Yesterday I got into an interesting discussion after Bobby Tewksbury proposed the topic of measuring a hitter's "streakiness". Daniel Tuckerinformed us bout Corr VOL which measures how evenly a hitter distributes their Run creation.
While I think it's more important because teams win Games not weeks, I still believe there may be value to understanding which hitters are streaky and which are not. I think there are potential coaching and player development implications to help understand who is more streaky and why.
So I went and looked at 3 different players( Merrifield, Harper, Machado) with the same WRC+ for the season, and looked at their wOBA and wRC+ Standard deviation for 7 game averages. The idea here is that the higher the 7 game standard deviation for wOBA or wRC+, The more "Streaky" a player is.
Whit has a wRC+ standard deviation of 38, Bryce 45, Machado 68. I was honestly expecting Harper to have a higher STD though I think his 2019 has been different than years past where it seems he had stretches of really strong weeks. His "Valleys" are also a lot higher than I was expecting.
Machado has a very large STD. He is seemingly either below average or God-like, whereas a guy like Whit has streaks where he performs slightly below average or moderately above average.
How much weight any of this holds, I'm unsure but I do see benefits for coaches understanding which guys are streaky and perhaps getting to know their player's process better so that they can limit the valleys more and help them find the things in their process during peaks.
A couple of notes: how you slice the games can affect the streaks numbers. I could have looked at the rolling week averages but because guys don’t play exactly 7 games in a week I settled on just using 7 game stretches instead of 7 day stretches because I wanted each stretch to have roughly the same number of at-bats.