NFL FORUM

Comments

  • JonBales

    RotoAcademy Lead Instructor

    • 575

      RG Overall Ranking

    • 2019 DraftKings FFWC Finalist

    • x3

      2015 DraftKings FFWC Finalist

    I’ve been doing some research on optimal lineup construction, and I’ve really started to buy in more and more to the idea of unbalanced teams. I mean ‘unbalanced’ in a couple different ways.

    One is cap allocation. How often do we see one top WR, one second-tier guy, and one cheap option in lineups? I think it’s natural to want to create a balanced lineup, though not always optimal. A specific week of NFL and certain values will dictate the best strategy, but in general, I think there’s a lot of value in potentially being unbalanced. One of the reasons for that is being contrarian; you can potentially create a really unique lineup without actually buying in on any super low-value contrarian plays, and being unbalanced is one way to do that because it’s unlikely to run into many lineups with three stud WRs, for example.

    The other way I’m thinking of being unbalanced is in terms of ownership; it seems like there are often a lot of “semi-contrarian” lineups that end up trying to sort of decrease ownership a tad and still maximize value for the most part, and that middle ground I think might be the worst spot to be. Instead, I’ve been thinking of how useful it might be to pair a couple very clear low-ownership options with chalk. You’re likely securing the top values, but then trying to differentiate your lineup with just a couple contrarian selections mixed in.

    I think that extreme type of approach – one of risk-minimization on one end and extreme risk-seeking on the other – is probably better than trying to “have it all” with each individual selection and create balance with your lineup.

  • cortiz50

    I think you may be right since many people have a natural instinct to want and balance things out given the salary cap constraints. In your research – did you go back and look at data on Fantasylabs to see if this holds true for winning GPP lineups. I assume that’s what you are talking about here.

  • JonBales

    RotoAcademy Lead Instructor

    • 575

      RG Overall Ranking

    • x3

      2015 DraftKings FFWC Finalist

    • 2019 DraftKings FFWC Finalist

    I don’t actually have that data (individual lineups) to examine, but I did have some data suggesting unbalanced lineups as a whole are more profitable than balanced ones. I do think there’s a little bit of a problem in analyzing points and even profitability because there’s a selection bias that favors unbalanced lineups; when there are one or two obvious punts (maybe a starter gets injured and the backup is min-priced), an unbalanced strategy is clearly optimal.

  • 8MileAllstars

    It seemed that last year on DK, P. Manning and D. Thomas were under-owned for their level of production and became quasi-contrarian plays because of their cost. I’ve put together a potential week 1 lineup with them, two of the more expensive RBs, a couple of “value” WR’s and some flyers at TE and Flex that I expect to be less than 5% owned.

  • Priptonite

    • Blogger of the Month

    @Jon Bales said...

    I’ve been doing some research on optimal lineup construction

    What does this even mean?

  • BeepImaJeep

    • 2015 DraftKings FFWC Finalist

    • x7

      2015 DraftKings FBWC Finalist

    @Priptonite said...

    What does this even mean?

    $$$

  • pmsimkins

    • 2014 FanDuel WFBC Finalist

    • 2015 FanDuel WFBBC Finalist

    I’m not entirely trying to be flippant, but the optimal lineup construction is the one that scores the most points. I recognize that system betting is comforting to people as it cuts down on tough choices and allows you to think “I did everything right, so must have been bad luck I lost”. The problem is that it limits your thinking and rules out options that may be correct, before you really consider them.

    If you dig deeper in your research, that pertains to actual on field performance, you’ll have a balance of contrarian and high owned guys naturally, without forcing it.

  • JonBales

    RotoAcademy Lead Instructor

    • 575

      RG Overall Ranking

    • x3

      2015 DraftKings FFWC Finalist

    • 2019 DraftKings FFWC Finalist

    I couldn’t disagree more with the idea that an optimal lineup is one that scores the most points. In any league, an optimal lineup is one that maximizes profit. Points are solely a mechanism to securing profit. In a tournament, a profit-maximizing lineup is very likely to resemble one that maximizes win probability – which more often than not won’t look anything like a lineup that maximizes points.

  • baseballs

    @Jon Bales said...

    I couldn’t disagree more with the idea that an optimal lineup is one that scores the most points. In any league, an optimal lineup is one that maximizes profit. Points are solely a mechanism to securing profit. In a tournament, a profit-maximizing lineup is very likely to resemble one that maximizes win probability – which more often than not won’t look anything like a lineup that maximizes points.

    well said

  • ATLSoulchild

    Maybe it is just me, but I have no idea what this entire conversation is about. Is it about choosing 2 expensive WR’s and then a cheap one like Eddie Royal? Or am I just completely off base?

  • deactivated79425

    In novice terms if a starter is hurt use the cheap back up to get better players. got it

  • kbarnhill7523

    • 408

      RG Overall Ranking

    • 2018 FanDuel WFFC Finalist

    • x2

      2017 FanDuel WFBBC Finalist

    But what happens when the chalk is the punt too?

    I can think of Tre Mason last year when he was in the high $5000s and then went off for like 100 yds and 3-4 TDs.

    Davante Adams at $5500 (on FD) comes to mind in Week 1. I think most people will gravitate to him to get a piece of the Packers against the Bears… but Randall Cobb might be lower owned because of it. Obv there will be Rodgers/Cobb/Adams stacks all over the place… I look at Week 1 and think there is no way I can fade the Packers in GPPs because if they go off and I don’t have them the day is over.

    I feel like this is what you are talking about though, right? Go with the chalk in this case, and then look for low ownership elsewhere?

  • crazypaul

    @ATLSoulchild said...

    Maybe it is just me, but I have no idea what this entire conversation is about. Is it about choosing 2 expensive WR’s and then a cheap one like Eddie Royal? Or am I just completely off base?

    I think what he is trying to say is that sometimes its better to stack a position with studs as opposed to the normal thinking people do when building a lineup. I’d guess on most days that the vast majority pays up for 1 top tier player at 2 different positions and then fills in around that with value/mid-range options.

    I believe Bales is saying that to counteract that, sometimes it might make more sense to pay up for 2 studs at the same position because most of your competition won’t. As an example of this, I made a lineup on DK yesterday that uses 3 of the most expensive WR’s available & filled in the rest of my rosters with some nice value. Could it blowup in my face? Sure, but by doing that I have an explosive lineup that could blow past everyone else.

  • chris1485

    I think more of what he is saying is most people ( myself included) like to go low tier, mid tier, and high tier to create a balanced lineup. Instead what I believe he is suggesting is go multiple low tier players in one position and multiple high tier players in another ,thus creating a unique unbalanced lineup. If your low tiers hit you’re going to be in really good shape. I understand this for the huge tourneys, but hitting on multiple low tiers is not an easy task. But like he said you will definitely maximize profit . If you fail 9 out of ten times you are more than likely going to make 10x your money on the one you hit and quite possibly much more than that. This seams like something that is worth experimenting with.

  • Yukerboy

    • Blogger of the Month

    Don’t be afraid to scrub out a position. An Adams/Lafell/Floyd WR set would differentiate you enough to where chalky Luck/Hill/Murray can win some monies.

  • draftcheat

    • x3

      2016 DraftKings FFWC Finalist

    • x7

      2014 DraftKings FBWC Finalist

    @Jon Bales said...

    I couldn’t disagree more with the idea that an optimal lineup is one that scores the most points. In any league, an optimal lineup is one that maximizes profit. Points are solely a mechanism to securing profit. In a tournament, a profit-maximizing lineup is very likely to resemble one that maximizes win probability – which more often than not won’t look anything like a lineup that maximizes points.

    90% of DFS players cannot, or refuse to, believe and act on this philosophy, but it’s completely true. In almost every DFS GPP, the top projected points per dollar players are not actually the best plays.

  • Yukerboy

    • Blogger of the Month

    @draftcheat said...

    90% of DFS players cannot, or refuse to, believe and act on this philosophy, but it’s completely true. In almost every DFS GPP, the top projected points per dollar players are not actually the best plays.

    90% lose. Correlation is not causation, but it’s a damn good indicator.

  • YoungFischer

    Profit is a function of multiple factors, including things you have no control over such as your opponent’s rosters.

    The goal is to create a roster that maximizes expected fantasy points produced—median expected points for a cash game and upper tail for GPPs.

  • pmsimkins

    • 2014 FanDuel WFBC Finalist

    • 2015 FanDuel WFBBC Finalist

    @Jon Bales said...

    I couldn’t disagree more with the idea that an optimal lineup is one that scores the most points. In any league, an optimal lineup is one that maximizes profit. Points are solely a mechanism to securing profit. In a tournament, a profit-maximizing lineup is very likely to resemble one that maximizes win probability – which more often than not won’t look anything like a lineup that maximizes points.

    Maybe we just don’t understand each other, but I’ve possibly never seen anything on here that makes less sense than this.

    Winning tourny = LU which scores the most pts

    Winning tourny = optimum profit

    I’m sorry but there’s no magic bullet LU construction. The solution is accurately projecting player performance and then fitting the project pts/$ jigsaw puzzle to achieve a maximum point total for your team.

  • jellish

    @pmsimkins said...

    Maybe we just don’t understand each other, but I’ve possibly never seen anything on here that makes less sense than this.

    Winning tourny = LU which scores the most pts

    Winning tourny = optimum profit

    I’m sorry but there’s no magic bullet LU construction. The solution is accurately projecting player performance and then fitting the project pts/$ jigsaw puzzle to achieve a maximum point total for your team.

    I’m not saying I agree one way or the other, but I’m very confused by this. I would think along these same lines, wouldn’t the LU that scores the most points maximize profit because it’s going to win tournaments? Or am I completely missing the point of this?

  • pmsimkins

    • 2014 FanDuel WFBC Finalist

    • 2015 FanDuel WFBBC Finalist

    @draftcheat said...

    90% of DFS players cannot, or refuse to, believe and act on this philosophy, but it’s completely true. In almost every DFS GPP, the top projected points per dollar players are not actually the best plays.

    Of course the lineup that should do best doesn’t often win. For every optimal play there are 20 sub optimal plays. This, obviously, makes it far more likely that a sub optimal play ends up being the highest scorer. That doesn’t make sub optimal plays the “right” decision.

    Keep in mind I’m coming from the perspective of 1-3 LUs not a mutual fund of 500.

  • ldavidjm

    @pmsimkins said...

    Of course the lineup that should do best doesn’t often win. For every optimal play there are 20 sub optimal plays. This, obviously, makes it far more likely that a sub optimal play ends up being the highest scorer. That doesn’t make sub optimal plays the “right” decision.

    Keep in mind I’m coming from the perspective of 1-3 LUs not a mutual fund of 500.

    But if you’re playing in a tournament (where you only pick one RB) with 1000 people in in, 500 people picked Jamaal Charles, 499 people picked Eddie Lacy, and no one is picking Frank Gore, you should pick Frank Gore.

  • pmsimkins

    • 2014 FanDuel WFBC Finalist

    • 2015 FanDuel WFBBC Finalist

    @ldavidjm said...

    But if you’re playing in a tournament (where you only pick one RB) with 1000 people in in, 500 people picked Jamaal Charles, 499 people picked Eddie Lacy, and no one is picking Frank Gore, you should pick Frank Gore.

    Agreed, but this is a hugely unrealistic scenario that has virtually no bearing on real life.

    Lets say ownership in a group of 3 total available players actually fell as you’re describing. This would imply the the .1% owned player has very very little chance of success. Lets say though that his chance of scoring the highest out of the 3 is 10x higher than his ownership. This would be massive in real life. Under that scenario your mean wait time would be 100 contests before he came through for you and did score highest. You’d have to have a bankroll to sustain atleast 300 lost contests in case it took longer for his 1% shot to come through. Now is your .1% owned play still smart?

  • Priptonite

    • Blogger of the Month

    @jellish said...

    I’m not saying I agree one way or the other, but I’m very confused by this. I would think along these same lines, wouldn’t the LU that scores the most points maximize profit because it’s going to win tournaments? Or am I completely missing the point of this?

    Bales is arguing that a lineup full of chalky plays that have great expected production in unlikely to win a large-field tournament because of ownership. Even if all your guys do well, you’ll be in a pack with all of the other players who also took those guys. The people at the very top of the tournament most likely got outlier performances from a few low-owned players.

    pmsimkins is arguing that there is no “optimal lineup construction strategy,” which I agree with. I think it’s stupid to try to teach people how to build a roster when every slate is different. The core skill in DFS is evaluating players, not how to construct a contrarian roster. Players are basically taught, “hey it’s ok to make subpar choices because other people won’t have them.” It’s not that that statement is necessarily false, but those should still be informed choices. Building a contrarian roster is easy, but you can’t do it intelligently unless you can accurately evaluate players.

    No offense to Bales, but a lot of the “53% of winning cash game rosters contain a stud RB” type stats that he reports are probably more detrimental to players than they are helpful. I don’t think his intention is to tell people “YOU NEED A STUD RB TO WIN CASH GAMES,” but that’s certainly how a lot of people interpret it. I don’t know how many times this MLB season I saw, “well Bales book said that the optimal stack is X players.” But not every slate is the same, not every team is the same, etc. You do yourself a disservice by blindly following lineup construction rules.

  • draftcheat

    • x3

      2016 DraftKings FFWC Finalist

    • x7

      2014 DraftKings FBWC Finalist

    @pmsimkins said...

    Of course the lineup that should do best doesn’t often win. For every optimal play there are 20 sub optimal plays. This, obviously, makes it far more likely that a sub optimal play ends up being the highest scorer. That doesn’t make sub optimal plays the “right” decision.

    Keep in mind I’m coming from the perspective of 1-3 LUs not a mutual fund of 500.

    The number of lineups one is entering into the tournament is not relevant to this discussion, imo.

  • Priptonite

    • Blogger of the Month

    @draftcheat said...

    The number of lineups one is entering into the tournament is not relevant to this discussion, imo.

    It absolutely is.

    For a player who can only afford to enter 1-3 NFL lineups per week, it would take them at least 2 seasons just to hit 100 entries. Don’t you think that player might be interested in entering lineups that give him a better chance of cashing, even if he’s giving up some top 0.001% finish equity?

    Compare that to someone who can submit that 2-seasons worth of entries into 1 contest.

    edit: lolmath

  • X Unread Thread
  • X Thread with New Replies*
  • *Jumps to your first unread reply

Subforum Index

RotoGrinders.com is the home of the daily fantasy sports community. Our content, rankings, member blogs, promotions and forum discussion all cater to the players that like to create a new fantasy team every day of the week.

If you or someone you know has a gambling problem, crisis counseling and referral services can be accessed by calling 1-800-GAMBLER (1-800-426-2537) (IL). Gambling problem? Call 1-800-GAMBLER (NJ/WV/PA/MI), 1-800-9-WITH-IT (IN), 1-800-522-4700 (CO), 1-800-BETS OFF (IA), 1-888-532-3500 (VA) or call/text TN REDLINE 1-800-889-9789 (TN).