Of course, Hank moved from a terrible home run environment to a great one in the middle of his career. His 2 year road totals:

Years	HRs
57-58	46
58-59	39
59-60	38
60-61	34
61-62	42
62-63	52
63-64	38
65-66	36
66-67	39
67-68	30
68-69	35
69-70	38
70-71	31
71-72	31
72-73	31
73-74	25

Aaron was 27-29 when he reached his peak, but maintained it exceptionally well. 

Barry's peak was a little later, you might say...he hit 63 homers on the road at age 37-38, and 32 homers on the road at age 27-28, during his great years with Pittsburgh.

I will expose my vast ignorance about stats with this question but why are the totals given for two years, with each year repeated? I understand it as it relates to hockey, since each season flows over two years. But in the case of baseball why can't the road totals just be given for each year?

Be patient with me.

Hi Flex: some analysts use two-year averages to make the sample sizes larger and even out the luck effects a bit.

As mentioned there's the issue of sample size, but using two year road/home totals have the advantage of being roughly comparable to full season statistics, so they can, at a glance, be mentally compared to one's internal notions of "a lot" versus "a few" HR.

How does using two year totals make the sample size bigger?   I like the notion that they're comparable to seasonal statistics.

Without the scale on the axes, those graphs aren't telling us anything. The scales are obviously not the same (different number of scale lines), and without the numeric reference, mean nothing.  As a statistician, I call foul.  If you want to compare the two visually and get a better idea as to why Bonds sticks out like a sore thumb, I suggest plotting both on the same plot. Then Aaron (1/20 to start, 1/10 at peak) looks almost like random noise jittering around in the middle of Bonds' massive upward trend (1/25 to start, peak at 1/7).

Deliberately mis-using graphics like this...it makes me sad.

Deliberately mis-using graphics like this...it makes me sad.

I believe the scale was removed intentionally, and the purpose was just to show the HR trend by age. The perception was that Aaron followed the "accepted" age trend, while Bonds had a moderately impressive level for much of his career, and drastically changed his HR rate at a late career stage.

The graphs don't really support that perception, although there may be various reasons for that.

Don't be sad.

Yeah, I know the scales were removed on purpose, but I still don't see the two plots as all that similar (although I do spend days staring at such things, and picking out trends). I guess I'm just annoyed by the comparison.

Firstly, for the all-time HR champ, we should expect at least a respectable late career plateau...if the player didn't maintain HR hitting ability, they wouldn't be the all-time champ. the ability to hit for power is sometimes the last thing to go. McGwire still hit HR (and only HR) at a high rate when he left. Ruth was much the same, as was Williams (although he kept up better overall hitting as well). Hitters like Mays, Robinson, and Killebrew followed what look like more "typical" aging patterns, which is why they didn't threaten the all-time HR records.

Secondly, Aaron's HR rate was helped by a bit of a platoon late in his career on top of the HR friendly home field, where it appears he faces a disproportionate number of lefties (that or the number of lefties suddenly increased in the early 70's). 

Thirdly, the new plateau late in Aaron's career also owes a lot to the emergence from the late 60's heavy pitching era (1969 expansion and rules adjustments increased scoring by half a run from 68 to 69).

Now, for Bonds, you can also list several reasons why his rate kept increasing in his late 30's, but most of those involve steroid/other enhancer use (overall HR totals in the league increasing, personal HR totals). Bonds even moved into a killer park for lefty homeruns, and still jumped way up. I still don't think Bonds should necessarily be ripped that strongly for merely being far and away the best at exploiting enhancing drugs (among hitters at least), when it's clear many others were as well, but the HR rate increases for Aaron and Bonds are really apples and oranges. I'm sure Magpie is aware of these things (Magpie has a better grasp of baseball history than most, me included), which is OK. Occasional stirring of the pot by the article writers isn't a bad thing.

Deliberately mis-using graphics like this...it makes me sad.

Yup, I was messing with you all. And the scales were painstakingly (well, not that painstakingly) removed with PhotoShop. And although the overall shape of the two players' development is surprisingly similar, anyone can figure out with a careful look, that the first one is Aaron and the second one is Bonds. (That 73 HR year sticks out rather dramatically, and I think it has affected our perception of Bonds in many, many ways.)

For the longest time, Aaron was the only significant slugger in major league history whose biggest run of HR production began at age 35. He's not the only one anymore, of course. Soon after the Braves moved from the unfriendly confines of County Stadium into the Launching Pad in Atlanta, Aaron made a conscious decision to narrow his focus and become a home run hitter. He thought he had a shot at Ruth's record.

Aaron is the guy who led the NL in HRs per AB for the first time at age 37, in 1971 (and again in 1972 and 1973).  Bonds had already led the league in HR per AB three times (1992, 1993, 1996) before his own late career HR explosion.

Bonds was obviously always regarded as an elite player, the best in the game from 1992 onwards - but I don't think anyone saw him as a threat to Ruth or Aaron's numbers. There were better contenders - Griffey, obviously, and for a while McGwire and even Canseco. Because Bonds has been walked in at least 20% of his plate appearances since the early 1990s, the fact that he was an elite power hitter was actually overlooked.

Aaron spent most of his career in the shadow of two players who seemed much better candidates to overtake Ruth. Mays and Mantle both had a three year head start on Aaron - while Aaron was actually keeping pace with both, it really went unnoticed at the time. (Mays and Mantle were both born in 1931 and arrived in the majors at age 20, in 1951. BY the time they were 34, Mays had 505 HRs, and Mantle 496. By this time, Mantle's career was obviously on borrowed time, but Mays was coming off a 52 homer season.

Aaron, with 398 HRs at the time, was an after thought. But he was three years younger.

Anyway, Mantle struggled through two more pain-racked seasons before retiring at age 36. Mays had one more year as a top home run producer, and then strangely lost a large chunk of his home run power. He had five more years left as a very effective regular, but he hit "only" 104 homers in those five years (age 36-40). Very strange. Aaron hit 179 homers in his age 36-40 seasons.

One of the great what-ifs in baseball history: suppose Willie Mays hadn't gone into the army in 1952?

Mays was called up by the Giants at the end of May in 1951 and hit 20 HRs in 121 games as a 20 year old. But he got to play just 34 games in 1952 and none at all in 1953. He hit 41 homers when he returned in 1954.

If he hadn't been drafted, I have to think he'd have ended up with about 725-730 homers, and would have broken Ruth's record in early 1971. Only to be overtaken by Aaron a few years later...

Incidentally, there is an extremely interesting piece by a writer named Michael Witte on the role of Bonds' notorious body armour. While everyone concedes that Bonds' unique armament (he is allowed to use it by way of a grandfather clause, the same way Dave Winfield wore a helmet without an ear flap) allows him to hang fearlessly over the plate and wallop the outside pitch, Witte says it also, among other things:

....allows Bonds to release his front arm on the same plane during every swing.

...The apparatus locks at the elbow... [and]   forms a rigid front arm fulcrum that allows extraordinary, maximally efficient explosion of the levers of Bonds' wrists.

...the weight of the apparatus helps to seal his inner upper arm to his torso at impact. Thus "connected," he automatically hits the ball with the weight of his entire body...

He notes how Bonds hasn't done all that well in Home Run Derbies, when there would be no good reason to wear the device, and also tracks its evolution over the last fifteen years from a one-piece forearm guard to a jointed two-piece elbow model, that grew larger and sprouted a flap that locked the two pieces together when the arm was extended....

A fascinating bit of work, I recommend it highly. Hat tip to Rob Neyer for bringing it to my attention.

You know, if I were Barry Bonds, I'd hate everyone too. No one can give this guy a break. He could save some kids from drowning and all anyone would say is that it was because he was juiced which means his efforts are tarnished.

Yup, I was messing with you all. And the scales were painstakingly (well, not that painstakingly) removed with PhotoShop. And although the overall shape of the two players' development is surprisingly similar, anyone can figure out with a careful look, that the first one is Aaron and the second one is Bonds. (That 73 HR year sticks out rather dramatically, and I think it has affected our perception of Bonds in many, many ways.)

Actually Magpie, I don't think it is just the 73 HR year that affects our perception of Bonds.  Aaron might have had a late surge in his career, but it wasn't near the dramatic scale that Barry has had.  Not to mention, even you can admit with no statistical backing that Aaron was seeing a heck of a lot more pitches to hit.  The Homerun record would probably be forever out of reach if Barry was getting the same kind of pitches Hank was.  The slugging % surge was another indicator for sure.  Someone also mentioned that Aaron also moved to a much more favorable hitting environment during his later years.

The thing is all these things can be accounted for logically, it isn't as if Barry moved to Coors and played Sandwiched beside Helton/Walker.

I'd like to oblige but I've completely forgotten how to insert a graphic into a comment. Help?

It all came back to me! So let me make some more pretty pictures and put everyone on the same scale this time. But it's not like I want to do a lot of work. Happily, I still have the big spreadsheet I built last summer when I looked at all-time HR leaders and which active players seemed likely to to claw their way into the upper ranks. The bottom line - Seasons - in each case begins at age 20. I would have liked to actually have the ages in that place, but my Excel skills... not the best.

 I think this is what we expect to see.
 

This is what we don't expect:

 

And here is Henry and Barry

 

I do think the 73 HR season sticks out just a little. But is it possible that 2001 was just a really weird year in the NL? Luis Gonzales with 57 HR? Shawn Green and Todd Helton with 49? Phil Nevin - Phil Nevin? - with 41?

Were there not some "juiced" baseball rumours going around at that time, which major league baseball catigorically denied? I mean if the fans got all hot and bothered with guys hitting 60+ wouldn't they line up en-masse to see 70 or 80?

After all, it happened in the 30's and worked pretty good for awhile unless you were a pitcher.

Looking at NL HR totals, 2001 doesn't look that odd. It does have the 2nd highest number of HRs in the past 30 years but then, 2000 has the highest and the 1999, 2004 and 2006 totals are all within about a hundred or so HR of the 2001 total. So in terms of total HR, 2001 doesn't stick out as an extreme outlier.

Re: Tellez, I hope the team isn't worried about years of control for a 24 year-old first basemen.