Mariano Rivera - Era-Adjusted Home Runs (HR/9) Using HR-Plus

An earlier post - Mariano Rivera's Nasty Statistics - discussed Mariano's high rankings in several statistical categories. Since statistics such as strikeouts, home run rates, and runs scored have varied so much across MLB history, I've been working on a mini-project to normalize these stats across eras, primarily for the all-important purpose of seeing where Mariano ranks. This post presents the career leaders in adjusted Home runs per nine innings (HR/9).

Using actual stats, Mariano's .47 HR/9 ranked a seemingly mortal 264th out of the 1037 pitchers who have pitched at least 1000 innings since 1901. However, since Mariano has pitched in the highest HR environment of all time, looking at adjusted stats underrates Mariano. For example, Hall of Famer Rube Marquard had a .291 career HR/9, ranking him well above Mariano. But Marquard pitched from 1908-1925, where the average HR/9 was .235. So Marquard was actually 20% worse than league average at preventing HRs, while Mariano's .47 HR/9 is more than twice as good at the league average from 1995-2011 (which was a 1.064 HR/9).

To see beyond such distortions, we need to adjust the pitchers' HR/9 stats to a common scale. In this exercise, I adjusted career HR/9 stats to the 1953 AL, which I chose because that league averaged a HR/9 of .715, which is almost identical to the MLB historical average HR/9 of .714. (Note: The method would rank pitchers the same regardless of which year and league used as the adjustment target.)

To adjust, I used the Home Runs Plus (HR+) statistic, which is one of the "Plus" statistics Diamond Mind Baseball uses to adjust for era differences. HR+ measures how far above or below average a given pitcher's HR/9 in a season was to the league average - a score of 100 is average, above 100 is above average, etc. For example, in 2011 Mariano's HR+ was 224, indicating he was 124% better than the 2011 AL league average in preventing home runs.

Once you have pitchers' annual HR+ stats, you can use that data to calculate pitchers' career HR+ stats. You can then use career HR+ stats to adjust pitchers HRs allowed and HR/9 for a given season's HR rate, which provides the needed common scale mentioned earlier. To spare those who are averse to math and are just interested in the results, the calculation method is described on a separate page - see http://www.guttygrittyyankees.com/p/adjusting-career-home-runs-allowed-hr9.html.

The following table lists the top 25 pitchers in MLB history in HR+,  for pitchers with at least 1000 IP through 2011, and also adjusted HRs allowed and HR/9 to the 1953 AL. Mariano ranks 4th - his HR+ of 228 indicates he was 128% better than league average at preventing HRs.

Pitcher	Debut	IP	HRs	HR/9	Expected HRs	HR+	Adjusted HRs	Adjusted HR/9
Corridon, Frank	1904	1216.0	7	0.052	18	262	37	0.273
Pfiester, Jack	1903	1067.3	6	0.051	15	251	34	0.285
Russell, Reb	1913	1291.7	7	0.049	17	244	42	0.294
Rivera, Mariano	1995	1211.3	65	0.483	148	228	42	0.313
Killian, Ed	1903	1598.3	9	0.051	20	219	58	0.327
Minton, Greg	1975	1130.7	43	0.342	93	216	42	0.331
Perranoski, Ron	1961	1174.7	50	0.383	105	210	44	0.340
May, Jakie	1917	1562.3	35	0.202	71	204	61	0.350
Wood, Joe	1908	1436.3	10	0.063	20	204	56	0.350
Rixey, Eppa	1912	4494.7	92	0.184	184	200	178	0.357
McDowell, Roger	1985	1050.0	50	0.429	96	191	44	0.374
Ruth, Babe	1914	1221.3	10	0.074	18	180	54	0.398
Tekulve, Kent	1974	1436.3	63	0.395	113	179	64	0.399
Veale, Bob	1962	1926.0	91	0.425	162	178	86	0.402
Joss, Addie	1902	2327.0	19	0.073	33	175	106	0.409
Forster, Terry	1971	1105.7	51	0.415	88	173	51	0.412
Quisenberry, Dan	1979	1043.3	59	0.509	102	172	48	0.415
Brown, Kevin	1986	3256.3	208	0.575	358	172	150	0.416
Donahue, Red	1893	1535.0	17	0.100	29	171	71	0.418
Plank, Eddie	1901	4495.7	41	0.082	70	170	210	0.420
Wight, Bill	1946	1563.0	74	0.426	125	169	73	0.422
Webb, Brandon	2003	1319.7	92	0.627	156	169	62	0.423
Lincecum, Tim	2007	1028.0	66	0.578	111	169	48	0.423
Richard, J.R.	1971	1606.0	73	0.409	123	169	76	0.423
Hollingsworth, Al	1935	1520.3	47	0.278	79	168	72	0.425

 I posted a spreadsheet listing these stats for all pitchers with at least 1000 IP - see https://docs.google.com/spreadsheet/ccc?key=0AkMu391-e5t0dDdlTjhaNlVpZWI1anBqVklWTlBLb3c.

There are many interesting things about the list. For example, apart from being exceptionally good at preventing HRs, what do Frank Corridon and Mariano Rivera have in common?




They both learned historically nasty pitches by accident:

• Frank Corridon is commonly considered to be the inventor of the spitball. The excellent book A Game of Inches: The Stories Behind the Innovations That Shaped Baseball: The Game on the Field (Volume 1) indicates Corridon was tossing with the catcher on a soggy day, and noticed that the wet ball was sharply breaking. This observation inspired Corridon to fiddle with various grips and moisture levels until he perfected the spitball.
• Mariano's cutter came into being in 1997. When he was playing catch with Ramiro Mendoza, the ball started cutting unexpectedly, even though Mariano was using a 4-seam fastball grip. The humorous part of this is apparently Mariano and pitching coach Mel Stottlemyre tried to straighten out the movement and restore Mariano's 4-seamer, since until Stottlemyre realized that having the ball break was more effective and said "You know, leave it like that."

Another name on the list, Jakie May, is the answer to a great trivia question. According to Jayson Starks, the only 2 pitchers to end their careers on a decision in the final game of a World Series are Jakie May and Pedro Martinez. We all remember Godzilla destroying Pedro a couple years back. Jakie May's last game, was also against the Bombers; in game 4 of the 1932 World Series May lost a 4-3 lead, giving up 6 runs in 3.1 innings, including a big 2-run single to Lou Gehrig and Babe Ruth's last World Series hit and RBI.

But let's get back to Mariano. What a great advantage to the Yankees to have had a closer with such versatile skills - control, power, and a historically great ability to keep the ball in the park.  Just compare him to Trevor Hoffman:

Pitcher	Debut	IP	HRs	HR/9	Expected HRs	HR+	Adjusted HRs	Adjusted HR/9
Rivera, Mariano	1995	1211.3	65	0.483	148	228	42	0.313
Hoffman, Trevor	1993	1089.3	100	0.826	124	124	70	0.577

Hoffman was stingy when it came to giving up HRs to batters other than Scott Brosius. Hoffman's .826 HR/9 was significantly below league average, and his HR+ ranks him 197th out of the 1037 pitchers with over 1000 IP.  Even so, Mariano is twice as good at preventing HRs.

This is where you can see the unique value of Mariano Rivera. I'd argue that if the Yankees had Hoffman instead of Mariano, the Yankees' odds of winning 5 World Championships since the mid-1990s would have decreased. For example,  having a pitcher twice as likely as Mo to give up a HR in the bottom of the 9th in game 5 of the 2000 ALDS or the top of the 9th in game 6 of the 2000 ALCS would obviously have lessened the Yankees' chances of getting to the World Series. This comparison means no disrespect towards Hoffman, who was a great closer;  rather, the comparison is just to highlight Mariano's relative greatness in terms of preventing home runs.

---

Correction: An earlier version of this post used a different adjustment method. I've since decided that HR+ is a better method than what I did back then.