Another night of data is in, and the bell graph is updated above. I would like to point out, too, that there is a solid correlation between Fanduel and Draftkings scoring pitching, so the data below could be applicable to both sites. Look at this lovely scatterchart :
Who doesn't love a r-squared of .9934? I know I do...
My thought today, after losing in last night's matchup with Atlanta starter Drew Smyly, was : is there a different way we can look at the data?
Recall the scoring rules for FD:
- Win = 12 pts
- Strikeout = 3 pts
- Earned Run = -3 pts
- IP = 3 pts
Today's thought was to look at three categories, quartile them, and look at average points earned by quartile. The three categories are defined as :
- XFIP (or fielding independent pitching) - a better gauge than ERA, or earned run average
- K/9, or how many strikeouts does the pitcher average per 9 innings pitched
- Ground ball %, or how many balls in play (or balls hit by opposing batter) are ground balls
I like the ground ball idea because more ground balls equal more outs. Let's see what we have with our limited data set.
Interesting. Blue cells highlight the highest average; red cells highlight the lowest average. So, in an ideal world, I'd want to target a starting pitcher that :
- Has a FIP lower than 3.44
- A K/9 greater than 9.19 ks/9 innings
- A groundball rate greater than 42.4%
Easy enough, right? Probably not, but it does give some guidance when wading through each night's probably pitcher slate.
Next week I'll start reporting m results; the goal is to enter a) 1 tournament and b) 1 50/50 entry, and record progress as the season plays out. As for my Brewers tonight? In theory, they should rough up the Reds, but I doubt it. Either way, I'll probably have a few beers during the game.
Thanks for reading; more to come, I promise...