NBA FORUM

Comments

  • lordwu84

    The ResultsDB tool has been a great help in doing research and finding leaks in my though process. I noticed something curious from the main 10/25 Fall Festival tournament. Otto Porter vs. Lakers was the highest owned player by the public on the slate at 45%. Across any projection site I looked at, Porter had one of the highest value and projected ceilings, and had been crushing it in the past couple of games, and was in a tremendous spot vs. Lakers. Yet Youdacao, ChipotleAddict, and Awesomo all completely faded Porter: 0% ownership.

    It seems to me they are using similar models/thought process here and I can’t figure what they saw to fade Porter. If there’s something so basic I’m missing here, there is a massive leak in my game, albeit I’m a low stakes grinder. It can’t be as simple as fading public chalk, as why didn’t that apply to all of the other high owned players.

    Any thoughts on what led all of them to fade Porter?

  • hoax

    @superstars92 said...

    How do you predict ownership? I know general ways like reading the forum and articles, but is there an exact science? I would say you are lucky if you are within 20% of the ownership . The best I can usually do is tell which players are chalky but I can never get exactly the right ownership. For example, tonight Steph will be massive chalk, but will it be 50 or 30%? No idea but that makes a difference. All I know is he’ll be massive chalk, and I’m sure everyone knows this.

    I believe leverage ownership only applies to choosing the sub 10-15% plays as like 40% plays. It’s hard to do a leverage ownership play in reverse, like with Steph tonight. A leverage ownership play would be like choosing Aaron Gordon over Draymond Green for example. However, completely fading Steph seems like not the point of this play (like completely fading Otto Porter Jr. in the OP’s example).

    That is the question! How do you predict ownership? Its near impossible since it is different for every contest you enter. The best success I have had with it is lumping predictions into ‘buckets’. You dont have to predict the exact ownership but rather, how many features does this player have that have lead to high ownership in the past?

    Basically every time one of these features matches with the same situation from the past you can move that player up a bucket. The more this player moves up in buckets the higher owned he is projected to be. You can assign each of these buckets a range of values but they are all arbitrary really. This seems to be a pretty good way to go about ownership from what I’ve been doing.

    So for example in my highest ownership ‘bucket’ for tonight I have 3 players, Irving,Curry, and Drummond. Now when I look at this I start with deciding how much of these players I want in my pool, assuming they will all be around 30% owned. Tonight I want to be overweight on Drummond and Curry. Even if I dont like Irving all that much, considering I have him in a high ownership bucket, I almost never fade those players entirely, but instead will be underweight to him only having him in say 15% of lineups. That make sense?

  • superstars92

    @hoax said...

    That is the question! How do you predict ownership? Its near impossible since it is different for every contest you enter. The best success I have had with it is lumping predictions into ‘buckets’. You dont have to predict the exact ownership but rather, how many features does this player have that have lead to high ownership in the past?

    Basically every time one of these features matches with the same situation from the past you can move that player up a bucket. The more this player moves up in buckets the higher owned he is projected to be. You can assign each of these buckets a range of values but they are all arbitrary really. This seems to be a pretty good way to go about ownership from what I’ve been doing.

    So for example in my highest ownership ‘bucket’ for tonight I have 3 players, Irving,Curry, and Drummond. Now when I look at this I start with deciding how much of these players I want in my pool, assuming they will all be around 30% owned. Tonight I want to be overweight on Drummond and Curry. Even if I dont like Irving all that much, considering I have him in a high ownership bucket, I almost never fade those players entirely, but instead will be underweight to him only having him in say 15% of lineups. That make sense?

    So yea the 15% of lineups actually makes me think that the players mentioned here just didn’t value Porter that high if he was indeed 0% owned overall over 150 lineups. Generally you would underweight a leverage ownership play not go 0% unless he also was also valued poorly (which definitely can be the case both situations occur simultaneously). Also, as I mentioned before, usually a leverage ownership play works in reverse, where you play a low owned player at like 5x the projected ownership, not the way Porter is described here.

    Yea I would have Steph and Kyrie as pretty highly owned tonight. Oh one thing to keep in mind is players of the same position are constrained by total ownership. For example, if like 8 PGs are all popular, it’s not possible to play more than 3 of them, so they naturally will constrain the overall ownership of players. Could be another feature in your model but I assume that’s more relevant in LP.

  • NoLimits0

    I really think you guys are just all overthinking this. Sometimes you don’t have a player becuase you don’t have him highly projected for that night. That’s usually the case. This ownership stuff is high level smoke. Good on paper no one really practically implements it becuase like you guys even acknowledge how do you even guess the ownership?

  • bhdevault

    • Lead Moderator

    • Blogger of the Month

    @NoLimits0 said...

    I really think you guys are just all overthinking this. Sometimes you don’t have a player becuase you don’t have him highly projected for that night. That’s usually the case. This ownership stuff is high level smoke. Good on paper no one really practically implements it becuase like you guys even acknowledge how do you even guess the ownership?

    I 100% use ownership projections when mass entering GPP. Our own team here, led by Gimino, does pretty darn well in the projections. I personally know many more that take ownership % into account as well.

  • billholler

    @NoLimits0 said...

    I really think you guys are just all overthinking this. Sometimes you don’t have a player becuase you don’t have him highly projected for that night. That’s usually the case. This ownership stuff is high level smoke. Good on paper no one really practically implements it becuase like you guys even acknowledge how do you even guess the ownership?

    At this point I’m going to assume you are new to DFS. No way you are an experienced player and think projected ownership is irrelevant when making lineups. Like bhdevault said, there are quite a few respectable sites including RG that give somewhat accurate pOWN%. Even when they are off, they at least give you a pretty good idea who the chalk will be. Then it is up to you to decide whether or not you should fade the chalk.

  • NoLimits0

    Oh so some of you guys used ownership%. Still doesn’t change my argument that the easiest explanation for why they had 0% Porter was that they just didn’t project him high.

    Like I said you guys are making it look to be like it was an ownership play that caused them not to choose Porter when the simple explanation was he wasn’t high on their models.

    Model prediction still takes massive precedent over ownership %. I’m almost sure in this case it was just model prediction that caused them not to choose him.

    Otherwise explain to me why youdacao never chooses Porter?

    You guys are making this to be a high level type of thinking when in fact the answer is quite simple. They just didn’t value Porter high and they were right.

    Again I stand by what I said. Most pros don’t really account for ownership stuff. They mostly choose base on projections. Think about it. If they chose based on ownership wouldn’t the predicted ownership always be wrong then by a lot?

  • billholler

    You are absolutely correct sir/ma’am. We have no idea what we are talking about. We will listen to you more now. Good luck and happy grinding.

  • NoLimits0

    @billholler said...

    Multiple people, including a few successful DFSers, have told you that you are wrong but you refuse to admit it. If you want to improve as a DFS player, you might want to accept your failings and listen to others.

    P.S. Yes they faded Porter because he was chalk. In regards to Mitchell and Jokic, they were both in great spots yesterday especially in reference to their values. I never said to ALWAYS fade the chalk. Sometimes it’s just more obvious.

    Sure whatever you say. Are you friends with youdacao and awesomo or something? Did you ask them to confirm why they faded? I think it’s actually quite obvious they faded because they didn’t rate Porter that high. You know the very simple explanation of why you don’t choose someone. And what successful DFSers are you referring to?

  • superstars92

    Uhh I mean it is reasonable they didn’t choose Porter because they didn’t project him high. I can see that happening. Your projections usually triumph any things you want to do with ownership %s, so I can see that being the case.

    I mean I looked through some of these pros tonight. Almost every single one had massive Alex Len exposure, but he was a legit person you could pivot off of (way better reasons on him then Porter, two are Gibson is 4.5k and Randle is 4.9k at the same positions but there are many more than the Porter example). Just FYI I didn’t, but I can see how you can since he’s 4.8k which is not that cheap. None of them did, so I don’t think Porter was necessarily a pivot play in that case if none did it for Len with a better reason to do so. Now, if I saw that a lot of pros did do the Len -> Randle pivot, I would start believing the Porter play was a pivot play rather than a projection play because the Len -> Randle pivot was better than any pivot you could do with Porter that night.

  • NoLimits0

    I wish a pro would just post on here to clarify. I just don’t understand why someone can’t choose someone simply because he had him projected low. Why does it have to automatically be becuase it’s a high level thinking type of play to avoid the chalk?

    If all pros did in fact think like that, how come the chalk players are EVEN CHALKIER in higher stake GPPs where there are more pros? Shouldn’t they get less chalkier if pros thought about ownership % the way people on here suggest they do? Chalk plays get progressively higher owned from the 3/4 dollar to the 8 dollar to the 33 dollar to the 333 dollar. Shows me it’s just a projection thing and not a avoid the chalk advanced game theory type of play. Most pros don’t care about ownership. Sure some do. Most don’t. Most want to optimize the best possible lineups. Hence why chalk plays get chalkier as you move up the stakes because chalk plays are chalk for a reason (good value) and there are a higher % of pros as stakes increase.

    I back up my statements with some data and facts. Billholler just is all heresay and a theorist. He even admits he doesn’t read my posts and just jumps to conclusions.

  • Stewburtx8

    • 2012 FanDuel WFBC Finalist

    @NoLimits0 said...

    I just don’t understand why someone can’t choose someone simply because he had him projected low. Why does it have to automatically be becuase it’s a high level thinking type of play to avoid the chalk?

    They certainly can fade someone because they had them projected too low. Maybe that was the case with Otto Porter. You are focusing too much on this one instance, Otto Porter, and not the overall point. To even remotely suggest that high volume pros who often max enter GPP’s do not take ownership into account when building their lineups is flat out wrong. I am sure some weigh it more heavily than others but it absolutely has to be a factor in successful long term GPP play.

  • superstars92

    I don’t know the answer to your questions but I know Otto Porter is playing the Lakers at home at 6.7k. I bet he’s not owned by many of these pros you guys talk about.

  • NoLimits0

    @Stewburtx8 said...

    They certainly can fade someone because they had them projected too low. Maybe that was the case with Otto Porter. You are focusing too much on this one instance, Otto Porter, and not the overall point. To even remotely suggest that high volume pros who often max enter GPP’s do not take ownership into account when building their lineups is flat out wrong. I am sure some weigh it more heavily than others but it absolutely has to be a factor in successful long term GPP play.

    I agree some pros would do this but I stand by my point that many high level pros do not take ownership into account.

    Again if they did why are the chalk plays even chalkier in the higher stake GPPs, where there are more pros. They are if they went by projections and less on ownership.

  • bhdevault

    • Lead Moderator

    • Blogger of the Month

    Many pro’s use their cash lineup (or a few pivots from it) in high stake GPP’s. The less people there are in a contest, the less contrarian you have to be.

    With that said, it seems at this point, nothing will convince you, and that is totally fine. Everyone is entitled to their own opinion. This thread has just become an attempt to convince just one person, defeating the purpose of it’s intention.

  • NoLimits0

    @bhdevault said...

    Many pro’s use their cash lineup (or a few pivots from it) in high stake GPP’s. The less people there are in a contest, the less contrarian you have to be.

    With that said, it seems at this point, nothing will convince you, and that is totally fine. Everyone is entitled to their own opinion. This thread has just become an attempt to convince just one person, defeating the purpose of it’s intention.

    Ok I was talking about 8 dollar vs 33 dollar which both have 150 max and a ton of people.

    The fact you say that pros just use their cash lineup and pivots in the 300 dollar contest (which I assume you mean) actually shows me they don’t care about ownership%. Just because there are fewer people it doesn’t mean you have to be less contrarian. That’s a huge mathematical flaw if people actually do think that way. In the extreme case, if you think LeBron will be 100% owned in a 10000 person or a 100 person contest, having 100 lineups without LeBron (which is what you are saying to do when you say 0% Otto Porter) in the first contest or having 1 lineup without LeBron in the second contest produces the exact same results. I’m assuming most pros understand this math so the fact that they don’t do that in higher stakes shows me they only care about optimizing.

    If you take a look at youdacao, ALL of his lineups whether in the 8, 33, or 300 dollars are slight pivots from his cash lineup. It doesn’t matter whether it’s 8, 33, or 300 he doesn’t care about ownership.

  • Razzle11

    @NoLimits0 said...

    Ok I was talking about 8 dollar vs 33 dollar which both have 150 max and a ton of people.

    The fact you say that pros just use their cash lineup and pivots in the 300 dollar contest (which I assume you mean) actually shows me they don’t care about ownership%. Just because there are fewer people it doesn’t mean you have to be less contrarian. That’s a huge mathematical flaw if people actually do think that way. In the extreme case, if you think LeBron will be 100% owned in a 10000 person or a 100 person contest, having 100 lineups without LeBron (which is what you are saying to do when you say 0% Otto Porter) in the first contest or having 1 lineup without LeBron in the second contest produces the exact same results. I’m assuming most pros understand this math so the fact that they don’t do that in higher stakes shows me they only care about optimizing.

    If you take a look at youdacao, ALL of his lineups whether in the 8, 33, or 300 dollars are slight pivots from his cash lineup. It doesn’t matter whether it’s 8, 33, or 300 he doesn’t care about ownership.

    I am just quoting your last post as I am too lazy. But you keep referring to using their optimizer to spit out 150 lineups and that they dont care about ownership%. Realize that nearly every optimizer I have seen publicly has a spot where you can dictate your ownership out of those 150 lineups. Just like with the RG optimizer, you can select your preferred OWN% for a given player prior to optimizing your X amount of lineups.

    Successful GPP players all take into consideration, some nights more so than others. Some nights they have to eat the chalk because pivots just arent there.

  • Stewburtx8

    • 2012 FanDuel WFBC Finalist

    @NoLimits0 said...

    Just because there are fewer people it doesn’t mean you have to be less contrarian. That’s a huge mathematical flaw if people actually do think that way.

    This is completely and utterly false. The type of lineup construction you generally need to win a 50-100 person contest is so much different than a 10,000+ person contest.

  • superstars92

    @Stewburtx8 said...

    This is completely and utterly false. The type of lineup construction you generally need to win a 50-100 person contest is so much different than a 10,000+ person contest.

    Actually I don’t agree with this. If you submit 150 lineups into a contest of 15000 people, it’s the same as submitting 1 lineup into a contest of 100 people if you are truly gong to do the 0% fade of a chalk player.

    I do agree you are supposed to consider ownership percentages though, although it’s not that simple. Usually it’s a tiebreaking factor. It should never be used as the first factor in choosing two players. So if you think tonight LeBron = AD (not saying he will, but let’s just assume you thought that), and you thought AD was gonna be 10% and LeBron 40%, you would go with AD. But if you thought LeBron > AD, I would go 100% LeBron (of course Harden messes that up, but just assume AD and LeBron were the only two studs available). With more studs avaliable and since the studs play different positions, that’s why you end up not having LeBron 100% even if you thought he was the best possible player. Hopefully my example makes sense, but somehow I think it’s confusing.

  • KindGuy

    Lmao how is this still going on

  • superstars92

    This thread actually got me curious about something. I hope someone in one of the other threads can help since they are better at math than me and it’s actually a mathematical equation at the end of the day.

  • Stewburtx8

    • 2012 FanDuel WFBC Finalist

    @superstars92 said...

    Actually I don’t agree with this. If you submit 150 lineups into a contest of 15000 people, it’s the same as submitting 1 lineup into a contest of 100 people if you are truly gong to do the 0% fade of a chalk player.

    Mathematically, yes it’s the same in terms of % of teams you have relative to the total field. But the roster construction for each contest is still very different as the score needed to win the 100 person contest is almost always going to be much lower than the score needed to win the 15k person contest.

    I generally agree with the rest of what you said. Ownership percentage is not the end all be all by any means. It’s just another factor that should always be considered. I am not saying you should ever blindly fade a player because they will be highly owned or blindly play someone because they are going to be less than 5% owned. There is obviously a lot more to it than that which involves weighing personal projections versus projected ownership percentages to determine your own ownership percentages across 150 lineups or whatever you play. But this is coming from someone who builds 15-20 lineups at most (usually in the 5-10 range) and does it all by hand.

  • billholler

    @elementasrat said...

    Lmao how is this still going on

    Any time one dude thinks he knows everything and is smarter than many successful DFSers, the thread never dies.

  • billholler

    @superstars92 said...

    This thread actually got me curious about something. I hope someone in one of the other threads can help since they are better at math than me and it’s actually a mathematical equation at the end of the day.

    Nothing to do with this thread but nice hit Tuesday night.

  • NoLimits0

    Look I’m not saying I’m smarter than other people. I just am saying projections usually is the number 1 factor and generally even pros largely ignore ownership levels. If you lie someone a lot who cares that he will be 50% oened.

  • billholler

    @NoLimits0 said...

    generally even pros largely ignore ownership levels.

    NO. JUST NO. STOP!!! UGGGHHHH!!!

  • 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).