Justin Verlander pitched his second career no hitter on Saturday May 7, shutting out the Toronto Blue Jays in a 9-0 victory. It was the 271st no hitter in Major League history (at least according to the Wikipedia list, and we know they’re always right). It was certainly a great accomplishment for Verlander, but something peculiar struck me about his feat… the fact that it happened a mere four days after Francisco Liriano threw the first no hitter of 2011 on May 3.
Last year, we saw a whopping six no hitters in 2010. Those dates were April 17, May 9, May 29, June 25, July 26, and October 6. At times in 2010, it seemed like there was a no hitter being thrown every week in the early parts of the season. The frequency of no hitters happening right after one another seemed quite peculiar. Looking at the last two years, it almost seems like no hitters breed more no hitters!
So, as someone who has spent a little time with math, I did what any geek turned sports blogger would do… I spent hours on the couch in my mother’s basement looking deep inside the numbers to see if this whimsical notion about no hitters breeding no hitters was in fact just a figment of my imagination, or one of those odd statistical wonders that only the sport of baseball could create. Even I was surprised with the results…
Looking at the last 15 years (1996-2011), there have been a total of 32 no hitters thrown. Here’s some of the interesting stats about no hitters thrown in the last 15 years…
1996-2011
-Number of no-nos: 32
-Min # of days between no-nos: 3
-Max # of days between no-nos: 841*
-Adjusted average # of days between no-nos: 101.2*
*To balance out offseason days (obviously a no hitter can’t be thrown during the offseason) I subtracted the 183 days between October 1 and April 1 as a quick estimate of offseason days that have inflated totals of gaps between no hitters from one season to the next. For example, the 841 days between Randy Johnson’s perfect game (5/18/04) and Anibal Sanchez’s no-no (9/6/06) was really only about 475 days of baseball action.
With 271 no hitters in the 135 years since the first official one in 1876 (by George Bradley, wouldn’t want to slight him), that averages out to right around 2 no hitters per year. So, the 101.2 day average between no hitters makes sense in about a six month baseball season. Even though 271 no hitters appear to be a lot (there have only been 15 unassisted triple plays and 15 4 HR games), it is still an incredibly rare feat considering the massive number of games pitched in MLB history. According to Baseball Reference, there have been 396,270 games played all time, which means .068% of games will include a no hitter. Every 1,462 games, there should be a no hitter statistically.
Of those 32 no hitters in the last 15 years, 18 of them followed a no hitter in the same season (aka, multiple no hitters in the same year) ala Justin Verlander this year. The astounding thing is that 7 of these 18 no hitters (38.9%) occurred within a month of the previous no hitter! Considering the adjusted average is roughly 100 days, this is an unreal statistic! Furthermore, the average number of days between in season no hitters dropped from 101 days to 49 days! Now, of course some of that drop is to be expected by removing the largest totals, but those numbers are still astounding. Consider last year’s amazing run in 2010 when six no hitters happened 22, 20, 27, 31, and 72 days apart. Earlier in the same decade, there were only four no hitters from June 11, 2003 to April 18, 2007 when it was 295 days or more between no-nos. What a strange phenomenon!
I thought that maybe this was just a bizarre trend in recent baseball and not indicative of no hitters throughout baseball history. So, as another random sample, I took a look at no hitters from 1900-1915. Here are those results…
1900-1915
-Number of no-nos: 36
-Min # of days between no-nos: 4
-Max # of days between no-nos: 733
-Adjusted average # of days between no-nos: 83.4
-In season average # of days between no-nos: 43.7
There were 21 instances between 1900 and 1915 when a no hitter was followed on in the same season like Verlander. Of those 21 times, an incredible 12 (57.1%) occurred within 30 days and 9 happened within 15 days. Can you believe that? A no hitter happened only 36 times in those 15 years, yet 9 times a no hitter was duplicated within only a couple weeks. Just to prove to myself that I wasn’t going crazy, I looked at the 15 year stretch between 1972-1987 that contained a similar number of no hitters…
1972-1987
-Number of no-nos: 36
-Min # of days between no-nos: 3
-Max # of days between no-nos: 719
-Adjusted average # of days between no-nos: 79.3
-In season average # of days between no-nos: 53.8
In this period, there were 22 no hitters that followed on in the same season. Exactly half of those no hitters, 11, happened within 30 days of the previous no hitter! So, over those three 15 year periods, 30 out of 61 no hitters that followed a previous no hitter in the same season came within a month. Considering the baseball season lasts upwards of six months, the fact that so many no hitters occur within such a small, relative time frame of each other is implausible, but somehow, a reality.
As a final “proof” of this theory that no hitters breed no hitters, I examined each of the 271 no hitters thrown in MLB history. In that study, I found 38 of the 271 no hitters (14%) took place a year or more after the previous no hitter. Naturally, 81 of the 271 no hitters thrown (29.9%) have occurred within 30 days of the previous no hitter! Therefore, it is more than twice as likely for a no hitter to happen within a month of another no-no than it is for a no hitter to happen a year after one.
So, perhaps Justin Verlander’s performance in doubling up Francisco Liriano’s no hitter in less than a week isn’t all that rare of an event. In fact, looking throughout baseball history, it’s actually quite common. Considering the miniscule probability of a no hitter, the fact that so many of them occur so close to each other is one of baseball’s great mysteries. While this quick statistical research is far from bulletproof, (I’m sure there are hundreds of sabermetricians that could speak to this above my head) it does make a convincing case for no hitters breeding no hitters. I don’t know how it happens, but watch closely, because another no hitter might be right around the corner.
