, 20, 30 times?

...thats what i did. all of that has been accounted for
You can estimate it, but who knows how the season will break. If your projections come true, then you know who you should grab. Projections are not 100% accurate, so it is hard to make conclusive statements like Stud + Scrub is definitely better than 2 middle level players. Precisely because the middle level players could break out or the Stud could get hurt or under-perform. The bottom line is that you could have a winning team, or a team that goes out in week 5. Same is true of my team and probably everyone posting in this thread. We just don't know what the winning strategy is yet since we don't know if players will live up to projections + what injuries will occur.
there are too many factors to solve this tournament in an equation? is that your point?
I think the point is that there are many ways to get to the winners circle, not just a single right way. For the first years of this contest people thought there was no way several small "stud" line up would ever do well. They said, yeah, maybe one or two will finish high but it'll mostly be the larger rosters that dominate. Then last year happened.
Keep in mind that small rosters were 10x of the population at the start.
10x might be a bit of an exaggeration but point taken. However how is that different than any other year? My point remains the same, each year is different and what has worked last year or the year before doesn't mean it's the ideal method or only way to win. A lot of different things can happen over the season, and depending on what and when, that'll shape the best strategy for this year.
 
Benson will probably have the highest % of ownership of any player in this contest. He is the cheapest RB who will start the year as the starter.
Beware of one Alex Green.
injury prone and not a full time back. 6 is a bit high for him, imo
Which $6 full time back do you prefer?
none that i like
Yeah, he probably ain't worth it.
 
I'm still waiting for what your definition of better is.
i define better as more ppg, higher floor, higher ceiling week by week and as season totals
But regardless, it makes no difference what's better on average.
yes it does. taking a top 3 qb based on YOUR rankings and taking the best cheap qb based on YOUR rankings is better on average than taking two 5-10 ranked qb's based on YOUR rankings who of which will likely cost in the $15-$25 range.if you think matt ryan at $19 is going to be the 4th qb, then i would say to take him. if you think rodgers is the best qb, ryan is the 7th best, romo is the 9th best and tannahill is the 27th best, you are better off taking rodgers and tannahill over ryan+romorodgers+tannehill will score more and is a lot cheaper.
By selecting a Stud+cheap guy or a couple of middle QB's, I get a specific pair. So all i care about it selecting the best pair.
im telling you how to increase your odds of selecting the best possible pair
So for one person specific projections, it COULD make sense to go with a couple middle cost QB's over a "Stud" and Cheap player.It looks like for YOUR specific projections, it's always (or almost) better to go with Stud + Cheap.
you clearly didn't read my earlier posts.i am not saying to take the $30 guy just because he is $30. i am saying take YOUR top guys at each position, even if that means rodgers, foster, calvin and graham.and use some common sense. do i really have to say that? if rodgers at $31 is your #1 qb and ryan at $19 is your #3 qb, then, yes. you should take ryan over rodgers.if rodgers is your #1 and ryan is your #8, then no. despite the way it may seem, you are better off taking rodgers and cheap guy over your qb 8 and qb 9
 
Last edited by a moderator:
Week 7 is friggin' killing me!

I can't decide who to move out, but I can't leave it like this or there is no way I survive with 1/3 of my team on the same bye.

 
Week 7 is friggin' killing me!I can't decide who to move out, but I can't leave it like this or there is no way I survive with 1/3 of my team on the same bye.
week 7 is the easiest week after the first 4. the most value is on bye that week imobut, 1/3 of your lineup out is probably a bit much
 
Last edited by a moderator:
I'm still waiting for what your definition of better is.
i define better as more ppg, higher floor, higher ceiling week by week and as season totals
But regardless, it makes no difference what's better on average.
yes it does. taking a top 3 qb based on YOUR rankings and taking the best cheap qb based on YOUR rankings is better on average than taking two 5-10 ranked qb's based on YOUR rankings who of which will likely cost in the $15-$25 range.if you think matt ryan at $19 is going to be the 4th qb, then i would say to take him. if you think rodgers is the best qb, ryan is the 7th best, romo is the 9th best and tannahill is the 27th best, you are better off taking rodgers and tannahill over ryan+romorodgers+tannehill will score more and is a lot cheaper.
By selecting a Stud+cheap guy or a couple of middle QB's, I get a specific pair. So all i care about it selecting the best pair.
im telling you how to increase your odds of selecting the best possible pair
So for one person specific projections, it COULD make sense to go with a couple middle cost QB's over a "Stud" and Cheap player.It looks like for YOUR specific projections, it's always (or almost) better to go with Stud + Cheap.
you clearly didn't read my earlier posts.i am not saying to take the $30 guy just because he is $30. i am saying take YOUR top guys at each position, even if that means rodgers, foster, calvin and graham.and use some common sense. do i really have to say that? if rodgers at $31 is your #1 qb and ryan at $19 is your #3 qb, then, yes. you should take ryan over rodgers.if rodgers is your #1 and ryan is your #8, then no. despite the way it may seem, you are better off taking rodgers and cheap guy over your qb 8 and qb 9
This is exactly what I've been saying, so I'm not sure why you ever had a problem with it. I never debated the fact that the #1 and #27 QB would out produce the #8 and #9 Qbs. The problem is we dont know the end of year ranks yet, so to say not taking a "stud" qb is the wrong strategy is faulty because we don't know who the stud Qbs are yet. I don't think any one is taking the 9th highest priced qb expecting them to end up QB9 at the end of the year. And if they are then I agree that's the wrong approach.
 
'Modog814 said:
The problem is we dont know the end of year ranks yet, so to say not taking a "stud" qb is the wrong strategy is faulty because we don't know who the stud Qbs are yet.
:goodposting:This was my point earlier - it hardly matters what the results of flc's analysis are, because it's flawed from the outset. He's plugging in points he assumes players will score. "This WR costs $20, I'm going to give him 17 ppg. This other WR costs $5, I'm going to give him 5 ppg." Of course that's going to show that taking "studs" is the better approach, because the analysis assumes that scoring correlates directly to price. But in reality, it doesn't. That's why three $10 WRs can outproduce one $30 WR - not simply because of variance or "quantity over quality," but also because there is some nonzero chance that the $30 WR will underperform his price and/or one or more of the $10 WRs will vastly outperform their price.
 
'Modog814 said:
'flc735 said:
I'm still waiting for what your definition of better is.
i define better as more ppg, higher floor, higher ceiling week by week and as season totals
But regardless, it makes no difference what's better on average.
yes it does. taking a top 3 qb based on YOUR rankings and taking the best cheap qb based on YOUR rankings is better on average than taking two 5-10 ranked qb's based on YOUR rankings who of which will likely cost in the $15-$25 range.if you think matt ryan at $19 is going to be the 4th qb, then i would say to take him. if you think rodgers is the best qb, ryan is the 7th best, romo is the 9th best and tannahill is the 27th best, you are better off taking rodgers and tannahill over ryan+romorodgers+tannehill will score more and is a lot cheaper.
By selecting a Stud+cheap guy or a couple of middle QB's, I get a specific pair. So all i care about it selecting the best pair.
im telling you how to increase your odds of selecting the best possible pair
So for one person specific projections, it COULD make sense to go with a couple middle cost QB's over a "Stud" and Cheap player.It looks like for YOUR specific projections, it's always (or almost) better to go with Stud + Cheap.
you clearly didn't read my earlier posts.i am not saying to take the $30 guy just because he is $30. i am saying take YOUR top guys at each position, even if that means rodgers, foster, calvin and graham.and use some common sense. do i really have to say that? if rodgers at $31 is your #1 qb and ryan at $19 is your #3 qb, then, yes. you should take ryan over rodgers.if rodgers is your #1 and ryan is your #8, then no. despite the way it may seem, you are better off taking rodgers and cheap guy over your qb 8 and qb 9
This is exactly what I've been saying, so I'm not sure why you ever had a problem with it. I never debated the fact that the #1 and #27 QB would out produce the #8 and #9 Qbs. The problem is we dont know the end of year ranks yet, so to say not taking a "stud" qb is the wrong strategy is faulty because we don't know who the stud Qbs are yet. I don't think any one is taking the 9th highest priced qb expecting them to end up QB9 at the end of the year. And if they are then I agree that's the wrong approach.
'Modog814 said:
The problem is we dont know the end of year ranks yet, so to say not taking a "stud" qb is the wrong strategy is faulty because we don't know who the stud Qbs are yet.
:goodposting:This was my point earlier - it hardly matters what the results of flc's analysis are, because it's flawed from the outset. He's plugging in points he assumes players will score. "This WR costs $20, I'm going to give him 17 ppg. This other WR costs $5, I'm going to give him 5 ppg." Of course that's going to show that taking "studs" is the better approach, because the analysis assumes that scoring correlates directly to price. But in reality, it doesn't. That's why three $10 WRs can outproduce one $30 WR - not simply because of variance or "quantity over quality," but also because there is some nonzero chance that the $30 WR will underperform his price and/or one or more of the $10 WRs will vastly outperform their price.
so whats the point of building a team? if you cant predict the outcomes, then we might as well pick names out of a hat.ignoratio, i've said time and time again to use your own projections. price doesn't have anything to with your projections.
That's why three $10 WRs can outproduce one $30 WR
it's possible, but the odds are overwhelmingly against that. i showed you this. just because it is possible doesn't mean it's equally as likely as the other option.
 
'Modog814 said:
The problem is we dont know the end of year ranks yet, so to say not taking a "stud" qb is the wrong strategy is faulty because we don't know who the stud Qbs are yet.
:goodposting:This was my point earlier - it hardly matters what the results of flc's analysis are, because it's flawed from the outset. He's plugging in points he assumes players will score. "This WR costs $20, I'm going to give him 17 ppg. This other WR costs $5, I'm going to give him 5 ppg." Of course that's going to show that taking "studs" is the better approach, because the analysis assumes that scoring correlates directly to price. But in reality, it doesn't. That's why three $10 WRs can outproduce one $30 WR - not simply because of variance or "quantity over quality," but also because there is some nonzero chance that the $30 WR will underperform his price and/or one or more of the $10 WRs will vastly outperform their price.
I don't think he is basing it off of price. He just projected Julio Jones out at 17 ppg. Well, Julio's numbers from last year don't represent that....so he picked another player who averaged 17 ppg to get variance in the week to week scores. DD more or less gives you a players average with a point or two more or less. That to me isn't a good indicator of what they really are going to score on a week to week basis.If flc projects a $5 guy is going to average 17 ppg, then he could compare it with the same person he chose for Julio Jones.
 
'Modog814 said:
The problem is we dont know the end of year ranks yet, so to say not taking a "stud" qb is the wrong strategy is faulty because we don't know who the stud Qbs are yet.
:goodposting:This was my point earlier - it hardly matters what the results of flc's analysis are, because it's flawed from the outset. He's plugging in points he assumes players will score. "This WR costs $20, I'm going to give him 17 ppg. This other WR costs $5, I'm going to give him 5 ppg." Of course that's going to show that taking "studs" is the better approach, because the analysis assumes that scoring correlates directly to price. But in reality, it doesn't. That's why three $10 WRs can outproduce one $30 WR - not simply because of variance or "quantity over quality," but also because there is some nonzero chance that the $30 WR will underperform his price and/or one or more of the $10 WRs will vastly outperform their price.
I don't think he is basing it off of price. He just projected Julio Jones out at 17 ppg. Well, Julio's numbers from last year don't represent that....so he picked another player who averaged 17 ppg to get variance in the week to week scores. DD more or less gives you a players average with a point or two more or less. That to me isn't a good indicator of what they really are going to score on a week to week basis.If flc projects a $5 guy is going to average 17 ppg, then he could compare it with the same person he chose for Julio Jones.
yup. i could use fitz, roddy, steve smith or harvin to represent "julio"if i project rookie jimbob as a 14 ppg wr, than i would use washington, bowe or vjax's numbers to simulate weekly varianceto get a more accurate outcome, i would use multiple variations of this. i'd use each of those players numbers and mix them up as well. after 20 trials of this, i should be left with a fairly good idea weather or not jimbob will improve my team enough to be worth the $12 he costs.
 
Looking forward to the lock so I can check out some other rosters. Will be interested to see who ends up the most heavily owned players. Some guys I think are obvious, others not as much.

 
I'm basically down to picking which STL WR I want on my team....or scrap them both and go with two $3 dollar guys. Do I roll with 3 or 2 kickers....maybe picking out of the $4 kickers to try to make my team more unique....Same for the defenses.

Really I just want this contest to lock up....the sooner the better....so I can stop obsessing with it.

 
Really I just want this contest to lock up....the sooner the better....so I can stop obsessing with it.
Agree--I think I've switched back and forth between Garcon and Britt about a dozen times.
Yeah...i can't figure out if I'm going to gamble at Britt or not. I feel like he has good value, but with how I constructed my team, I feel like like it's a gamble I can't take. Then I have flash backs to his first two games of last year (on my team). Then I think about his injury (off my team). Then I think about having a possible top ten guy on my team for the second half of the season (on my team). Then I think about the likelihood of him getting another DWI or something stupid like that (off my team).
 
'Modog814 said:
'flc735 said:
I'm still waiting for what your definition of better is.
i define better as more ppg, higher floor, higher ceiling week by week and as season totals
But regardless, it makes no difference what's better on average.
yes it does. taking a top 3 qb based on YOUR rankings and taking the best cheap qb based on YOUR rankings is better on average than taking two 5-10 ranked qb's based on YOUR rankings who of which will likely cost in the $15-$25 range.if you think matt ryan at $19 is going to be the 4th qb, then i would say to take him. if you think rodgers is the best qb, ryan is the 7th best, romo is the 9th best and tannahill is the 27th best, you are better off taking rodgers and tannahill over ryan+romorodgers+tannehill will score more and is a lot cheaper.
By selecting a Stud+cheap guy or a couple of middle QB's, I get a specific pair. So all i care about it selecting the best pair.
im telling you how to increase your odds of selecting the best possible pair
So for one person specific projections, it COULD make sense to go with a couple middle cost QB's over a "Stud" and Cheap player.It looks like for YOUR specific projections, it's always (or almost) better to go with Stud + Cheap.
you clearly didn't read my earlier posts.i am not saying to take the $30 guy just because he is $30. i am saying take YOUR top guys at each position, even if that means rodgers, foster, calvin and graham.and use some common sense. do i really have to say that? if rodgers at $31 is your #1 qb and ryan at $19 is your #3 qb, then, yes. you should take ryan over rodgers.if rodgers is your #1 and ryan is your #8, then no. despite the way it may seem, you are better off taking rodgers and cheap guy over your qb 8 and qb 9
This is exactly what I've been saying, so I'm not sure why you ever had a problem with it. I never debated the fact that the #1 and #27 QB would out produce the #8 and #9 Qbs. The problem is we dont know the end of year ranks yet, so to say not taking a "stud" qb is the wrong strategy is faulty because we don't know who the stud Qbs are yet. I don't think any one is taking the 9th highest priced qb expecting them to end up QB9 at the end of the year. And if they are then I agree that's the wrong approach.
The problem is we don't know the end of year ranks yet, so to say not taking a "stud" qb is the wrong strategy is faulty because we don't know who the stud Qbs are yet.
no. just because we don't have a crystal ball doesn't mean all strategies are equal. 1 strategy is slightly better than the other. now maybe making a stud roster may give you a 10% chance of making the 250 as opposed to an 8% chance, but it still gives you an edge.your projections have nothing to do with what i am saying. if you don't get them right, it doesn't matter what strategy you use
 
Last edited by a moderator:
What are the odds that the MJD/Jennings situation is cleared up by the time this contest locks??? That might have the potential to change a lot of people's rosters, including mine.

 
What are the odds that the MJD/Jennings situation is cleared up by the time this contest locks??? That might have the potential to change a lot of people's rosters, including mine.
I think I read in another thread that MJD has to sign by Tuesday or he wont be able to play week 1 (probably not going to play regardless the first week) and is going to lose a lot of money. Contest locks at midnight on Tuesday. Should be an interesting day for your contest team.
 
What are the odds that the MJD/Jennings situation is cleared up by the time this contest locks??? That might have the potential to change a lot of people's rosters, including mine.
I think I read in another thread that MJD has to sign by Tuesday or he wont be able to play week 1 (probably not going to play regardless the first week) and is going to lose a lot of money. Contest locks at midnight on Tuesday. Should be an interesting day for your contest team.
i remember when vjax held out, there was a reason he needed to sign by week 11 (or something). is there anything like that with mjd?
 
What are the odds that the MJD/Jennings situation is cleared up by the time this contest locks??? That might have the potential to change a lot of people's rosters, including mine.
Saw in another thread that he is reporting today. Jennings will NOT be on my roster this year...
 
That's why three $10 WRs can outproduce one $30 WR
it's possible, but the odds are overwhelmingly against that. i showed you this.
No you didn't. That's my point.Edited to further clarify your error:
ignoratio, i've said time and time again to use your own projections. price doesn't have anything to with your projections.
OK, so I'm going to project that every $10 WR is going to average 20 PPG, and every $20+ WR is going to average 8 PPG. Plug those into your Excel worksheet and voilà! I've "proven" that a combination of cheaper WRs is better than stud WRs. Does that seem like a pretty crappy analysis? Of course - but it's plagued by the same exact flaw as your own conclusions. You haven't actually demonstrated anything with your analysis, other than the relatively obvious, "Players who score more points are better than players who score fewer points."
 
Last edited by a moderator:
I was a bit late to re-up my subscription this year, but still managed to get back in time to start tinkering with my contest roster for the first dozen or so times last night. After missing the cut in week 11 by less than a point in my rookie attempt last season, I can't wait to give it another go. Great to see that the perennial arguments still haven't been settled (and likely never will be). At least the winning strategy of loading up on kickers and cheap defense seems to still be intact.

My current rendition is sitting pretty at 27 players (down from 29 last year), and has used up every last dollar of cap space. As it stands now, I'll have 2 players making a repeat appearance on my roster. The rest of it is all fresh value. Feeling pretty good about it right now. My only regret is that Shayne Graham was excluded from the kicker options this year. If I would have been able to add him to the kicker stable, we'd already be talking about a guaranteed berth into the final 250.

 
That's why three $10 WRs can outproduce one $30 WR
it's possible, but the odds are overwhelmingly against that. i showed you this.
No you didn't. That's my point.Edited to further clarify your error:
ignoratio, i've said time and time again to use your own projections. price doesn't have anything to with your projections.
OK, so I'm going to project that every $10 WR is going to average 20 PPG, and every $20+ WR is going to average 8 PPG. Plug those into your Excel worksheet and voilà! I've "proven" that a combination of cheaper WRs is better than stud WRs. Does that seem like a pretty crappy analysis? Of course - but it's plagued by the same exact flaw as your own conclusions. You haven't actually demonstrated anything with your analysis, other than the relatively obvious, "Players who score more points are better than players who score fewer points."
No you didn't. That's my point.
yes, i did.
OK, so I'm going to project that every $10 WR is going to average 20 PPG, and every $20+ WR is going to average 8 PPG. Plug those into your Excel worksheet and voilà! I've "proven" that a combination of cheaper WRs is better than stud WRs. Does that seem like a pretty crappy analysis? Of course - but it's plagued by the same exact flaw as your own conclusions. You haven't actually demonstrated anything with your analysis, other than the relatively obvious, "Players who score more points are better than players who score fewer points."
maybe you will understand better if i keep my answers simple. no, i did not say that. we both know i did not say that so why would you even try and slip that in? i said a stud+scrub is better than two mid players. before i looked into this, i had the same opinion you did. i did the research and realized i was wrong. unlike you though, i am able to accept that and adjust.i know you have a very good understanding of numbers. if you bothered to do some work yourself, you would come to the same conclusion as i did. if you don't want to do that than i'm not going to waste any more time on this. :bye:
 
I think I'm done tweaking. I went with 23 players with zero bye week issues. No multi stars out during bye weeks... meaning two of my main lineup at one position.

Have to take some chances and my QB spot is where I am by only having two as of right now.

 

Users who are viewing this thread

Top