Factoring Consistency into your Fantasy Draft Picks
When you scroll through the list of players in selecting your fantasy lineup, think about the main statistics that are available. Regardless of the site you choose, you always see salary and average fantasy points per game prominently displayed. The reason for this is apart from matchup characteristics, this is the main criteria most fantasy managers use to asses player value. While season averages have been proven to be best predictor of performance on any arbitrary day, they must be taken a step further to determine a player’s true value.
Compare Randal Grichuk to Mike Trout. Over the past 10 games, these two players have hit similar fantasy point averages (around 3.9 points per game under Fanduel scoring). However Mike Trout is going to run you almost twice as much as Russell Martin to put him in your lineup. Is this simply a mis-valuation by the daily fantasy sites? No, they are factoring in a couple other things into their valuation.
First off, Mike Trout has been performing right on his season average over the past 10 games but Grichuk on the other hand has been performing about 50% better than his season average. Therefore, his lower salary is indicative of how he has performed all season, not solely the past 10 games.
Second, and more importantly, Mike Trout is a more consistent performer. The standard deviation of his fantasy points over the past 10 games is 3.41 points compared to Grichuk’s more volatile value of 5.13 points. This means, you can better predict the performance of Trout while Grichuk’s may be very high or may be very low.
However, although this shows up as a positive (increase) in salary, this may me good or bad depending on what your goals are for your lineup. If you are entering your lineup in a 50/50 contest then consistency is an important aspect you would like since you don’t need a huge point total, you just need a solid and reliable performance from your players.
If you are shooting for a more aggressive MLB fantasy tournament type such as a winner take all or high stakes tournament however, you need those volatile players. Volatile means they come at a bargain and although they usually don’t, on the right night they could score big points for your lineup. Furthermore, those risky players usually have lower ownership percentages within the cash draft so if they score big, you are likely the only one benefitting.
Below you see a chart where each player is plotted on their consistency versus their average points scored all over the past 10 games that player has played. Let’s rethink of these two factors as risk and reward instead of points and standard deviation. Points are the reward since they are what we expect to get from that player on average and standard deviation are the risk since it is what makes the players performance more volatile and harder to predict.
The general trend of the data is that as risk increases, reward increases as well. This is depicted by the best fit line (red line) that seems to split the data in two parts. Points below that red line take on too much risk for the reward they offer, and points above the red line provide extra reward than what you would expect for that level of risk. Hence, players above the red are preferable.
You’ll notice the points way up in both risk and reward, greatly exceeding the red line. The player at the top is Manny Machado who has averaged 6.55 fantasy points over his last 10 games (despite a season average of 3.2 fantasy points) and a standard deviation of 4.41 fantasy points.
Therefore, the more reward a point provides for a given level of risk, the more fantasy value that player brings. The threshold for that is depicted by the blue line. For each level of risk, that line shows the maximum reward that can be attained. The dark blue area shows the spot where reward most exceeds risk and is the place of greatest fantasy value.