I received the following email this afternoon:
Just a thought. It would be interesting to look at the correlation between Starting Pitchers K/9 and the likelihood of a no-hitter. The natural intuition is that an increased amount of K’s limits the chances of lucky hits. With strikeouts pushing upper limits, you’d expect no-hitters to become much more common.
Figured you’d have the data to do it.
Here is the data in graphical format presented by decade, covering the years 1957-2012 (the range of the Day by Day Database). Only starter stats are considered here, and for a game to be considered a no hitter the pitcher must pitch at least nine innings and complete the game. Decades run from year 1 to year 10, so we are in the second year of the 2010 decade.
I was surprised to see no-hitters trend down as as a percentage of starts as strikeouts increased. It could just be that there are so many more games due to five expansions, you’re just bound to see one more often. I was surprised by this result, since more strikeouts should make no hitters more likely, since there is less of a chance of a ball going by a diving fielder. It will be interesting to see if the rest of the 2010s regresses toward the 0.03 to 0.04 percent no hitters of previous decades.
Sometimes K/9 is negatively related to IP/G. This could swamp out the effect you expected.