4 Tips For Making Your Best Lineup
The majority of the time I spend in daily fantasy NBA is doing research and thinking about my projections for the players of the day. But knowing who the best plays are does not necessarily mean you will make the best lineups. Making the best lineups given your projections involves understanding some concepts of optimal lineup building.
I’ve set up some dummy salaries and projections based on the 2 games on 2/19/15. There are only 36 players to choose from and the projections are quite reasonable: No player is vastly better than the others and the highest projected player is Russell Westbrook at about 45.8 fantasy points. Jamal Crawford is the best value at 4.83 fantasy points per 1000 dollars. You can view the salaries and projections here. The best lineup given these projections end up being this one.
Below are my thoughts on the optimal lineup, the best lineups from the generator, as well as concepts we can learn from different sets of projections.
Use All Your Salary
The optimal lineup given these projections uses all but $200 of your salary cap. The average salary remaining in the top 10 lineups in this projection set ends up being $260, which is quite negligible. Even if we change around the projections to give some very strong value plays, we see the same type of results. I changed Dion Waiters and Nick Collison to 6 fantasy points per $1000, and the average salary remaining actually moved down to $180. Overall, it may be tempting to leave $500-$1000 on the table to get everyone you like in your lineups, but chances are there are players you have picked who are worth upgrading, even if you think they are strong values.
Value Absolute Fantasy Points Over Fantasy Points Per $
An interesting experiment we can do is to see what happens when we change projections to include a position with a strong value plays and high projected high salary players. Let’s say we change Russell Westbrook and Chris Paul to have projections of 5 FP per $1000, but also change two low salary PGs to the same values, Cory Joseph at at 3.5k and Devin Harris at 4.5k. All 4 players have the same value, so which players should we prioritize in our lineup? Does it matter?
When the numbers are crunched, it turns out the top 10 lineups all include Westbrook and Paul, despite them having the same Pt/$ value as the lower salary players. When Joseph and Harris are upped to 5.5 FP per $1000, Paul and Westbrook are 8 out of 10 of the top lineups, while Harris and Joseph make one appearance each. Even when we up Joseph and Harris to 6.5 FP per $1000, we don’t find both Harris and Joseph in the top 10 lineups, although one of them appears in 7 out of 10.
The reason a strong high salary play is always prioritized over a low salary play is simply because high salary plays contribute to a larger percent of our lineups score. A $12000 player on FanDuel is going to contribute around 20% of our lineups total fantasy output, while a $4500 player will only contribute around 6.7% of our lineups output. If we whiff on a high salary players, we’re pretty screwed, but if a low salary player does badly we still can easily have a good night.
With That Said, Strong Value Plays Are Your Bread And Butter
Let’s go back to examining the optimal lineup given the dummy projections. If we arrange our sample NBA projections spreadsheet by Pts/$1000, you’ll find that an overwhelming amout of top value plays are in our lineup. The optimal lineup ends up being the best Pts/$1000 plays at every position, except for PF, where Serge Ibaka and Nowitzki are the 3rd and 4th best values.
Assuming there is not a plethora of extremely good low salary plays, your best lineup is going to be close to the best values at every position. Almost all of the top lineups include exclusively players who are the top 4 Pts/$1000 plays at each position. On the other hand, the optimal lineup only includes 2 out of the top 7 expected fantasy point producers of the night. This is mostly because of the salary texture of the players we have to choose from doesn’t have a lot of strong, low salary plays, and…
High Salary Plays Must Be Matched With Low Salary Plays
The optimal lineup in this projection set has only one player, Chris Paul, who has a salary above $8000. The issue with the this days players is there are not a lot of good, low salary values. The top plays with under $5000 salaries are Diaw, Redick, and Hawes, all of whom are not overwhelmingly good. What happens if one of these players were to have a very good projection? Let’s change Diaw to 23.4, or 6 FP per $1000. The majority of lineups now include Kevin Durant or DeAndre Jordan, as well as Paul or Westbrook. The amount of high salary players in our optimal lineup ends up being proportional to the amount of strong low salary options we have. It’s easy to understand, but quite important to making your best lineup.
- DFS players focus mostly on top picks, but knowing the best plays doesn’t mean you will make the best lineup.
- In general, you should try to use your entire salary cap. You may love all your plays in a lineup with $800 remaining, but chances are there is somewhere you can upgrade.
- Top high salary plays should be valued over great value plays.
- With that said, your best lineup will mostly consist of the top 3-4 Pts/$ plays at every position.
- High salary players are only appealing if they can be matched with feasible low salary plays.
View all posts by Daniel Steinberg