A Guide to Winning Tournaments on FanDuel
Tournaments are a big draw for any daily fantasy player. Cash games might be good for building a bankroll, but the allure of the big payout is what makes tournaments an attractive (and fun) game type.
They aren't easy to win, but here are some tips to win big tournaments on FanDuel.
Look for Ceiling over Safety
This is probably the most basic tournament advice there is, but it is important to remember. To win a tournament, you need to have the highest scoring lineup (duh). But like I said, tournaments are attractive to pretty much every one, so almost any lineup that gets made by a player will likely be throw into a tournament. This is especially true at the lower stakes. So to win big tournaments, your lineup really needs to be one of the very best lineups on that site on a given night.
The safe choice isn't going to get you there, so you need to find guys that have the potential to finish with crazy high scores. Now obviously you won't be able to get guys with huge upside in every spot and stay under the salary cap, so you are really just looking for guys that can outperform their price.
In NBA, this is the oft-talked about concept of value. For cash games, you need 5x the salary (so a $9,000 player would need 45 points to reach "value"), but in tournaments you should be looking for players that could potentially hit 6x or even 7x.
Other sports aren't viewed with the same strict emphasis on "reaching value", but the concept remains in the same. In football, hockey and baseball, there is a big fantasy emphasis on a single event (touchdowns, goals, home runs respectively). Ideally, you want your lineup filled with guys that can get multiple touchdowns, goals or home runs if everything falls right.
In all sports, there tends to be a category of players that can best be described as "boom/bust". These guys have the potential for huge fantasy games, but they also bring the risk of absolute duds. It is this type of player that you should be trying to get into your lineups. If you get them on a boom night, you will have a chance to win big tournaments. If you catch them on a bust night, you are out of luck. But that is the risk you have to take. By choosing a safe player instead, you are limiting your chances of coming in 1st before the games even start.
Correlation is Key
Let's play a game. To win the game, you need to get 2 points. Points are awarded for correct guesses of a coin flip. You have two choices: you can guess the outcome of one flip with a correct guess being worth two points or you can guess the outcome of two flips with each correct guess being worth one point. Which option do you take?
Obviously you take the first option: one flip worth two points. Your chance of getting any one flip correct is 50%. The likelihood of you going 2 for 2 is 25% (50%*50%).
This can be applied to DFS. If you take a player from a different game in every spot of your lineup, you are betting on 9 (or however many roster spots there are for that sport) independent events. Research and skill can increase the likelihood of being correct on a particular event, but the odds of you going 9 for 9 are very low. Even if you have a 95% chance of being right on each spot, your chances of going 9 for 9 are just 63%.
The easiest way to improve your odds is by lowering the number of independent events you have in your lineup. If you get that number down to 4, and you have a 95% of being right on each of the 4, the chance of you getting every one right (going 4-4 in this case) jumps to 81%.
So how do you get the number of independent events down from 9 to 4? Through correlation.
This takes on different forms in different sports.
In football, you can pair a quarterback with a wide receiver. Now the correlation coefficient isn't 1 (which would mean perfect correlation) like it was in my hypothetical example above (and it won't be as strong as that for any sport, I just used the example to show the impact cutting down independent events). But these two spots now have at least some correlation and at the very least are no longer independent events.
The WR is obviously extremely dependent on the QB. Barring injury, any points that WR gets are going to involve points for the QB. The QB though has other WRs to throw to, so he can get points that don't involve the WR (which is why the correlation coefficient isn't 1). But by pairing a QB and a WR, you are essentially getting double points any time the WR makes a play. And if the WR has a big day, you know the QB did as well. You can even pair a QB with two of his WRs or a WR and a TE. In this case, a huge day for the QB would likely result in 3 players on your team posting very high scores.
The alternative to this is selecting a QB from one team and a WR from another team/game. This situation has no correlation. For the QB to do well, his WRs (who you don't have on your team) need to do well also. Same goes for the WR and his QB (who isn't on your team). You also need the game flow in both of those games to dictate heavy passing for both teams you have players from.
So by pairing up the QB/WR, you need a lot less things to go your way in order to have them finish with a high score.
Correlation in baseball comes through stacking a particular offense. The hope is that one player on your roster will reach base and then his teammate (also in your fantasy lineup) will drive him in, giving you points for both the run and the RBI. If you don't stack, you are counting on the teammates of every player on your roster to get on base ahead ahead of your hitter and to drive him in if he reaches bases.
In hockey, +/- makes stacking a particular line a great way to correlate your team. If that line scores, you will get a + from every player and you are also likely to have the guy who scored the goal and one or both assists on it. if you take 3 forwards from 3 separate lines, you need 3 different lines to score a goal. If you take 3 forwards from 1 line, you need 1 line to score a goal. This goes back to the coin flipping example. Predicting one event correctly is much easier than predicting three.
NBA is probably the least important sport in terms of correlating. There are so many more opportunities to put up fantasy points compared to the other three sports. You could pair a PG with a SG in hopes of the PG assisting a lot of the SG's field goals, but that isn't really a common strategy.
Find Your Edge
This may seem obvious, but I think it is something a lot of people overlook. If you are doing the same thing as everyone else (reading the same articles, checking the same key factors, listening to the same experts), how can you expect to win consistently?
In cash games, maybe there is enough casual players who do limited to no research and you can finish in the top 50%. But to cash in a tournament, you need to beat 80% of your competitors. To win that tournament, you need to beat all of them. You simply aren't going to do that by stacking the game with the highest over/under or rostering the player that has been lighting it up for the past few games. You need to find an edge, something that everyone else is overlooking.
Again, this will vary by sport, but you essentially have to dig deeper than Vegas line and expert advice. Here is an example of what I am talking about in each sport (there are obviously many ways to do this for every sport).
For MLB: Ignore BvP (batter vs. pitcher) and hot streaks, focusing on the underlying metrics that drive player performance (BABIP, Line Drive %, xFIP). For example, in a recent article, I recommended a Brewers stack against Chris Sale. I believed that Sale was hurt due to a few statistical indicators and decided to play the hunch. He wasn't hurt and he pitched great, but that is the kind of thing you have to do to separate yourself from the pack. And if I see a pitcher with similar injury indicators, I will stack against them again.
Instead of playing the guy that is hot, play the guy you think is about to get hot based on your research.
For NBA: Granted, I barely play NBA, but WOWY (With or Without You) analysis is a good place to start. You can analyze player's usage rates when certain units are on the court. So if a player is injured, you can look at the WOWY numbers with him off the floor to see who actually replaces his production.
For NHL: Advanced stats like Corsi are gaining popularity but they still aren't commonplace, especially in DFS circles. Embracing them and using them to your advantage is a great way to separate yourself from the pack in NHL.
For NFL: This is the toughest because the NFL is the most popular sport. There are a lot of people that are very knowledgeable about football (and countless others who think they are). The best way to gain an edge in NFL is by trying to anticipate game flow and gameplans. Or by trying to identify coverage matchups. For instance, if you know a WR takes 85% of his snaps out of the slot and his is facing a team with a terrible nickelback, you can take advantage of that.
I have written about going contrarian a bit before. That article has the mathematical explanation of why going contrarian can work as well as some thoughts on how it applies to hockey.
But to summarize, going contrarian is essentially taking a low-owned player/team/stack. Contrarian picks generally come with a trade-off. You are taking a player/team/stack that is less likely to come through, but the payoff will be huge if they do come through.
For NHL, there are multiple ways to do it, as outlined in my previous article. You can take a 3rd line from a popular team or take the top line from a team that everyone expects to get crushed.
For NBA, a contrarian pick might be something as simple as paying a little more for Russell Westbrook when Steph Curry is a slightly cheaper and has a better matchup. If Westbrook manages to outscore Curry, you have an advantage over everyone that used Curry.
NFL is similar to NBA. If an obvious play emerges throughout the week, a contrarian move can be to fade that player and take a similarly priced guy you believe could potentially outscore them. Or if there is a value RB that everyone is talking about, a contrarian move would be to pay up at RB and use a value player at a different position.
MLB requires a similar approach to NHL. Contrarian stacks can either be less-popular choices (guys that hit low in the order) from the popular teams or the main guys from a team no one is talking about.
The Recipe to Win Big Tournaments on FanDuel
All of these tips will help get you closer to winning a tournament, but you also need a little bit of luck to go your way. Winning a big tournament is extremely hard for even the best players. The strategies outlined in this article can help you get into the upper regions of the standings, but you will need a bounce or two to go your way. And you have to understand that when playing a tournament. Obviously you want to win, but cashing should be your goal. Cashing keeps you in the game and gives you the chance to play more tournaments. And if you keep getting lineups into that upper tier, you are giving yourself more chances to have a bounce go your way, enabling you to win the whole thing.
*Photo courtesy of FanDuel