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.

Help! Need some shark advice (1 Viewer)

jdoggydogg

Footballguy
I need help from this group. I only play GPPs, so that's how I'm gearing my questions. So here's the deal: I've won enough over the years to still be way up money wise, but I have not had a winning week so far this season. Last season wasn't good, either. I've tried paying close attention to rostership percentages, that didn't work. I've tried just putting together killer lineups with just a couple cheap low-owned players, that didn't work.

So: those of you that are winning in GPPs, what is your winning formula?

 
So, there is a brilliant discussion in the Week 2 and week 3 Fanduel threads below. But here's what I do each week.

I find the 4 games with the highest O/U from Vegas. I play GPP lineups built as follows:

20% of my lineups (LUs) for each QB from the team winning each of those games, 5% for the QB losing those games. I stack each QB with his RB and WR, and then also the opposing WR. Then I find 6-8 RBs with the highest volume (attempts + targets), and set them to 30% or so exposure (leave the rest of the RBs around 5% to get a tiny bit of exposure on some other guys). Then I put this in the FBG LU builder and let it build my LUs. Been consistently putting a couple GPP lineups in the top 5% each week.

For cash I take a look at the pool of players above, and try to build a lineup of ONLY those 4 games. So, I'll make a judgement call on a stack, usually based on QB value, and stack that game, then stack a RB-WR from a second game, then try and fill the DEF in with something cheap that will play a crappy team, and round out the LU with players from the 4 games that seem like high floor types.

EDIT: I usually play 25% of my stakes in GPP and 75% Cash.

 
Last edited by a moderator:
The reasons I landed on that style of building LUs was some data that one of the FBG folks (@KarmaPolice I think) that did a boatload of research and basically distilled it to this:

1) The QB-WR1 stack has a VERY high correlation with high-scoring LUs in FD GPPs.

2) The teams that won the Sunday Millions the first 2-3 weeks this year all had a stack of QB-RB-WR, opposing WR, and another stack of RB-WR. That caught my attention big time. I found it really hard to build LUs that way, so I pared it down a bit.

The real key to the GPPs seems to be able to limit the RB pool you choose from to something manageable but almost solely high-volume guys.

 
This week was weird because of Davante and Lockett. They ended up being SO valuable that you had to have them to cash. That's not usual. You need to score over 8 TDs with a lineup to have a chance at money in GPP.

 
That's helpful. Although you might need to do a zoom tutorial for me to understand the lineup optimizer. It was confounding to me when I tried it. 

 
I need help from this group. I only play GPPs, so that's how I'm gearing my questions. So here's the deal: I've won enough over the years to still be way up money wise, but I have not had a winning week so far this season. Last season wasn't good, either. I've tried paying close attention to rostership percentages, that didn't work. I've tried just putting together killer lineups with just a couple cheap low-owned players, that didn't work.

So: those of you that are winning in GPPs, what is your winning formula?
I'm in the same boat. I only play GPPs.  I did very well last year and ok in previous years. This year has been awful, including the game just now. I've only had one winning week. 

This week was weird. My best lineup had 201 points and barely cashed. I had the Wilson/Lockett combo and AJ Brown and H. Bryant, but also less than 6 points combined from Drake and Claypool. 

I would say the stacking strategy is still the best option for GPP. Start with the QB/WR or TE stack from Team 1 and bring it back with one WR, RB, or TE from Team 2. This week the milly winner followed that strategy with Wilson/Lockett/Hopkins. I will say this strategy seems to be working less this season than previous seasons. But it did work this week, if you picked Lockett instead of Metcalf. 

On that note, the QB/RB stack almost never works unless it's a monster PPR back like Kamara or McCaffrey. And bringing it back with the opposing team's RB is lower success rate too. 

But just when you think you have some sort of formula figured out, the backup backup TE (Harrison Bryant this week) will feature in the winning lineup. And yes I had Bryant on multiple teams too, just had a hunch. But only 1 lineup with Lockett. Way too many with Metcalf. 

 
I'm in the same boat. I only play GPPs.  I did very well last year and ok in previous years. This year has been awful, including the game just now. I've only had one winning week. 

This week was weird. My best lineup had 201 points and barely cashed. I had the Wilson/Lockett combo and AJ Brown and H. Bryant, but also less than 6 points combined from Drake and Claypool. 

I would say the stacking strategy is still the best option for GPP. Start with the QB/WR or TE stack from Team 1 and bring it back with one WR, RB, or TE from Team 2. This week the milly winner followed that strategy with Wilson/Lockett/Hopkins. I will say this strategy seems to be working less this season than previous seasons. But it did work this week, if you picked Lockett instead of Metcalf. 

On that note, the QB/RB stack almost never works unless it's a monster PPR back like Kamara or McCaffrey. And bringing it back with the opposing team's RB is lower success rate too. 

But just when you think you have some sort of formula figured out, the backup backup TE (Harrison Bryant this week) will feature in the winning lineup. And yes I had Bryant on multiple teams too, just had a hunch. But only 1 lineup with Lockett. Way too many with Metcalf. 
I hear you. I've been stacking for years. But perhaps I'm not running back the stack with enough players from the opposing team. For example, I had a Seattle stack this week with Deandre Hopkins on the other side and I didn't do well. 

 
The reasons I landed on that style of building LUs was some data that one of the FBG folks (@KarmaPolice I think) that did a boatload of research and basically distilled it to this:

1) The QB-WR1 stack has a VERY high correlation with high-scoring LUs in FD GPPs.

2) The teams that won the Sunday Millions the first 2-3 weeks this year all had a stack of QB-RB-WR, opposing WR, and another stack of RB-WR. That caught my attention big time. I found it really hard to build LUs that way, so I pared it down a bit.

The real key to the GPPs seems to be able to limit the RB pool you choose from to something manageable but almost solely high-volume guys.
#1 is the key most weeks. It offers the highest probability of success. And usually bring it back with WR1 from the opposing team. It worked again this week with Wilson/Lockett/Hopkins, but there are some weeks where the formula doesn't work at all. 

 
I hear you. I've been stacking for years. But perhaps I'm not running back the stack with enough players from the opposing team. For example, I had a Seattle stack this week with Deandre Hopkins on the other side and I didn't do well. 
yes same. If you didn't have Adams and/or hit on the high scoring cheap RBs like J.Williams or F.Wilson Jr, then it may not have even cashed. Also I had way too many teams with Wilson/Metcalf/Hopkins. 

And you know what will happen next week if you start Lockett. 

 
Last edited by a moderator:
So, if I'm reading your advice right, game stacks are the key to victory in GPPs?
I posted in one of the threads a breakdown of some of the top LUs in the milly maker.  I am pretty sure after 5 weeks it was a game stack dominating the top of the leader boards.  Not only that, but all had a 2nd correlation in the lineup too.  

If I remember right, there were a lot at the top with a gb/houst or az/Seattle game stack, plus another correlation like clev te + Boyd or Gibson + wash D or Lockett + kirk, etc.  

 

Users who are viewing this thread

Top