Babe Ruth's aging vs. Barry Bonds' aging

I recently read the book Cooperstown Confidential by Zev Chafets. 

On pages 184-185 it has

"Cole and Stigler (two academics) pointed out that Babe Ruth hit 198 homers in the last six years of his twenty-two year career, 28 percent of his career total of dingers. In the last six years of his career, Barry Bonds hit 195, or 26 percent. "There is no convincing way," the study said, "to demonstrate that Bonds' performance owed to drugs more than Ruth's did to his prodigious use of alcohol and tobacco." Which, of course, was nothing."

The first problem is that the last six years of Bonds' career were the years 2002-2007. That leaves out his 73 home runs in 2001 (all stats I present here come from Baseball Reference).

If we go by the last 7 years of his career, Bonds hit 268 home runs. That is 35.17% of his career total (268/762 = .3517). For Ruth, in the last 7 years of his career, he hit 244 home runs or 34.17% of his career total (244/714 = .3417).

This is still close, but Bonds has over taken Ruth just with this one adjustment.

Now Ruth did not hit many home runs before age 24 because he was mainly a pitcher early in his career. This holds down his career total. Other wise, he would have larger denominator and therefore an even lower percentage for his last 7 years.

I don't know how the numbers would change. Those early years in Ruth's career were part of the dead ball era, so even if he had been a full-time outfielder from ages 20-23, it might not have been many HRs.

But, from age 24 until the end of his career, Ruth his 694 HRs. So 244/694 = .3516. In his last 7 years, Ruth his 35.16% of his age 24-40 HRs.

What about Bonds? Bonds hit 697 HRs from age 24-42 (the end of his career). So 268/697 = .3845. In his last 7 years, Bonds his 38.45% of his age 24-42 HRs. The gap in favor of Bonds has grown a bit. But they are still close

One thing that holds Bonds' HR totals in his last 7 years were all the walks he got. He drew 1,011 walks in those 7 years including a record 232 in 2002, of which 120 were intentional (the most walks Ruth ever had in one season was 170).

In Ruth's last 7 seasons, he drew 704 walks. So Bonds has an extra 307. That is 307 times he had less of a chance or no chance to hit a HR. 368 of Bonds' 1011 walks were intentional. Baseball Reference shows 26 for Ruth, but I am not sure how much we know about IBBs in those years.

But just to have an idea of how much the extra walks mattered, with 368 being intentional and Bonds having .112 HRs per AB in his last 7 years, that projects to an extra 41 HRs. Even we use the gap of 307 walks, that would be an extra 34 HRs.

What this suggest to me is we should look at HR%. The table below shows Bonds' HR% each year from age 24-39 along with the league average. The last column is his relative HR% or his HR% divided by the league average.

Bonds' Age	HR%	Lg HR%	Rel HR%
24	0.0328	0.0207	1.5795
25	0.0636	0.0231	2.7577
26	0.0490	0.0219	2.2407
27	0.0719	0.0192	3.7449
28	0.0853	0.0252	3.3810
29	0.0946	0.0278	3.4015
30	0.0652	0.0278	2.3491
31	0.0812	0.0286	2.8437
32	0.0752	0.0280	2.6837
33	0.0670	0.0289	2.3179
34	0.0958	0.0325	2.9468
35	0.1021	0.0339	3.0147
36	0.1534	0.0335	4.5769
37	0.1141	0.0296	3.8617
38	0.1154	0.0306	3.7677
39	0.1206	0.0321	3.7567

You can clearly see something happen at age 36 (and perhaps as early as age 34). His Rel HR% jumps from 3.01 at age 35 to 4.57 at age 36, by far the highest of his career. After 5 years in a row with a Rel HR% under 3 from ages 30-34 (and well below 3 in 4 of those years), it starts rising slightly. But even after the very large increase at age 36, his Rel HR% from ages 37-39 are all very high. The only year before age 36 in their range was when he was 27.

The Giants play in what is now called Oracle Park. It opened in the year 2000 when Bonds was 35. The RF wall is 24 feet high. It was only 309 feet down the line when Bonds played but right-center field was 421. So it is not an easy place for a lefty (which Bonds was) to hit a HR.

From Bill James Handbooks, the HR rating for the park for lefties in 2001 was 57, meaning lefties had a HR rate of only 57% of what it was in the rest of the parks in the league. Over the years 2002-04 it was 75. The years 2001-04 cover the years when Bonds was 36-39 years old.

So Bonds still had this great HR% surge despite hitting in a tough park for lefties.

Now the same table for Ruth

Ruth's Age	HR%	Lg HR%	Rel HR%
24	0.0671	0.0064	10.4577
25	0.1179	0.0088	13.4142
26	0.1093	0.0111	9.8026
27	0.0862	0.0124	6.9452
28	0.0785	0.0105	7.4663
29	0.0870	0.0094	9.2487
30	0.0696	0.0125	5.5623
31	0.0949	0.0102	9.3501
32	0.1111	0.0104	10.6598
33	0.1007	0.0115	8.7906
34	0.0922	0.0141	6.5350
35	0.0946	0.0157	6.0268
36	0.0861	0.0132	6.5314
37	0.0897	0.0163	5.5097
38	0.0741	0.0142	5.2063
39	0.0603	0.0160	3.7609

I don't put much stock in those sky high Rel HR%s when he was younger. The dead ball era had just ended and maybe not many other guys had figured out how to hit like Ruth to generate so many HRs (or maybe they were still being conservative and swinging for line drives-it might not be easy to change batting styles very quickly). That would make the league average HR% pretty low and boost Ruth's Rel HR%.

But look at Ruth when he was 32 (that was 1927 when he hit 60 HRs). His Rel HR% was 10.65. But then it dropped 3 straight years with a slight increase at age 36 (but a far lower increase than Bonds saw at age 36). Then it falls quite a bit over ages 37-39. Those years are well below what it was for ages 31-33. But for Bonds, his Rel HR%s from ages 37-39 are much higher than for ages 31-33.

Bonds' Rel HR%'s from age 37-39 are very high compared to most of his career while for Ruth they are pretty low compared to his earlier years.

So it looks like Bonds had an unusual aging pattern in his late 30s. Perhaps very unusual.

Some stats on how pitchers have fewer IP than about 10 years ago

I used Stathead to search this.

From 2011-13, 53 pitchers had 486+ IP & 22 had 600+ (2 had 700+). Verlander led with 707. 

From 2021-23, only 25 pitchers had 486+ IP and only one had 600+ (Alcantara, 619). About a 52.8% drop in guys with 486+ IP.

12 pitchers from 2011-13 were above 619. Cole was 2nd from 2021-23 with 591. That would have been 28th from 2011-13.

I used 486 IP since that is 3 times 162, what it takes to qualify for the ERA title for just one season. So I am using 486 as a qualifier for a 3 year period.

For position players I used 3*502 = 1,516 PAs since 502 is usually the number of PAs it takes to qualify for the batting title for just one year.

From 2011-13, there were 104 guys with 1,516+ PAs. From 2021-23, it was 94. That is a 9.6% drop, far less than what happened for pitchers.

At the 2,000 PA level, it fell from 15 to 10. That is a much smaller drop than it was for pitchers with 600+ IP.

The highest anyone had from 2011-13 was 2,111 (both Alex Gordon and Starlin Castro). Two players from 2021-2023 actually exceeded that: Marcus Semien (2,201) & Freddie Freeman (2,133).

At the 1,600 PA level, there were exactly 82 guys in each time period. So it sure looks like something different has happened with the pitchers.

Update April 14:

Some data on pitcher games started. Number of guys with 90+ & 75+ starts.

2011-13:

90+) 41
75+) 64

2021-23:

90+) 19
75+) 46

Norm Cash's 1961 season

He won the AL batting title that year with a .361 AVG. Yet he never hit .300 or higher again and his lifetime avg was just .271 (all data is from Baseball Reference and Stathead).

His OBP that year was .487. His career OBP was .374 and his next best was .402 (in 1960 in only 428 plate appearances).

His SLG was .662. His next best was .531 and it was .488 for his career.

His OPS+ was 201 that year and his next highest was 149. Lifetime it was 139.

I wondered if his flukiness was balanced against both lefties and righties.

This table shows his OPS vs. righties relative to lefties for each of his 14 full or close to full seasons

1960          1.50
1961          1.57
1962          1.43
1963          1.36
1964          1.54
1965          1.15
1966          0.94
1967          1.31
1968          0.97
1969          1.56
1970          1.21
1971          1.27
1972          2.11
1973          2.39

The 1.57 in 1961 is the highest until late in his career when his performance against lefties went down quite a bit. But it was just a bit higher than 1960, 1964 and 1969. So this does not indicate a great imbalance.

But, I also calculated his OPS vs. righties relative to the league average of all left-handed batters vs. righites (and the same was also done for vs. lefties).

Here is his year-by-year OPS vs. righties relative to the league average of all left-handed batters vs. righties:

1960          1.21
1961          1.62
1962          1.25
1963          1.26
1964          1.19
1965          1.24
1966          1.15
1967          1.22
1968          1.20
1969          1.24
1970          1.13
1971          1.30
1972          1.24
1973          1.18

The ratio in 1961 is by far the highest at 1.62 with the next best being 1.30. So a great year for him vs. righties.

Now for his year-by-year OPS vs. lefties relative to the league average of all left-handed batters vs. lefties:  

1960          1.02
1961          1.21
1962          1.04
1963          1.03
1964          0.94
1965          1.30
1966          1.39
1967          1.08
1968          1.44
1969          0.92
1970          1.16
1971          1.23
1972          0.65
1973          0.55 

He had 1.21 in 1961, but that is only his 4th highest ratio. He had four that were higher: 1.44, 1.39, 1.30 and 1.23.

So he had, compared to the rest of his career, a fantastic season against righties. But against lefties, it was just good.

Update March 25: From 1960-73, Cash had an OPS of .918 vs. righties while all left-handed batters had .733. His ratio is 1.25 (.918/.733). So his 1.62 ratio in 1961 was far above this.

Over the same period, his OPS vs. lefties was .696 while all all left-handed batters had .625. This ratio is 1.11 (.696/.625). His 1.21 ratio from 1961 was only slightly above this.

I only looked at 1960-73 since he did not get many PAs in 1958, 1959 & 1973.

There can be some idiosyncratic things going on here. For example, 20 of his 162 PAs vs. lefties in 1961 were against Whitey Ford. He had just a .417 OPS vs. him that year.

Cash only had 52 career PAs against Ford. So it is possible that 1961 was an unusually tough year for him in terms of the quality of the lefties he faced. But to conclude that would require looking each of his seasons to see who he faced. Also, in 1961, Ford seems to be the only good lefty that he faced fairly often.

Interesting new article by Bill James: The Competitive Advantage of the Pitcher’s Park

It is in the latest issue of By the Numbers: The Newsletter of the SABR Statistical Analysis Committee, edited by Phil Birnbaum.

Click here to read it.

Here is a synopsis from Phil:

"Bill James finds that teams who play in pitcher's parks have had better records, historically, than teams who play in hitter's parks. He presents the data showing the effect, and then offers a suggestion for why this may be happening."

Also in the issue, Charlie Pavitt reviews several recent studies from the academic literature.

Factors that might influence the difference between ERA and FIP

My last post mentioned that Aaron Bummer had a 6.79 ERA last year while his FIP ERA was 3.58 for a differential of -3.21 (FIP - ERA). That was the largest absolute differential last year for any pitcher with 50+ IP and it was 0.64 larger than the next largest.

So to look at what might explain why ERA differs from FIP (fielding independent ERA estimated using SOs, BBs and HRs), I ran a regression with FIP - ERA as the dependent variable and the following three independent variables:

SLG Diff (a pitcher's SLG allowed with runners on base minus the SLG they allowed with no runners on)

BAbip (the batting average a pitcher allows on balls in play, so it is dependent on how good his fielders are)

BQS/9 (Bequeathed runners that scored per 9 IP).

Bequeathed runners represents the number of runners left on base by a pitcher when that pitcher leaves the game. Any bequeathed runner who scores an earned run after a pitcher has left the game will be counted against that pitcher's ERA (from mlb.com https://www.mlb.com/glossary/advanced-stats/bequeathed-runners).

 

I used SLG Diff because some pitchers might have gotten hit pretty hard when they had runners on base, making their ERA higher than what we might otherwise expect based on their overall numbers.

I used BAbip because this is not controlled very much by the pitcher. A guy can have a low FIP but if his fielders can't catch the ball, his ERA will be high.

I used BQS/9 because a pitcher cannot control what happens after he leaves the game. Some pitchers get lucky and their bullpen bails them out. For others, it is the opposite.

I looked at all the guys who had 100+ IP last year. All data came from Baseball Reference and Stathead. There were 127 pitchers.

Here is the regression equation:

FIPERADIFF = .789*BQS/9 + 13.63*BAbip + 2.5*SLGDiff - 4.32

r-squared = .639, so 63.9% of the variance in the dependent variable is explained by the model.

standard error = .33

Here are the t-values for the three independent variables:

 

BQS/9)  6.7 (The p-value is < .00001)
BAbip)  11.6 (The p-value is < .00001)
SLG Diff)  5.04 (The p-value is < .00001)

The r-squared seems fairly high but it still means that 36.1% of the variation in FIPERADIFF is not explained.

The standard error seems high. I wish it was lower. The average absolute differential was about .44.

The t-values are all pretty high so each independent variable is significant. I used a website that converts t-values into p-values.

There may be some other variable that I should include. Maybe I could find the estimated FIPERADIFF for each guy and look at the 10 or so guys with the biggest differences between the estimated value and the actual (FIP - ERA). Maybe something that would be obvious to include would pop up.