Revisiting Bill James’ “Baseball’s Best Player”

There’s an old article in the Bill James Online archives where Bill tries to determine who the best player in baseball was for any given year. He used Win Shares and a four-year weighted average to calculate the top-five players for every year, beginning in 1904. I actually wrote about his article over a year ago, when the blog wasn’t even a blog. From that post:

Without going too deep into the method or the findings (that’s what Bill’s site is for, after all), I’ll say that Bill used a player’s four-year (weighted) average of Win Shares to see who was the best player over those past four years (in the “mirror” article, he looked ahead to the next four years, using real numbers of course, not predictions). The method seemed pretty sound to me, and gave some really fair and mostly expected results (Babe Ruth rules for a few years, then Gehrig, then Williams, etc, with Bonds ruling for much of the last 10-12 years, even before BALCO).

What makes the list most interesting, I think, are those years between the true greats, when the historically great players are either too old or too young to have dominated the league for three straight years. The players who emerge during these times as the “best player in baseball” make you pause for a minute but, upon further reflection, seem sensible. These players include: Ron Santo in 1967; Bobby Murcer in ’72; Dave Parker in ’78-’79; Tim Raines in ’87; and Will Clark in ’89-’91.

I still find myself thinking back to that article when I hear people debate the best player of a given era (such as Cyril Morong’s look at the mid-1960s and Ron Santo from last week). It’s such a simple and effective way to measure the top players in a short timeframe that I find myself drawn to it. The problem, though, is that, with its reliance on Win Shares, the article has quickly become dated. Wins Above Replacement, or WAR, does a much better job of quantifying a players full worth and, as such, has become the preferred way to measure something like this (as you can see by the numerous WAR articles I’ve written recently).

But we can fix that. With the Rally WAR database and a little bit of SQL, I was able to replicate Bill’s process for every year between 1904 (when we first have four year’s worth of data available) and 2009. The best part of the data is how easily it breaks the last century down into different eras, by the dominant players of the time. You can click here to view the five best players for every year from 1904-2009, or read below for the breakdown by era.

1904 – 1919: The Early Greats


Click to enlarge.

Aside from Hall of Famer Joe McGinnity sitting in the number one spot in 1904, there isn’t a single year where Honus Wagner, Ty Cobb, or Walter Johnson isn’t considered the best player in baseball in this early era (Eddie Collins is tied for first with Walter Johnson in 1915). Which sounds just about right to me. Others finding themselves in the top five in this era are: Nap Lajoie, Christy Mathewson, Rube Waddell, Ed Walsh, Pete Alexander, and Tris Speaker. If you look close, you’ll see pitcher Babe Ruth creeping up the list in those last couple of years.

Oh, and a note about the charts… Sorry for not being to give you something a little more flashy. When it comes to charting data, I’m kind of stuck with my old copy of Excel, and it’s not too great when it comes to this kind of thing. Still, I think this does a pretty good job of showing just how much better the top one or two players were compared to the rest of the list from year to year. The points are coded by rank, with the #1 ranked player always a red dot, the #2 always a blue, and so on. If anyone has any better suggestions, please let me know.

(Click “Read More” to continue reading.)

1920 – 1937: Ruth, Hornsby, and Gehrig


Click to enlarge.

And 1920 is where Babe Ruth takes over as the best player in baseball. Remember, this ranking uses a weighted average over the player’s last four years, with the most recent year being weighted the most, and three years ago weighted the least (on a 40-30-20-10 weighting). In Ruth’s case, then, it wasn’t until he was able to get two full years in as a batter that his average pulled him above Cobb and Johnson. And even in 1922, when he had a weak year due to injury, his ’20 and ’21 seasons keep him atop the charts alongside Rogers Hornsby in the weighted average. Hornsby does pull ahead in ’25, though, when Ruth plays only 98 games. Lou Gehrig, Jimmie Foxx, and Charlie Gehringer fight it out for the top spot as the era ends, though Gehrig wins every year. It’s only as Gehrig’s health starts to deteriorate that he falls of the charts. Some others who make repeat appearances during this era are George Sisler, Lefty Grove, Tris Speaker, and Al Simmons.

1938 – 1955: Pre- and Post-War


Click to enlarge.

This chart provides a pretty good picture of just how badly the war affected the major leagues. Not only is it the first era that isn’t clearly dominated by two or three great players (on a year-to-year basis), but the few players who are atop the ranks clearly aren’t as great as the other eras. Just look at the peaks of this chart compared to the peaks of, say, the Ruth/Hornsby/Gehrig chart. Only Ted Williams’ weighted 1942 and 1949 seasons would even compete for best player in baseball in the previous era. Most of the names who do float to the top in this time, though – DiMaggio, Feller, Williams, Musial, Robinson – are generally listed among the all-time greats. All of the others, save Charlie Keller, are still Hall of Famers, just not quite as revered. The Mick makes a cameo in the #2 spot in 1955.

1956 – 1972: Mantle, Mays, and Others


Click to enlarge.

Mantle makes a cameo on that last chart, but he flat-out dominates this one – at least for those first few years. His weighted-’57 and ’58 campaigns are in the 11-12 WAR range – the only two other players to have a weighted average of 11+ WAR in any given season are Babe Ruth and Barry Bonds. It’s safe to say that few peaked as highly as Mantle. But for as high as the Mick peaked, Willie Mays makes a pretty strong case for being the top player of this era, with his seven consecutive years atop the list and his five consecutive years of 9.5+ WAR. The era ends with a five different players wearing the crown in six years: Ron Santo, Carl Yastrzemski, Bob Gibson, Fergie Jenkins, and Steve Carlton, with Gibson the only repeat winner. Other players who make a strong showing during this era are: Duke Snider, Ernie Banks, Eddie Matthews, Hank Aaron, Sandy Koufax, and Juan Marichal.

1973 – 1990: The Age of the Infielders


Click to enlarge.

Joe Morgan takes over as the best player in baseball in 1973. His four-year run ends after two great years in ’75 and ’76 that find him as far above the second best player as anyone else is history. The 1977 crown goes to Mike Schmidt, who beats Rod Carew and Morgan by barely 0.05 WAR. After a two-year stint that finds Phil Niekro atop the charts, the crown spends the next 11 years among some of the best shortstops and third-basemen of all time: Schmidt, George Brett, Robin Yount, and Wade Boggs. Only Rickey Henderson’s ’85 and ’90 seasons keep it from being a clean sweep by the left-side of the infield. Some other names on the list during this time are: Tom Seaver, Gaylord Perry, Andre Dawson, Cal Ripken, Tony Gwynn, and Roger Clemens. As with Mantle and Ruth before him, Barry Bonds makes a brief appearance as the second-best player in the final year of the era.

1991 – 2009: Bonds and Pujols

Click to enlarge.

Which brings us to the modern era. I know it’s almost trite to say these days, but Barry Bonds was easily a Hall of Fame player before 2001, and this chart is a perfect illustration of that. From 1991 through 1998, there was no better player in baseball, and it was hardly even close. Cal Ripken had a case in 1991, and Greg Maddux was only 0.02 WAR behind Bonds in ’95, but that eight-year stretch belonged to no one but Barry Bonds. It was only an injury in 1999 that held Bonds to 100 games that kept him from owning the whole decade. Jeff Bagwell and Pedro Martinez filled the void in those years, but it wasn’t long before Bonds found himself back on top. The three-year stretch from 2002 to 2004 is historic.

When Bonds finally relinquishes the crown for good in 2005, it’s to none other than Albert Pujols, who hasn’t lost it yet. It is pretty interesting to note, though, that Pujols’ top weighted average so far is 9.05, a number that Bonds topped three different times in the 1990s alone. For as great as Albert has been these past nine years, he’s never skyrocketed above everyone else in WAR. That may be a product of his being a first-baseman (having to overcome the big positional debit in the WAR calculations), but it’s also likely a product of his sheer consistency (putting up six consecutive years of WAR values between 8.3 and 9.6, with his best season of 10.9 WAR in 2003). Other players who find themselves among the top-five players of the last decade are Ken Griffey, Alex Rodriguez, Roger Clemens, Randy Johnson, Jason Giambi, Sammy Sosa, Jim Edmonds, and Joe Mauer.

I don’t know if this is a perfect metric – I’m sure there are plenty of people out there who can tell me why a weighted average of the last four years isn’t the best way to judge the best player in baseball – but it’s simplicity and sensibility really appeals to me. It shows us what we already knew about the greatest players of all-time, but it also gives us some good insight into a few other players, like Ron Santo, Phil Niekro, and Jeff Bagwell. Bill James also did a variation of this, where instead of weighting the last four years in descending order, he weighted the next four years – effectively telling you who the best player in baseball over the next four years will be. I have the data run for that as well, but I have yet to make the charts for it. If others would like to see that, I can probably find some time to make the charts. For most of the century, the names remain the same. There are a few years, though, where someone new shows up at the top (for example, Cal Ripken takes over the top spot in 1983 when you weigh his next four seasons instead of his last four).

One last thing, because I’m sure some people are going to look at the lists here and see that A-Rod never made it to the top. I said this in my initial post, and I think it’s worth repeating here:

It’s important to note, however, how A-Rod has never been able to bust through the reigns of Barry Bonds and Albert Pujols as the best player in baseball. He’s always 2nd or 3rd, but Bonds was putting up insane numbers when A-Rod was in his 20s, and now Pujols is putting up insane numbers as A-Rod plays in his 30s. The comment I made on the article itself: “Using these numbers, it’s clear that A-Rod had the great misfortune of playing his early years opposite Barry Bonds and his later years opposite Albert Pujols, two of the greatest talents since Mantle & Mays. Does the fact that Hank Aaron never controlled the crown before he was 35 keep him out of the conversation of greatest player of his generation/all-time? Of course not, and I don’t think A-Rod constantly finishing just behind Bonds and Pujols should mean anything different.”

Let me know what you think.

Larry Granillo

About Larry Granillo

Larry Granillo has been writing Wezen Ball since 2008 and has dealt with such touchy topics as Charlie Brown's baseball stats and Ferris Bueller's day off. In 2010, he got the bright idea to time every home run trot in baseball; he has been missing ever since.