STRATEGY FORUM

Comments

  • MrMadness001

    I have played DFS now for about 3 years. I am mostly a very small stakes, single entry GPP player. Over the 3 years I have lost more money then I have won. Roughly around $160 for each year. Now I know most people will immediately say you need to also mix in cash games with your play and use bankroll management. Which I have read a lot about and have tried over the past few years of playing. Some how I always find myself playing way to much during the NFL season. Probably since it’s a much shorter window.

    When I build my LU’s I look at various different stats for teams and players. But I always seem to wind up paying some website for projections, articles, recommended picks and the like. My question is by doing that am I not really getting better as a player? And do I need to really dig deeper in researching the sports themselves for a better chance at winning consistently?

    Is it worth my time to try to create my own data models and projections instead of jumping from one site to the next trying to find one that works best for me? I have read a bunch of older threads here on the RG forum and most people say creating models of their own is a tough task and it could take a while or even never happen that the model is very successful.

    Overall I do enjoy playing DFS because it gives me way more knowledge about the players themselves and I have a vested interest to now follow not just my hometown team but the entire league. I guess I am just curious as to what other small stakes players or even those who play say $100-$200 a night do when they go about creating their LU’s? How difficult would it be to create my own projections at the least?

  • Heterodox

    @MrMadness001 said...

    Is it worth my time to try to create my own data models and projections instead of jumping from one site to the next trying to find one that works best for me? I have read a bunch of older threads here on the RG forum and most people say creating models of their own is a tough task and it could take a while or even never happen that the model is very successful.

    A lot of the people making those comments are people who don’t want you to do it.

    First off, I should say that I have tons of time. No girl, no kids, job isn’t too demanding. When you ask if its worth your time to create models, for me, that was part of the point. When I decided dfs was more than just a fun way to gamble on sports, I realized I would need some technical ability to properly organize information and evaluate players, not just in terms of ability but in terms of dfs value for any given slate. I had never used excel before, but I dove in and started learning all the functions, and where to get information, and how to combine it, etc. This was part of the fun of playing dfs for me.

    What I ended up with was a sheet that is very similar to the tables that sites like this one offer, or that people have discussed in threads on the subject of making spreadsheets. So then I was like, what’s the point of that? Sure, I developed a skill I didn’t previously have, but how does that help me, when everyone else is using the same thing, and in many cases what they have is more robust than mine. I was still just relying on a presumed ability to use the same info everyone else has, better than everyone else is using it.

    So, now I’m learning R, and SQL, and anything else. Those are also real-life skills which I will be glad to have, even if they amount to nothing. I don’t know what I’ll get out of it in terms of dfs, but, again, I have tons of time, and I enjoy learning it, so why not?

    In the meantime, I have done one simple thing with my spreadsheets. Aside from making sure that I’m more thorough than what I see on sites like RG in terms of splits and de-emphasizing less relevant information, I took the simple step of, rather than just ranking everyone or assigning values, or applying filters based on certain criteria, I line up the match ups, combine everyone’s averages, and come up with baseline projections. I know they aren’t as robust as professional projections. I have to mentally or manually adjust for some situations, and I don’t always have a precise mathematical value for those adjustments. But they are much more convenient to use than the previous versions of my spreadsheet, and they account for every bit as much information. I used to sort by rankings, then filter, then sort again by another ranking, and I was never all that comfortable with what I ended up with. And weighting the rankings and sorting that way just felt clunky and imprecise. The projections I have, which are not that difficult to make, are as good as anything anyone is starting with, and probably better than the standard spreadsheets most people are working with. If not, they’re just as good, and easier to work with. The only thing that separates vegas or the pros is what they add to them, and the processes they have for doing so.

    Baseline averages don’t give you floor or upside estimates, or probability distributions, but those are the next step.

    I don’t know if it’s worth your time, specifically, but I know it is worth mine, and it’s also part of the game for me to keep developing skills that will help me improve, and that’s part of the enjoyment I get out of this. Hope this was on topic, got distracted part way through, just my two cents.

  • aramirez24

    Sounds like you and I are in the same boat. I don’t play enough volume to consider making my own models. I have done pretty well using the RG lineup builder and projections.

    I recently got a subscription to Fantasy Labs because they allow you to create your own models and create your own trends without you having to worry about the scripting stuff. It’s been quite the learning curve to figure it all out but it does give me the sense that I am finding the value and plays myself rather than reading a picks article. I know there are a few other sites that let you do the same thing like RotoQL and what not. You may want to try out those and see what you think.

  • Jvanspro

    I have my own model but still do my lineups with pen and paper. For me it’s just another tool in the daily research. To be honest though, just using projections and a solver is not a winning process. It’s much better to use your own brain to build lineups. Use a model to find plays you hadn’t thought of or to confirm what you are thinking about a player. Not just build the lineup for you.

  • BigRay

    When you use other peoples projections/processes you never really improve. you just keep searching for someones elses picks. If you come up with your own process, even if it sucks, you can keep tinkering with it , learning, adapting, and that will help you improve as a player

  • FantasyTime69

    @Heterodox said...

    What I ended up with was a sheet that is very similar to the tables that sites like this one offer, or that people have discussed in threads on the subject of making spreadsheets. So then I was like, what’s the point of that? Sure, I developed a skill I didn’t previously have, but how does that help me, when everyone else is using the same thing, and in many cases what they have is more robust than mine. I was still just relying on a presumed ability to use the same info everyone else has, better than everyone else is using it.

    Care to share any links or a push in the right direction for one whos interesting in learning how to make models but doesn’t know where to look?

  • maxeernst

    @FantasyTime69 said...

    Care to share any links or a push in the right direction for one whos interesting in learning how to make models but doesn’t know where to look?

    I think Seth yates did some stuff on this for beginners… This was the only one I could find, but I think there is something more extensive.

    https://rotogrinders.com/blog-posts/creating-a-projection-system-for-beginners-399030

  • Heterodox

    @FantasyTime69 said...

    Care to share any links or a push in the right direction for one whos interesting in learning how to make models but doesn’t know where to look?

    Well, I’m not the best guy to give you tips because I’ve taken an insanely scattershot approach. I have no background in programming or statistics, so I’ve had to go back at various points and fill in the gaps on one side or the other. I’m also incredibly unfocused in general. But I find myself on sites ranging from econometrics blogs, or regression analysis tutorials, youtube series on data anlysis in R, to old textbooks that are online for free, “For Dummies” books that always seem to come up on Google for free, help forums on a variety of topics. Basically, I just google whatever question I have, I’ll open up a ton of links, and just go through them. It’s not an efficient process as I wind up spending a lot of time wading through things that either aren’t totally relevant or that cover topics I haven’t gotten to yet, but I’m an immersive learner so that’s how I do it.

    The more deliberate approach comes in the form of MOOCs. MOOCs are awesome. EdX and Codeacademy have a number of courses that cover various aspects of the math and programming involved in predictive modeling. EdX has a course on Sabremetrics which is great because it not only gives you a grounding in the statistical concepts, but it works you through the basics of getting set up and working with your data in R and SQL. A lot of the tutorials I watch on youtube, or the blogs I read, not to mention the MOOCs, are either general, or they’re geared towards the more common applications of data analysis, whereas the sabremetrics course is specifically focused on baseball, so it’s more natural to apply it.

    Also, there’s a guy who posts here, Shockermandan I think is his name. I randomly found some youtube videos that turned out to be by him, where he basically does the kind of simple projections in excel I was talking about, particularly in his “top-down analysis” video.

  • slcseas

    Information is only as good as the use you find for it. Financial information for every public corporation is available to the masses, but 90% of people can’t do dick with it. The skill is drawing inferences from the information, not being able to build a spreadsheet.

    It’s ok to pay for a service if you don’t have the time / energy / ability to build your own. I think it’s important to understand the data you are looking at though, what it infers, and what it might fail to consider. Said another way, try to figure out why you are being told to use a particular player, rather than just plugging him in without a second thought. This should lead to you to start thinking about the finer points, which is where profit is going to come from.

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

Use our links to sign up and deposit on sites listed in this thread to get these bonuses:

  • FanDuel

    Get 1-month of RotoGrinders Premium for FREE (a ~$40value) by signing up through one of our links!

    Learn More
  • DraftKings

    Sign up for DraftKings using a RotoGrinders link & receive our DraftKings Premium content FREE for 1 month. That’s a ~$40 value! No DraftKings promo code necessary!

    Learn More
  • FantasyDraft

    FantasyDraft strives to put players first, with a mission to “provide a fun and fair experience for all.” To this end, the site has a well-built, easy-to-use interface and a the first of its kind in offering “Rake-Free” fantasy contests.

    Learn More

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. Our goal is to help all of our members make more money playing daily fantasy sports!

Bet with your head, not over it!
Gambling Problem? Call 1-800-Gambler