Daily Baseball Strategy: The Formula For Success

It’s pretty easy to lose sight of the objective when researching trends and strategies. It seems like there’s no end to the amount of numbers that you can crunch when constructing a lineup.

But in reality, DFS scoring only incorporates a handful of stats, which narrows down the amount that are relevant.

The ultimate goal is to discover what factors relate to fantasy point production. I set out to do exactly that by finding the correlation coefficient between Average Fantasy Points and 25 stats that I felt to be the most relevant.

Luckily the tricky calculations from back in college stats class are now obsolete thanks to Excel. Correlation coefficient is the measure of a relationship or dependence between two variables. A good example that I often hear used is the strong relationship between temperature and ice cream sales.

Anyway, I took the Average Fantasy Point totals for all 30 teams and ran the correlations to see which stats proved to have the greatest link to DFS production. Fantasy Labs stats are highlighted:

R Chart
 

Any result greater than 0.7 is generally considered a strong relationship. Ranking at the top of the list is Fantasy Labs “Upside”.

From a logical standpoint, it makes sense that Upside (the frequency with which a team/individual doubles their salary-based expected point total) would correspond well with overall fantasy production. The stat does well to reflect on past success.

upside
 

I do think there is value in incorporating Upside and Consistency as a prerequisite for player selection. Using Upside for tournaments allows you to start with the players that most frequently achieve elite point totals, while consistency for cash games separates out the players that most often perform up to expectation. It’s a great place to start.

From there, I recommend moving the selection process to OPS, wOBA and/or wRC. The correlations chart indicates that a solid performance in any of these categories aligns well with fantasy production. These stats are more predictive in nature due to the tangible outcomes that are needed to achieve a high mark (mostly extra base hits and run production).

Here are two charts from Fangraphs that you should reference when using wOBA:

wOBA graphs
 

The correlation numbers are even more favorable when weighted for matchups by handedness. For example, when right-handed batters face lefties, the correlation coefficient for wOBA increases from .789 (overall) to .826.

The moral here is to always remember which stats matter in DFS, and those are the ones with a proven link with fantasy production.

Start with the Fantasy Labs metric (upside or consistency) that corresponds with the type of contest. This will give you a foundation of players with the greatest tendencies to perform well within certain formats.

Next, sort your pool of consistent or high-upside players using wOBA, OPS and/or wRC (preferably by handedness).

Finally, win lots of money!

It’s pretty easy to lose sight of the objective when researching trends and strategies. It seems like there’s no end to the amount of numbers that you can crunch when constructing a lineup.

But in reality, DFS scoring only incorporates a handful of stats, which narrows down the amount that are relevant.

The ultimate goal is to discover what factors relate to fantasy point production. I set out to do exactly that by finding the correlation coefficient between Average Fantasy Points and 25 stats that I felt to be the most relevant.

Luckily the tricky calculations from back in college stats class are now obsolete thanks to Excel. Correlation coefficient is the measure of a relationship or dependence between two variables. A good example that I often hear used is the strong relationship between temperature and ice cream sales.

Anyway, I took the Average Fantasy Point totals for all 30 teams and ran the correlations to see which stats proved to have the greatest link to DFS production. Fantasy Labs stats are highlighted:

R Chart
 

Any result greater than 0.7 is generally considered a strong relationship. Ranking at the top of the list is Fantasy Labs “Upside”.

From a logical standpoint, it makes sense that Upside (the frequency with which a team/individual doubles their salary-based expected point total) would correspond well with overall fantasy production. The stat does well to reflect on past success.

upside
 

I do think there is value in incorporating Upside and Consistency as a prerequisite for player selection. Using Upside for tournaments allows you to start with the players that most frequently achieve elite point totals, while consistency for cash games separates out the players that most often perform up to expectation. It’s a great place to start.

From there, I recommend moving the selection process to OPS, wOBA and/or wRC. The correlations chart indicates that a solid performance in any of these categories aligns well with fantasy production. These stats are more predictive in nature due to the tangible outcomes that are needed to achieve a high mark (mostly extra base hits and run production).

Here are two charts from Fangraphs that you should reference when using wOBA:

wOBA graphs
 

The correlation numbers are even more favorable when weighted for matchups by handedness. For example, when right-handed batters face lefties, the correlation coefficient for wOBA increases from .789 (overall) to .826.

The moral here is to always remember which stats matter in DFS, and those are the ones with a proven link with fantasy production.

Start with the Fantasy Labs metric (upside or consistency) that corresponds with the type of contest. This will give you a foundation of players with the greatest tendencies to perform well within certain formats.

Next, sort your pool of consistent or high-upside players using wOBA, OPS and/or wRC (preferably by handedness).

Finally, win lots of money!