Accurate projections are the holy grail of DFS, the crux of success, and the primary concern of top-ranked players. Used in conjunction with game theory and diversification strategies, accurate projections will give you the best chance of coming out ahead in all your tournaments.
One option is to use third-party projections. There are a lot of sites that offer projections and others that test the accuracy of these sources. Be aware that thousands of other lineups are generated by other players using the same projections, which can lead to certain player combinations being over-owned. The third-party projection approach is optimal if you have limited time to invest into DFS, but if you go this route I recommend implementing a more contrarian strategy than you would use with my preferred approach of developing your own projections. Third-party projections still have a place in my process—in the form of an accuracy check—but I develop my own projections because I trust myself more than anyone else, and there is a contrarian element built into your process when no one else is using the same information.
To create your own model for every sport from scratch, however, requires an impossible degree of specialization. Luckily, Vegas is here to help. Sportsbooks are a vital source of information for modeling unfamiliar sports. Since the books are willing to take money on set odds, you can infer much from the bets they offer. Incorporating figures like team totals into your model for fantasy sports is a great way to speed up your process. These numbers might not be as accurate as you could make in an ideal world, but they free your time up to focus on other aspects of your model.
The process of creating a game total or a money-line is known as handicapping. Sportsbook employees, with complex algorithms in hand, set the opening total or line. Yet the true accuracy of Vegas doesn’t originate with a single algorithm or even the collective efforts of all sportsbooks. Vegas owes its accuracy to highly-sophisticated sports bettors who are identifying inefficiencies in betting markets and taking advantage of them.
Sportsbooks set the line at whatever will make them the most money, so a high volume of money being placed on one side of the bet, will change the line the sportsbooks offer for subsequent bets. The more money they allow on a bet, the more efficient the lines will be as professionals make up a higher percentage of the total bets. The accuracy of betting lines is maintained as information continues to emerge, for example, in the form of player-injury or weather updates. In fact, this is probably the greatest benefit of using betting lines—the wealth of information that is incorporated into a single line, meaning less factors that you will need to adjust for in your own model. To get the full benefit of this information, you should update these numbers close to game-time, when they are the most accurate.
Because the margins of error at any given time are small, these bets are reliable sources of information. As I’m writing this, the Pacers vs. Pistons game tonight has an over-under of 208 and a spread of -2 favoring the Pistons, according to vegasinsider.com. Using this information, you can deduce that the Pacers are expected to score 103 and the Pistons 105 (these numbers are often referred to as “implied team totals”). Furthermore, you can use implied point totals to make educated guesses about the pace of play and other stats. It’s also important to consider the odds offered on each bet. You should adjust the numbers you use in your spreadsheet towards whichever side has odds farther in the negative. For instance, in NHL if the over-under is 5.5 with the over bet being -125 and the under being +105, you could enter a number slightly higher than 5.5 in your spreadsheet. You can calculate exactly how much to adjust it by a comparison of different sportsbooks over time, like if one set a line of 6 in the same game skewed to the under.
Once you estimate the statistics for the entire game, you can compare those to a teams’ average performance and adjust individual players expected performance accordingly. The baseline for players’ projections should, of course, be their historical stats. Although traditional per game or per minutes stats work well, to gain an edge, you’re ideally going to want to sprinkle in some advanced/non-traditional statistics. Not only do some have superior predictive value but they’re also another way to differentiate your lineups from your competitors. There are a plethora of advanced statistics available on sites such as ESPN, NBA.com or NFL.com—many of which I have yet to study—but some of the more helpful stats I’ve found so far are usage for basketball, isolated power (often denoted as ISO) for baseball and routes run for football. But be selective, the vast majority of non-traditional statistics are useless for DFS projections.
With advanced statistics, it’s also important to ensure that you aren’t accounting for the same factors multiple times, which will diminish accuracy. It’s essential to know exactly how each stat is defined so you can properly use it; be careful, some sites use different calculations for the same stats! One example is pace of play in NBA, which has several different definitions depending whom you ask.
Not only will the type of statistic you use be crucial but also the size of sample you select. Larger samples are better than smaller ones, but you’ll have to balance that against decreased accuracy associated with a player being in different contexts or the improvement/decline that occurs with age. As a general rule, you’ll need larger sample sizes for statistics with higher volatility.
Finally, when you create your projections, you’ll want to check them against other sources. Even the best prepared DFS players miss updates that significantly impact projections. If you don’t want to shell out money for sophisticated projections (be wary of sites offering them for free), you can reference the letter grades I assign to players each day in my rankings. These rankings are based on the actual projections I used to generate my lineups and are updated frequently throughout the day.
Another free resource are the “prop” bets listed on sportsbooks, covering things like passing yards for quarterbacks or points for a NBA player. These numbers are informative as they wouldn’t be offered if they were easily outsmarted. However, these bets should not be treated with the same degree of respect as team totals or money-lines, as the market for them is less efficient than traditional bets. Prop bets aren’t as ubiquitous, sportsbooks cap the dollar amount of each bet, and they ban sharp users. As a result, they don’t have equivalent accuracy of a game total. If you are confident in why your projections differ from other sources, that will help guide who you put in your lineups.
If you want to compete in DFS at an elite level, it’s essential that you create your own projections. Because the profit potential of accurate projections is much higher than anyone would reasonably pay for it, these projections are not publicly available. Although the process of creating projections and incrementally upgrading your algorithms has many more growing pains than using third party sources, this approach will give the greatest chance of success in the long term.