Everyone knows sports are unpredictable day by day and game by game. Accuracy across small samples is an exercise in futility. However, DFS players love to try anyway. And, they drive themselves nuts if they don’t have realistic expectations.
The chase is great as a night winds down, but in all honesty, the results from one contest aren’t particularly helpful even if your bold call hits big, or if your lock of the day fizzles. To gather a real sense of your level of skill and success, you need data. Lots of data! And, over large samples that reduce outlier performances (good and bad) to isolated incidents that have little to no effect on your overall results. Diversification very much helps in this regard.
Just like a financial advisor will recommend, in many cases, mutual funds over stocks, a DFS coach will recommend playing more sports or more slates. But, those aren’t the only ways to gain diversification in DFS…
More Sports and More Slates
Let’s get the obvious out of the way first. Different slates on the same night can be considered unique events. Ownerships and key plays change when a 15 game main slate becomes a 5 game turbo or 2 game after hours slate. The chalk shifts and leveraging that chalk leads to different lineup constructions, sometimes VERY different. Any slate is fair game. From the big main slates to the smaller mini-slates and the notorious showdown, single-game slates (provided you have a viable strategy).
Since the results of each individual slate are independent of each other, the lack of correlation provides diversification to your bankroll. You can play as much as you would on any other slate without incurring much, if any, additional risk.
The same applies to adding new sports. If you invest 5% of your bankroll on an MLB main slate, you can double that by adding PGA. And, you can triple it by adding NASCAR, and so on.
More Lineups Are Better For Diversification
Most of us operate within a budget, or bankroll. Bankroll Management is key to long-term survival in the daily fantasy sports hobby. It wouldn’t make a lot of sense to add more and more $20 tourney lineups to your nightly play if you can’t afford to do so. Instead, dropping from the $20 level to the $1 level allows 20 unique lineups instead of just one. And, if you want to get really crazy, 80 lineups of quarters is $20. In fact, playing a full 150 lineups of nickels ($7.50), then adding 50 lineups of quarters ($12.50) gives you a total of 200 unique lineups for the same $20 investment.
Adding lineups (and thereby lowering your buy in) reduces variance. It also increases return on investment. For one, you are playing weaker competition as a whole. For another, you are playing bigger contests with bigger winning prizes in relation to your buy in.
Making a lot of different/unique lineups allows you to spread your risk over a wider player pool instead of relying on the unpredictable outcomes of just a few. Yes, many people love the “all in” nature of playing just a few players heavily, but too many beginning players fall in love with the “optimal” from an optimizer and play it nearly exclusively. This strategy leads to some truly disastrous nights. And, it’s not just the single-entry players doing this. In many cases, the full mme players also recycle their same 150 lineups across all their contests, too.
Given the high levels of variance and unpredictability in sports on a given night, the difference between the top projected lineup (optimal) and second projected, or third or fourth or fifth or hundredth, is mathematically insignificant when it comes to each lineup’s winning chances in these larger field contests. Furthermore, that top projected lineup often isn’t even the optimal in terms of profitability when you adjust for game theory.
Diversification and Reducing Lineup Inter-Correlation
When switching from making a few lineups to making more lineups, you must take steps to ensure there are significant differences between each lineup or your results/performance will be highly correlated on a given slate. The issue with correlated lineups being too similar is your entries bunch together. Bunching lineups is great when things go right, but devastating when things go wrong. Bunching lineups increases volatility and actually serves to decrease your expected profit. When you narrow your range of outcomes, you actually start competing against yourself for the top spots.
To decrease lineup correlations, you need to carefully build by hand (great for 10 or fewer lineups), or you need to become proficient in using DFS optimizers for proper diversification. Let’s talk about some features intentionally invented to combat lineup correlation.
Randomizing features typically allow you to apply a percentage range to the median projections provided. Some sites allow between 1% and over 50% adjustments. If you hypothetically choose 10%, every player’s projection will recalculate as the algorithms run and assign every player a new projection between 10% higher and lower than the median projection. For example, Player A carries a projection of 10 pts tonight. Each run (150 runs if using 150 lineups) will generate a new projection for Player A between 11.0 (10% higher) and 9.0 (10% lower) as it builds your lineups for you. The result is slightly different projections for every lineup you create, adding diversification to every lineup automatically.
Small differences in projections can lead to dramatically different players in lineups. As previously mentioned though, these differences don’t dramatically lower your chances to win big tournaments in unpredictable sports from slate to slate. The ideal settings vary from slate to slate and sport to sport. And, these settings are widely discussed between our coaches and players on a nearly daily basis.
Minimum and Maximum Player Exposures
Many optimizers allow for a maximum player setting. Very few allow for a minimum (Hint – DFS Army has minimums for more finite control!). When used in conjunction (but not overused), these features combine to deliver a lot of diversification.
Optimizers, based on projections, definitely fall in love with specific players on some slates. By placing a maximum exposure of, say, 25% on a player, the optimizer will remove this player from your pool once the player exceeds 25% of your current lineups built. And, the optimizer will add the player back in once the player dips back below 25% and more lineups are optimized.
To achieve maximum diversification, applying a different maximum or minimum for each player will achieve the best results. PRO TIP – Some of the lineups using this method will be very poorly constructed and projected. To combat this, simply generate more lineups than you need and eliminate the bottom projected lineups. Removing the bottom 20-25% of these lineups should suffice.
Applying Unique Player Settings
The unique player setting specifies the number of overlapping players any lineup can contain. This is perhaps your best weapon when prioritizing the reduction of inter-lineup correlation.
For example, if you specify three unique players, every lineup you generate will have a minimum of three different players from every lineup previously constructed. Unique player settings can ensure significantly different lineups across an entire build from 20 lineups to over 200 lineups.
The ideal setting will range from around two players to upwards of five players. Bigger slates favor more unique players to cover the larger player pool available. And, the more volatile sports (think hockey and baseball) will also benefit from a larger number of unique players used, again, to cover more viable options.
Final Words About Diversification
Another great strategy is backtesting your settings across multiple slates. Backtesting allows the serious player to hone in on patterns winning lineups are using to create very big scores. The results from slate to slate may not be predictable, but the patterns continually showing up in winning lineups are. Duplicate the settings that create these winning lineups and give yourself increased chances at winning the contests you play.
Having a strong diversification strategy in DFS is like choosing mutual funds over individual stocks. Diversification alone will keep you in the daily fantasy game longer, buying you more time to hit the big one. Using increasingly diversified lineups will increase the chances you have a good day, but it will also decrease the chances you have a phenomenal day. This is a trade-off in such unpredictable sports I will always take.
Entering multiple slates per day will allow players to increase exposure to prize pools without significantly increasing the risk they are taking. And, having more diversification between your individual lineups will spread your results out throughout the tournaments bringing you more consistent results…and decreasing the chances you compete with yourself for that coveted top spot so critical to success in 2022.