• Welcome to OGBoards 10.0, keep in mind that we will be making LOTS of changes to smooth out the experience here and make it as close as possible functionally to the old software, but feel free to drop suggestions or requests in the Tech Support subforum!

Official Daily Fantasy Sports (DFS) Thread

DeaconBrews said:
How do you get this type of info? Is it just PFF?
Click to expand...
DEF ranks is a combination of sources. I make my own using DVOA from Football Outsiders and PFF data. Football Outsiders also provides data for team match-ups vs WRx. And then specific match-up analysis comes from Rotoworld.
 
StarsDraft makes a big announcement today.

Update to List of Accepted States

Following a review of recent daily fantasy sports developments within a number of jurisdictions, we regret to inform you that StarsDraft is now limiting its operations to a select list of states. Starting today, only players physically located in Kansas, Massachusetts, Maryland and New Jersey will be able to participate in our fantasy sports contests.
 
Is Gurley going to be in every lineup this week?
 
for the first time ever I picked a reliever (osuna) on dk, maybe it will work out
 
buckets said:
Is Gurley going to be in every lineup this week?
Click to expand...
I expect his 50/50 usage to be around 70-75% which is ridiculous high. Tourney usage probably around 45%. I'll probably build a lot of tournaments with him and Freeman as the chalk, and then high contrarian stacks.
 
AnonymousDeac said:
I expect his 50/50 usage to be around 70-75% which is ridiculous high. Tourney usage probably around 45%. I'll probably build a lot of tournaments with him and Freeman as the chalk, and then high contrarian stacks.
Click to expand...

Is that usage info listed somewhere? Or is this just gut?
 
dot com said:
Is that usage info listed somewhere? Or is this just gut?
Click to expand...

If you're doing tourneys starting on Sunday, the ownership percentages become available after the start of the Thursday night game. It's not a perfect indicator because there's one less game to choose from and fanduel vs DK salaries differ slightly, but I've found it to be pretty close. You can find this info on dfsreport.com and rotogrinders.
 
Bourrc9 said:
If you're doing tourneys starting on Sunday, the ownership percentages become available after the start of the Thursday night game. It's not a perfect indicator because there's one less game to choose from and fanduel vs DK salaries differ slightly, but I've found it to be pretty close. You can find this info on dfsreport.com and rotogrinders.
Click to expand...
Yes, this is correct. Those percentages were my guesses based off where I'm seeing talking heads rank them and their prices across sites. Other than FantasyAces, he is priced as the #10 to #15 RB. And then on FantasyHub he is somehow priced as the #25th most expensive. However his rank this week is a clear top 5 play, probably even top 3.
 
If the advanced stats people are using to rank the players point out clear values, what advanced stats are the sites using to set prices? Are they just going by prior usage or a more primitive model than the high rollers are using?
 
PhDeac said:
If the advanced stats people are using to rank the players point out clear values, what advanced stats are the sites using to set prices? Are they just going by prior usage or a more primitive model than the high rollers are using?
Click to expand...
The big sites don't actually have a model directly creating the prices. They have guys on staff (usually ex-professionals) that are manually creating the prices. Now do those guys use some models? Sure. But I don't believe their goal is to specifically price the guys according to their supposed rank.

The goal of each site is to acquire players and get them to participate in as many contests as possible, right? So when people see that Todd Gurley is very cheap but a great match-up, they feel a lot better about their line-ups. Sites try to find that balance that allows people to choose the most popular players with still keeping pricing competitive so it's not just all elite players in line-ups.
 
anyone have much experience / success with NBA DFS?

i feel like it's easier to do at least some rudimentary modeling and the more popular dfs gets, the bigger advantage just doing those basics will have.

is there an accessible database anywhere that has individual game results by player? I'd imagine if you could basically assign players into clusters / personas (low usage 3 pt shooter, stretch 4, post scorer, etc.) it wouldn't be too hard to figure out favorable matchups.
 
I don't know if there's something equivalent for NBA, but I use NFLDB which is a python library and database to do something similar. In my experience it's really hard to accurately measure matchups until you have a large sample size and by then the season is well along.
 
I looked into NFLdb when building a model for DFS, but ended up not using it. Having a relational database that I can query would have been ideal but I don't know anything about Python. Rotogrinders has a lot of really good info so ended up just going with that. Unfortunately some of their stats aren't in a great format to bring into excel using web queries. I bring in half of the data manually each week which kinda sucks. I built some VBA macros to compile and analysis everything from there. So it doesn't take too long to do some decent analysis. Still wish I could get a more complex model going that would be made easier with a database and SQL/SAS scripts.
 
Last edited:
NBA was wayyyyyy easier for me than NFL last year. You just have to pay close attention to if/when a player is sitting due to a minor injury or rest and plug his replacement in. I'm sure there are some crazy algorithms for it too but it's much easier to get an edge.
 
Gurley only 35% owned on fd

People going big in Martavis Bryant... 32%
 
You must log in or register to reply here.
Back
Top