Bill James on Guidry and Rice

Plus — the origins of Runs Created... and WAR

Bill James on Guidry and Rice

If there was one article I really wish I could read in full, this one would be it. I haven’t been able to find a source anywhere for this one, either online or in compilations of James’ prior work.

Unfortunately, unless and until I can find a reprinted copy of The 1979 Bill James Baseball Abstract, we’re going to have to stick with Rich Lederer’s summary.

Apologies for the small font size — and, yeah, this is one reason why I really wish I had the original article instead of a web summary.

This short excerpt is apparently the first time that James brings up the concept of runs created. The idea here is pretty simple. Hitters are trying to use their abilities to create runs, and the things that we traditionally measure in baseball — singles, doubles, triples, home runs, walks, times hit by pitch, stolen bases, bunts — are all ingredients used to create those runs.

The conventional wisdom at that time — and for most of baseball history — was that batting average was the ultimate measure of offensive success. Home runs and power hitting was useful, but most baseball analysts and fans through the ages would take a man who hit .350 or .375 or even .400 above a guy who hit .270 with a lot of power.

“Advancement percentage” is a stat I’ve never heard of, and was clearly dropped at some point in time. However, the concept is important to understand when you look at James’ first runs created formula:

It’s pretty easy to understand. James used at bats plus walks as a quick estimate of plate appearances (which wasn’t an official stat at the time), and added in caught stealing to the denominator.

The numerator consists of hits plus walks, which is the numerator for on base percentage. James subtracted times caught stealing from that total — which is where my first question comes in. Why punish the player for being caught stealing twice in the formula?

Then James multiplies that total by total bases after adding in 70% of stolen bases. I’ve got no idea where that 70% comes from or what the theory is behind this. This is one of those Jamesian arbitrary decisions, one of those things that come out of left field that “just work” and that his followers need to accept on faith.

This is different, by the way, than the linear weights approach that Pete Palmer would later take in The Hidden Game of Baseball. Palmer’s approach was to actually measure how important each of these ingredients is to run creation, using play by play data from World Series games (this was long before Retrosheet was a thing). This is the same approach Tom Tango used in The Book, which actually validates a lot of Palmer’s preliminary findings.

We’ll talk about this later — but, in hindsight, it’s always struck me that Palmer’s work wound up being more significant than James’ work on this subject. Palmer’s approach to weighing offensive statistics led directly to WAR, while Runs Created and its arbitrary weighting had been largely ignored over the last 20 years or so.

Anyway, back to James.

Okay — this is the part where the concept of wins above replacement starts to come into play.

James calculates Rice as creating 70 runs above average as a hitter. Note that James made no attempt to calculate how effective Rice was as a fielder, and that James didn’t use any advanced calculation for pitching. He basically concluded that all 61 of those runs that Guidry gave up were his own responsibility, and concluded that Rice was the superior player.

And then he adds in more.

This is where I really wish I had a scan of the original two articles.

It’s not clear to me that James actually figured out a way to incorporate park effects in his calculation. It’s good that he had the thought — but it’s hard to tell from this summary just how important it was in his thinking at the time.

The argument that “defense is probably 70% pitching and 30% defensive play” is another arbitrary statement with no real basis in reality. I’m not sure what the sabermetricans currently think, though I would argue that pitching is actually mostly defense, not the other way around. This shows, however, how traditional baseball thinking was to treat defense as a given and focus instead on the performance of the individual pitcher.

It’s also interesting to see that James continued to claim that Rice was the rightful winner of the AL MVP in Win Shares.

We’ll take a closer look at that AL MVP race tomorrow.