Game Theory: Learning How To Win At Daily Fantasy NBA



ladies and gentlemen welcome to the inaugural roto grinders game theory show I am Fast Eddie fear with me is Bret's killaby how are you doing Bret good man and how are you today doing well man glad to be here so just to start us off about half the questions in chat so far have been what the hell is this a couple words about the show there's a variously attributed adage that says you can give a man a fish and feed them for a day or you can teach a man to fish and feed them for a lifetime and to that end whenever I host daily shows I always tried to more so than just saying I preferred this player to that player I try to explain why and give them and why so that you're not just giving out picks but you're also giving out some of the reasons why you make those picks so the overall goal of the show I guess more so than like a heuristic discussion of game theory and mathematics is to make those of you in attendance teacher to fish as opposed to bean fish in the ocean so game theory is a bit of a buzz term it's like gamification which is something else I worked with for a long time that gets overused basically what game theory is is it's the mathematical modeling of any situation in there's a lot of economic applications or biology or just everyday life into a game that has players and moves in strategies which you can then use to short possible outcomes and then hopefully determine an ideal one so some simple examples are should I run this red light the players in that game are you and the hypothetical police that may not even be there and or should I attempt to get my wife into the sack tonight in you know tonight I'd really like tonight but if I wait until Friday that's when I get paid she likes that and that's also usually when she buys the wine from the grocery store so what's the right way to put it or when is the right time to ask your boss for a raise on Mondays he tends to be grumpy on Fridays he's happier he's going home for the weekend and then so all those are very simple game theory applications almost binary in their nature but more complicated you could take things like that and combine them to create with more variables things become more complex but to really chart away through all those scenarios so for example using that above example you could say that maybe I will wait till Friday ask my boss for the raise which I think I have the best shot at on Friday and then my wife will be super super interested plus she's got the wine and now that I know something good is waiting for me when I get home that would be a really bad time to run the light and get pulled over and get the ticket so basically that's that's what we're looking at and trying to do so I love the guy in shadow just yellow who do we play that's so that's what we're looking at when you're picking players many many more points of data and I'll move on to this point or real quick so Brett any any follow up no I mean I actually wrote my senior thesis in college about game theory so economics major at Madison so this is right up my alley this is actually gonna be my 11th year being a professional poker player so there's plenty of game theory in any anything I mean each decision in life is kind of like a game theory decision if you want to break it down like that so yep yeah I mean just I think it's I think this is probably the most important aspect of profit excuse me a profitability in DFS it's a lot less about knowing the sport and a lot more about knowing the theory behind who to play when to play them why to play them so I think that it's really cool to have I mean we're going to talk about some slate specific stuff within this show but it's cool to have kind of a show like this where we can really touch on things I know that that it was this sort of information to help me a lot early on when I first got started in DFS and and you know I think that this goes a long ways and especially for me being yeah Brett today was who's done really well in NBA and Jeff Collins is gonna be our normal third member on this show and so I know Eddy has a really dip in depth understanding of game theory and Jeff and Brett are really good at NBA and most of you who read my stuff know that I'm still kind of in a transverse transmission transition phase with with NBA so so yeah it's really cool to kind of get all these different perspectives and to kind of learn what what goes into the thinking behind as opposed to just which players are on the winning lineups right so sticking with you for a minute JM one of the things we were gonna talk about today and hopefully I don't have the production up but hopefully we're moving down the menu now is we wanted to talk a bit about the transition from NFL to NBA the difference between those sports the difference in the other players in the game for example if you're speaking about in a game theory perspective you have probably in general a much better educated populace in NBA than you do an NFL that's not to say that there aren't a fair number of fish but NFL is a huge number of fish when you spend couple hundred million dollars on advertising you get people coming out of the woodwork who have no idea what they're doing so I was hoping you would share with us a bit of your own experience in transitioning from NFL to NBA yes so I think each week we'll kind of try to touch on a different topic that specifically relates to trends from NFL to NBA and and we use that label as kind of a catch-all but this is not just people who are coming from NFL to NBA but it's also people who feel they can improve an NBA or people who are essentially nonprofit organizations in NBA and I'm just giving charity to other players while making no money yourself so basically if you feel like there's room for improvement in your NBA play this section is about so I have kind of two topics we can talk about today and we can gauge timewise but yeah I'm gonna present one that we can all talk about it and then if we have time for a second one we'll get into that one so this first one ties directly into my NFL NBA story and that's that I started in DFS in 2014 in baseball and I was fortunate enough to have some some nice wins early on before I really knew the game theory I actually won a qualifier on draft Street before I even discovered roto grinders so like I didn't really know my stuff so much as I knew the sport of baseball really well and and that you know translated to enough good lineups and then as I started really studying the game theory and doing well I had a great MLB season and then I probably wasted a year's worth of my life if you take all the hours and put them together I probably wasted like eight years worth of my life studying the NFL to no end throughout my lifetime outside of like season-long fantasy sports and just enjoying the sport so then I was able to transition to NFL and put that knowledge to use and had a really nice NFL season as well and got two NBA and I had a couple losing weekends to close out NFL wasn't planning to play NBA and kind of got some emails from draft kings trying to entice me to play some NBA and I was like yeah I'll check it out and so I had one good night and it was like yeah this is it I lost some money last couple weekends of NFL I'm gonna win that money back so I jumped into NDA hit playing it like the same stakes I've been playing MLB and NFL and by the time NBA ended most of my MLB and NFL winnings were gone and I stuck spending all of 2015 MLB building back up to have the bankroll I wanted for NFL's bill to attack NFL the way I wanted so now that I'm back basically above where I was at the end of 2014 I'm recognizing that I've reached a point where I'm a break-even NBA player but that's not good enough to be putting in large chunks of money hmm and so I think that one of the most important things as you transition is it's you know the it's an old Booker term of identifying your leaks and plugging your leaks it comes into play here that if you know we all get excited about playing every slate and playing every sport and playing every night and if if you see a slate that you can't figure out an NBA or if you've recognized the NBA's not your best sport yet like I'm playing I was playing you know five to six K a weekend and NFL I'm playing like between sixty and a hundred dollars a night reason being I've identified NBA as something that was previously leaked and is not yet profitable and I'm working really hard to learn it and once I get to a point where it's profitable to really start ramping up then so I think it's important to identify NBA as a leak if that's the case and then just work on learning it ya know two things I would say about that one without getting too heavily into it because you can do an entire bankroll management show but yet just the basic concept of varying the amount of investment you have based on how they do feel about the slate or the support is obviously a key contest concept and then secondarily I would say that your attitude towards improvement is extremely healthy I've spent a lifetime designing and playing games and there are in all games or contests people who don't really they don't understand it but they have capped their ceiling as a player because they are not looking at themselves in an introspective way trying to improve so for example if you're playing a card game if you're playing on Magic the Gathering or something and you're constantly going oh I would have I would have won that game but I got someone lucky does this happen and if you're always looking for excuses or oh that guy got hurt too I didn't win it DFS last night as opposed to constantly looking at what you're doing and trying to get a little bit better you're gonna be capped as a player so if you're not improving you can look at your past and presume your future is gonna be extremely similar so I think it's I think it's it's really healthy to you know whether you're not quite profitable or breaking even right now or pulling back you know 2% a month on your investment there's no reason you shouldn't be trying to make that 3 and 4 by really looking over every aspect of what you do in trying to improve on it a little bit and and Brett I know obviously you play a good bit of volume you've spoken about your poker experience and obviously I've had a lot of success at DFS so I would think you have something valuable to end yeah I mean just like with anything in life I think Jim is right he's got a lower Laura's volume until he feels comfortable that he's got this board down and you know something that I would suggest people doing I don't know how popular this is is um but don't even watch the games one night just so you have your results and then research that slate after all the games have happened so whether that be in the morning or at night and then you can still projective what you think is going to happen because you have no knowledge of what just happened and then instantly look at the box book and see how close you are with that line up it's it's an interesting way to go about it but that's kind of how I did it I tracked who I thought was the best in the NBA I definitely took my licks last year but I also feel like that's the price of getting better is you know it's kind of like a feed-in poker it's very similar trying to learn a new game I you want to play at stakes that matter or that you feel like you're you're learning somewhat top-level strategy obviously in DFS we don't need to do that it takes one click to get into $1.00 leak it takes one click to get in a you know $1,000 leak so it makes sense to start out there's even even to the today I still don't have the best cash game strategy versus GPP strategy when it comes to MBA because to me when I do look at game logs sure there's those guys that what they can get hot like the Klay Thompson's you know that's the perfect GPP guy right but you know I'm always tracking minutes and to me minutes means usage and then usage means fantasy points so there are certain guys that I have dubbed as strictly GPP plays but you know like last night like Tyler Johnson kind of he got he still got minutes even though he ended up with three fouls early in that game he still ended up log in 34 minutes he did not produce though in those 34 minutes so should be interesting in today's like where his ownership level will be yeah definitely I think that represents an opportunity because people I mean basic game theory people will get really hung up on last night and people get hung up on the money they've just lost look at his several games before that his production was around a fantasy point a minute or at least somewhat close to that and again if he still expected me getting 30 plus minutes at his price he still should look like a good point something on that that Brett off of what you said was I mean what you essentially said at one point was basically you should either be playing you should be playing for a reason so if you're playing it low stakes you should be playing this wasn't specifically what you said but what I took from it that I think is an interesting thing for people to think about is you should be playing each slate for a reason so if you're playing it you know states where you're playing to profit that's why you're playing but if you're playing at lower stakes it can't just be like oh well I'm okay if I don't win this money because it's not that much it has to still be like I'm playing today in order to learn this sport more and in order to improve more I think that's an important aspect is that each day that you play as an opportunity to learn the sport more fully like the place where I came from for NBA is like I don't watch SportsCenter I don't watch NBA games so I came into last NBA Slyke last January when football and I came in blind like learning the players learning the strategy learning the roster construction learning everything so I mean you know even right now I'm still like I don't know the NB a half as well as as I know Major League Baseball or NFL as far as all the players and stuff but I think you it's important to take each slate that you play as an opportunity to learn something and to that end one of the one of the best ways to learn something is just track high-dollar tournaments and you don't have to be in them to I don't know if and who has this put on draftkings there's a tab on this site on the desktop site that has on the top right corner it says live and you click on that it's all live contest and so you can go to the all star which is the 1k tournament in an NBA and and you know once you start paying attention it's not which lineups are winning its who are the good players cuz the best players don't win every night but once you start identifying who the good players are and you starting their lineups each night why did they make these decisions that can help a lot and people will ask me during football people would want to play me head-to-head for $5.00 or something to to see what my lineup was like and sometimes I tell I'm just going the the 1k tournament you can see it live and you can see you know everybody's lineups right there so it's you don't have to spend that money to see everybody's lineups right and that's probably the best way to get to know us for it is to not only break down one you know one lineup but get into a tournament where you can deconstruct everybody's lineup understand what they're thinking but also understand that tournament strategy and cash game strategy is totally different cash game you're looking extremely just for value whereas tournament you're looking upside low ownership and things of that nature so I definitely think it's about consuming as much as you can as far as information and data goes whether or not it's reading articles or listening to us or looking at other people's tournaments the more sources of data you come from the better because these little form your own opinions like you should almost never accept as fact almost anything that is said to you unless it's a truly unimpeachable source and I always like it when someone you know turns to me and goes you know Eddie I don't like your analysis today I'm good that means you've come up with a theory and you think your theory is better than mine and you're considering it like that's absolutely the way it should work people in chat are saying that hasn't been a ton of specific game theory applications let me throw one out there and again no one prep for this so take your time and sorry for the the pop quiz but let's say for example last night we had four games in MBA and in a little bit we're gonna review I think one of your personal lineups Brett and yes how that went and how it was put together and tonight we have 11 games now for the purpose of simplifying this example let's assume that everyone falls into chalk or not chalk so everyone is either and by chalk again I guess maybe everyone isn't familiar with that term that means plays we expect to be widely played because they do have an expected high output so Tyler Johnson last night for example someone who is going to be played by a lot of people you're building a line up on let's say FanDuel you need nine players to me it has always seemed like when you're building a GPP lineup when you're trying to build a large GPP there's there's a huge value to being contrarian and taking players who are less owned but the number of players to take who are off the beaten path is something that varies to me somewhat proportionally with the number of games on the slate so for example if you take nine players all of whom are less than 5% own I don't think you can ever win now will almost never work out because those are all long shots they will almost never all come in so are you looking at seven truck and two long shots are you looking at six chalk and three long shots and again this is a vast oversimplification because 0.1 percent and five percent would both be long shots in this estimation but those are far far different things but for this simplistic thing it seemed to me that pulling in a GPP on a four-game slate involves being much more chalky and you can get there with one or two contrarian picks whereas pulling in a large DPP on say an 11 game slate like tonight involves being further afield in more picks I'm curious if you have an opinion that we'll start with you bread uh yes so I mean I guess what I'm gonna I'm gonna kind of just be somewhat site-specific here now each site is definitely different and I know I know I think I've heard you talk about position scarcity when you have a guy who is D Chuck like Tyler Johnson last night I think it's okay to fade a guy like that even though we know he's getting minutes if there's another guy that's gonna get similar amount of minutes so my I ended up fitting him last night and going with Emmanuel mudiay who has been getting questionable minutes but he was in a very pace up game against OKC and so my turn of strategy was the stack that game you know so I just took moody n Gary Harris and then came back with both Russell Westbrook and Durant so just hoping that game would shoot out now I I knew that Tyler Johnson was going up against a buffs team which is a very slow pace team so there's a I guess I guess this is when I start just doing an almost like like thinking of like what could go wrong and what could go right and then it's more like uh like a hypotheses of like okay so this happens what what needs to happen for this to be a good GBP play so for me it was okay okay see I'm gonna go strong on OKC so that's Westbrook and Durant I ended up just paying for those two guys now for those guys I think to go off that means Denver needs to keep it a close game so that means Denver needs to produce how is Denver gonna produce let's and then you kind of dive into that I liked moody eight so I was like okay I know people are gonna be on Tyler Johnson I try and listen to enough pods throughout the day while I'm working out and whatnot to get an idea of what the industry is and then and then go off of that so I heard all Tyler Johnson and every single podcast was like one of the very first names that people talked about so yeah I mean if I know he's going to be somewhere especially in a four-game slate I think he's somewhere around eighty percent owned in most of my tournaments so if I know he's gonna be that high Lyon and I know moody is gonna get similar minutes in a big time pace up game which guards like that's one thing you want to look at like on that base is just pace pace is really big even more so than DVP I would look at pace up games because especially studs like Russell Westbrook it doesn't matter how strong the defense they play if they play a pace up game they're so good they can go off whenever it's also a brief analogy the better a player is the less relevant a matchup is I know that sounds a little silly and again it's an oversimplification but right I'm much more worried about a value play running into a buzz saw of the DVP matchup or an especially efficient defender than I am Westbrook or Durant running into a tough defender because their guys have proven over again they can do it under tough circumstances one more thing I want that real quick and then I'll throw it back to you Jam was I think it's worth reiterating what you said about finding a narrative and again this is this is easier to do on a four-game slay than it is on an 11-game slate it's still doable on 11-game sleeping on a for games like you can really engineer to the okay if this happens this is the team I want to pick there's only four games this is the pace of one if this game a simple example is if this game goes to overtime well now all of a sudden most of your players are picking up an extra 15-20 percent value or in this case if the the the Denver no KC game stays closed and everybody plays four quarters as opposed to a three which was a real possibility then these guys make extra value so it's not just about picking players based on percentages it's about saying in the event that these again pure game theory in the event that these things occur I can expect this result and so I'm going to attempt to forecast that and get the best result I can jam yeah I think I think another one of the biggest things to in in game theory when we talk about ownership is a huge part of that and somebody amongst all the criticisms and complaints in chat somebody pointed out that contrarian played is not all long shot which is absolutely true and I think that's one of the things I I try to really hard bond especially in an NFL MLB is a lot of times you know if you wanna if you want to get some good game theory when we get to MLB maybe jeeps tournament lineups you know because if you look at his his pictures each night it's usually guys who if you look at all the numbers they're one of the top two or three pitching options on the board but they're also loaned because a lot of times people get set on a certain play and so you know a couple things I wanted to talk about here was I didn't play last night I mean I'm in Santa Fe and I was down out of cell phone range on a brutal 10-hour or 10 mile hike yesterday but NBA was was setting up to play but I played two days ago and I used Donald Sloan 4500 on draftkings for Brooklyn he probably would have been thirty thirty-five forty percent owned until Tyreke Evans was out and Norris Cole was 4100 filling in so Donald Sloan didn't become a worst play than he'd been earlier in the day um it's just that people have this tendency to whatever's the latest value play to come out everybody feels like they have to gravitate toward that all right I think that's one one thing that's really important to realize in identifying game theory is is if several value plays are available whichever was the latest one to come out especially if one was people knew about it for a full day in advance and then one comes out you know an hour before the game start everybody's going to switch to that one that just came out and so that I think Norris Cole was 40% home and Sloan was 10% owned something like that and then Cole was a great play to you know I thought about switch he had myself but the point is if you think about it from a game theory perspective you know that Sloan is also a good play and will be lower owned and then that same day you know just in in looking at things and I'm trying to block out the noise I identified Ilya sova as my favorite power forward low cost power forward I like talking about as much as who everybody else was on 900 more and I like Brook Lopez a lot of center and had the money to use him nobody would have said those were bad plays that day but Ilyas OVA was three percent owned and Brook Lopez was two percent owned so the fact they both had good games is irrelevant compared to the fact that they were good plays and were both under five percent owned so I think that a lot of times the strategy side of things is is identifying who are the guys that everybody's overlooking there aren't necessarily bad plays that's all a high floor to go with their high ceiling nobody's on them for whatever reason we're getting a bit behind on schedule one thing I want to cover real quick because I had several people ask it to me today and people are still saying things in chant that indicated still to some degree controversial is the value of being contrary a lot of people with whom I discuss DFS especially those they were or relative neophytes just simply don't get it I'll be like yeah and I'd like this guy cuz I think he's gonna be a lot of work lower owned and it gives me an advantage versus the field in this way they don't quite get it they're like why don't you just pick the best guys you just pick the best guys you want to score the most points so if you pick you know these low on guys and get 337 points then I pick all the well own guys and I get 338 points I beat you I don't get it Brett why don't you give me the 30 second reason I sound stupid yeah well I think Jam to win brought up a really good player in beep I'm a Jeep the reason why he wins probably more often than the next guy in tournaments is because he is taking a very highly contrarian so I guess if we can't we can't discuss Major League Baseball and NBA in the same same sense because it doesn't matter how great you are in baseball if you're Mike Trout you can go over five of five strike yutz right so I that's not gonna happen with Russell Westbrook Westbrook has a floor and has a ceiling but in MBA a guy like Anthony Davis or a boogie cousins or legitimately is zero Anthony Davis can hurt himself within the first three minutes which we have seen this year and boogie cousins can get a double technical and I've owned both of them on multiple occasions and have lost every single you know cent that I invested in on them so you just got to know I guess I'm kind of getting lost in what I was talking about but here just here's here's something that I want to talk to about Jam – when said that you know later news a lot of people scramble to the late news they then put in backups now I know as of last year when I first started getting into MBA it was if that news was out for a full day we knew that there was going to be plenty of ownership in that specific player a great example was Tyler Johnson yesterday the older the news is I think the higher the the more popular the play will be now in today's DFS it might be a little bit different with all these news tracking sites you're you're able to scramble for that news a little bit more well let's just say the rotor grinders community is still not a drop in the bucket but let's just say you know thirty to forty percent of the action I would say the other other sixty percent still makes their lineups and they're not waiting to the very last second like we are waiting on news they're there they've already bought the Tyler Johnson's Tyler Johnson stuck you know if something happens at 5:30 these guys aren't going in in their lineups and changing changing out to different guys so as much as we'd like to think that we are kind of like our own opponents that's not that's not the case I think I think a lot more lineups are more stagnant or more set in stone a good 90 minutes before a line of lock and you really think so I mean to me late news is gold for me and I used to be the guy when I dipped my toes into em being last year that was like oh who is the backup now I'm more like Oh what does this mean for usage like with the other players so I mean when you have a sport where there's only ten individuals on on the playing field you take one oh it changes the whole dynamics of the game so you you really have to analyze matchups also understand that season-long stats don't mean much when a big name is out because that changes defensive that changes rotations that changes everything so MBA is so different but if we're talking from a game theory perspective like a guy like beep on a Jeep like he goes so contrarian in baseball and it the thing is you're right it's not gonna hit very often he's gonna be last place a lot but then when it does hit boom top of the leaderboard and it's the perfect tournament strategy because who cares about you know last place to even you know a twentieth percentile you make your money in the top 1% yeah just one real quick thing is just you know you ask what you know the whole idea people saying no you want the team that scores the most points that would all be good well if we could predict with 100% accuracy what everybody would score that's the point of you know taking some contrarian plays is if you take the 5% guy and he's the one who has 50 points well you know the 40% own guy the same price to scoring 20 points that's regain the big edge and so if we could say for sure like we know for a fact this guy he'll score 50 points well then yeah obviously everybody's on but it's not you can't predict with hundred percent accuracy and so that's why when two players are close you take the guy whose projection might be a little bit lower we'll also be far lower owned okay so what do you say we hold on to your second topic which I don't think you even started no no I didn't I think we've kind of touched into strategy two haven't we pretty heavily yeah we sort of got into there so something wanted to discuss let me uh hopefully I still have that link so we'll see if that works no stars out that was I left the HTTP in or put a string of obscenities in chat and start once more boom okay I put a link in chat and someone else copy it a couple times I'll throw it in there one more time just because I can but anyone who wants to follow along with that I'm going to talk about for a couple minutes here can I can click on that so um again I came to DFS I've been playing season-long fantasy sports for God nearly 20 years I'm old as the hills but III think I inherently looked at it from a game perspective I have been playing games way way more than I should have for my entire life and then I owned the games burn tournament center for a dozen years and then I designed games for a number of companies for about a decade after that so when I came to DFS I had sort of sort of trained myself to look for edges in games I'm a min/max er as I think we all do but what I found was the earliest thing that helped me was taking advantage of playing on multiple sites and could do a couple hours on this topic alone but there's a lot of good reasons to play on multiple sites there there are pitfalls to it as well Oh awesome okay cool yeah all right well I have a strategy thing I wanted to talk about if bread if you don't have anything to fill in with yeah I can probably go right along with what he was saying cuz I know exactly okay yeah you go ahead okay so fast that easier so okay he his article was really good now this is straight about looking at the industry's value i play seven sites during NBA so i totally know where he's going with this article it's to try and find your best value by those guys on those sites right so since i play seven sites there's gonna be times where hey I want Russell Westbrook on 60% of my action right so I know exactly what I have in play now I have to kind of figure out what sites I'm buying them on I want to buy him on the site where he's the third highest price not the top price guy you know so it makes sense to get your money in where you're finding the best value how many sites you play yeah usually one okay no and and that's that's great so the reason why I across seven sites will throw it back to Edward here in a little bit but the reason why I play seven sites is so I can invest more of my bankroll and lower that variance now I if you go from site to site I'm not buying the same guys like my lineups are different because each each site has different pricing like Edward is talking about you so I do want to get my money in good I want to buy the guys where there's best value and I'll let Edward finishes a couple of points to get to that's the first time that's happened to me in about a hundred shows a couple of feet of snow on the ground as you were saying tons of great reasons to play on multiple sites don't know if you touched on there are promotions and deposit bonuses you get to take advantage of all of those across the industry most sites will overlay at one time or another and overlay again when it turn is paying out more money than it's taking in you can take advantage whenever that rears it said but the one thing I wanted to focus on today is site-specific pricing and again he said 7-7 was the number of sites I played on for about for the for the longest time that's what I tried to do and as I was saying accepting when news comes out at six fifty nine and thirty Seconds and three or four of those sites don't have late swap it can get a little bit shaky but let's I'll come over the simple example to elucidate the point a little more clearly let's assume that you're playing on seven sites but all seven sites while they have different prices have the exact same structure as FanDuel so let's say all those sites have a floor price will talk and be a gain of 3,500 and a cap price of 11 to 12,000 but they have different pricing on those sites so on one site LeBron is 9700 on another one he's 9900 on another one he's 10.1 on another he's 10.3 etc etc when I make my lineups on all those sites I'm gonna design at the beginning about how much LeBron I want to have and then I'm gonna buy my shares of LeBron where he's cheapest so I'm gonna play him on the sites where he cost 97 and 99 and 10.1 and I'm gonna pay people I'm gonna play people across all those sites many of whom are paying for him at ten point three and ten point five and ten point seven and over time factored into tons of players and tons of sites and tons of games this represented what I thought for me was my most significant advantage as a player over time so to look at and again I haven't had any time whatsoever to actually disseminate that data that I shared today as a matter of fact I'm having trouble even finding it haha there it is so if you look at the that spreadsheet that I populated today and I haven't even looked at the shade it takes me a few hours to put it together the trickiest thing is actually getting all the names to agree some places your JJ some places your Juan Jose some places your jr. some places you aren't but basically if you're looking at that site I've used basically top seven sites here and for each side is going to show you each guy's salary and their rank and their rank is simply how expensive they are so the number rank one guy is the most expensive guy on that site and so on down you can look at this and see proportionately how expensive a guy is site-to-site so if you're playing on all these sites again you're purchasing him where he's proportionately cheaper now there are other tools rotor grinders has a tool right now I'm trying to get them to upgrade or change to be more like my tool which i think is an improvement they have a percentage of cap and percentage of average and you'll see other tools that do them but the problem with those is they're only referential to a single data point like we need to talk about all guys is in percentage of cab well then the cap is incredibly important like if if russell westbrook costs ten thousand two hundred everybody's percentage of cap is going to be proportionately higher than if he cost ten thousand seven hundred so everybody gets better or worse depending on Russell Westbrook's price even though they're Manship and price didn't actually change at all so what you're looking at here is and I'm just picking a name around let's look at Damian Lillard Damian Lillard is between the 6th and 12th most expensive player on all the sites Damian Lillard is only the 12th most expensive overall player on FanDuel everywhere else so if you're gonna play tonight you're gonna play it fan doulas as many of you do that would be a good place to start buying your shares of the Damian Lillard and in further application I would try to actually play the exact same amount of cash game so I'll just start as cash games as the simplest on each site so if let's say I would play and frequently I would put $500 down on each side if I could in cash games so basically I'm looking at $500 and head-to-heads and 1550s on each of these sites and I'm determining how many shares of Damien Lord do I want if I want one share of Lillard I just want about one seventh exposure I'll buy him insane if I want a second share of Damien Lillard well and I would go back you check and see you know he's ninth most expensive on both draftkings and penisy and so on and so forth and if you look through a couple other nifty things a tool like this can do for you is it can make you aware of value you might not otherwise so for example and again I wish I had more time to look at those of you look at Gordon Hayward is 21st and 23rd most places there's 125th and one 30th at Fantasy aces so you may not like Gordon Hayward tonight but seeing him that far depressed you might think oh that's that that's a guy I should maybe have some exposure to or if you look at who's this with a 47 is that klay Thompson yeah klay Thompson is randomly at my alma mater yes if you'd be 47th most expensive player he's 20th to 22nd most other places so he may not have been on your radar but that's it oh make sure you get him in there you might have otherwise missed that piece of information so this was a tool I used for a number of years and you guys say about it this this would be a great tool obviously when you do when you do have a bankroll on multiple sites I guess coming from my poker background I think what makes me a step ahead of most my opponents is that I'm pretty good with bankroll management so I understand that I can be pretty aggressive and a lot more aggressive than what most people would do just for example or some gym you play one site on NFL – no NFL play I play four or five or six sites oh nice okay um do you do you have like a set percentage of what your DFS bankroll that you would invest per week in NFL yeah I mean I have a general range budgeted and then I'll fluctuate a little bit up and down depending on what I'm seeing in the slate and how many contrarian plays I like okay so I mean I potentially will invest somewhere around 40 to 50 percent of my bankroll at times in in the NFL just because I am making so many different light-up lineups on so many different sites with so many different salaries that I'm actually not taking a big risk you know if you're running all of the top 30 players in some proportion everywhere I mean you may execute a total fade sometimes when you really hate someone but if you have enough cash lineups on enough sides usually if some exposure to everyone so right know game theory extrapolation of conclusion would involve you losing all your money's we've now invent a $10,000 it would be impossible to correct and and just to give you an example poker wise like let's just say I have ten thousand to invest in poker games right I could play 150 hundred No Limit game and put all ten thousand on one table right so that would be just one site or I could play ten five ten games where I might win six hundred here I might lose 300 here I might win eight hundred there so it's it's also very similar that that it's actually a very low variance like I said I'm not these lineups are not similar by any means and then I also create multiple lineups and tournaments so it really lowers variance and I think a lot a lot of people are way too risk-averse if you if you wanted to be a little bit better with bankroll I would just say joining another site is a way to get more action in and also lower Barry's yeah absolutely JM did you have a follow up yeah something on on that most people who are watching know that I'm pretty well known for building basically one lineup per site each weekend especially in NFL MLB where I feel like I know the sports really really well and you know NFL I had three losing weekends all year and I think when you you know there's certain sports when you know them really well you can go more all in on on one group of players obviously you wouldn't want to play 40 or 50 percent of your bankroll if you're doing that but point being you can go more all in on certain players and one thing I've noticed from a game theory perspective is pricing differences from side to side can actually work the other way too as far as ownership changes is so if I like a guy a lot these lower priced on draft kings and I still like the guy the same amount and feel really confident about the way I broke down the play and the chances of him having a good game I know that his higher price on FanDuel is actually going to lower his ownership then that can also be a way to create you know a difference in ownership from a game theory perspective is I'm actually taking a guy who is slightly higher priced on one site which is an interesting way to look at pricing from site to site and with all of this a hugely important thing is actually studying ownership percentages I mean none of this matters if you're not looking at ownership percentage to each night and seeing who's highly owned and figuring out why they're highly owned but if you start identifying okay these certain guys when they're higher priced see a big drop in ownership then that can also be a way to gain an edge by comparing prices side to side yeah there's there's and again you can start firing up the qqq's in chat producer can you give me a sign in chat or auditory whether or not we're four twenty five or four thirty I need to run out of here for twenty five okay we're really gonna go here one of the questions in chat was why would you want to deviate from what you think is the optimal lineup if you are seeking optimal returns and the answer for that is basically your opinion of what will be the optimal lineup will almost always be incorrect even in the most predictable sport which is NBA we could do a whole show on the fact that NBA is a more predictable and has less variance per player than any of the other major sports you don't know what the optimal lineup is is going to be so it it's you're sort of walking to the casino taking all of your roll and putting it on one number on roulette and you're you're slightly advantage to that it's not completely random that you're throwing at it but it's you don't know enough to state what's gonna be the album on so that's that's really the reason usually usually percentage owned in like a high-stake 50/50 whoever the highest owned is is the best value like if you just pull up a hundred person $100 50/50 off a FanDuel on an NFL weekend you will find out who the best values but once you have that information that's a great cash game ah that's a great way to make a cash game line but is it a great way to make a tournament lineup if if that's the most likely lineup out there no because the odds the odds of that lineup being duplicated or or that line up being the not line up with also somebody not having those exact same optimal players with somebody that's an outlier it's just it's so much easier if you know you're picking Lowe own contrarian place to go up the leaderboard and then to have one of those guys be an outlier than to win it and and to win a tournament absolutely yeah and and the other thing to to mention is this is further exacerbated by the fact that tournament payouts and some of the larger tournaments are so top-heavy when you have a three million dollar tournament with a million dollars to first place you want to be even more contrarian because anytime that your local guys patch you over a hundred guys it's more than a proportional increase in their value getting further towards the top is greater returns the further you are away from the field so I'm gonna grab some more of these qqs I'm gonna throw them to you guys while we can can you talk the pros and cons of making one lineup and pros of cons of making several lines Jam you want them yeah for B I mean the reason I play one lineup in in MLB and NFL is because I discovered that further I got away from by optimal lineup trying to cover more bases the like there was to lose money I think if you if you you know really study things you can get to a point where you can narrow down your list of players at each position to a small number of players and then you're not necessarily having to go far outside of that but I think it's also different for every different type of player I think it's cool that I you know I do mostly one to two lineups and I am successful with it because most people have this concept that you can't be successful with that but I think that it's also obviously well proven that you can be extremely successful building lots of latinum if you do it the right way and so I think it's finding what works best for you is what's really important and and I think again to bring it back to where is the fewer games late more find I am consolidating lineups because there are fewer valuable players I mean inherently empirically when you have your ideal lineup and start making other ones to some extent each of those lineups is worse and how much worse they are is based on how many other decent plays they are so on a four-game slate it's gonna be much harder for me to make seven viable cash game lineups but on an 11-game slate like tonight I think I could make seven cash game lineups I would be perfectly happy with again just want to reiterate less lineups will increase your variance so if you only make one cash game lineup let's say normally normally an NFL I make one cash game lineup one league lineup leagues being anything that's paid out from 33% to like 20% is a league lineup for me and then a GPP lineup I'll have Optimum's of all those but sometimes what I'll do is I'll do an optimal 50/50 lineup and in a hospital heads-up lineup there is a difference between the two a lot of times in a 50-50 all you need is like you know that 45th percentile or a better line up and there's strategies where you can pair a quarterback and running back to get the points of that offense in basketball you compare Kevin Kevin Durant Russell Westbrook which raises your floor do things like that in a 50/50 which I think was the optimal play yesterday was the paired Durant Westbrook just to hit that floor in the 50/50 but it might not work the best especially in basketball because obviously one player works against the other player a lot of times the only time where it doesn't would be like when you pair like a rage on Rondo and a buddy cousins like that true assist guy to a true score is there like you know a really positive correlated play but I think I think like an NFL there's been plenty of times where I create 50/50 lineup and then an optimal heads-up line but I I'm also a one line of guys so we have to run a great job everybody we neglected to cover one of the things on our list of things to do which was talk about some of the excellent grinders tools but we will get to that and much much more these obviously a lot to cover so we will get back to this next week or next month or Never I don't know how we'll choose exactly as on I think we should have sort of a Hunger Games type thing it can be wrestling but you know if we have three other people who want to get into the cube maybe we could run a contest where all the hosts or co-hosts you know we have a GPP with the six or seven of them in the top three get to be on next week's shows that would be kind of fun I'll see if we can pitch that but um I'm back on tonight at 6:30 for Last Call with eMac I don't know when these first wild gentlemen are returning again any final thoughts boys now it's fun ah thanks for having me yeah and I saw any readers I'll see you guys on Friday I'll have an Oracle up covering some transition from NFL to NBA stuff on there as well all right well thanks all for coming guys you can send us all questions on Twitter if you have them and it was a great time good luck tonight know your contest

7 thoughts on “Game Theory: Learning How To Win At Daily Fantasy NBA”

  1. Fasteddie I been knowing you exactly 3 minutes and already I like you and your metaphorical way of "splaining" stuff. You remind me of me.

  2. "Should I attempt to get my wife into the sack tonight?" DFS has taught to me stay away from contests I have no chance of winning.

  3. Would be nice to get the links of the things you posted while doing the show for us guys that are watching on YOUTUBE.

  4. Perfect example re: low ownership not necessarily being longshots…tonight (Sharpshooter) has both Durant and Westbrook at sub 7% ownership! I love it when 90+% of my competition does not have my heavy lumber! 🙂

Leave a Reply

Your email address will not be published. Required fields are marked *