Over the weekend, I got the following email from a reader, entitled “WAR, wRC+, and Iso”:
I’m one of the “old guys.” The Yankees have been my team since 1960. I’ve used all the same measures since then. My take on the Trout-Cabrera conversation is different from my friends. I “see” Trout (I live in Southern California), and all he does and my instincts are he’s the best guy and if I was starting a team I’d choose him. I want to learn more about all the metrics you use. It’s hard for me to digest unless I see them when two players are being compared. Help me out here, whats the best way to get a handle on this.
On an impersonal level, I’d go to the following resources: the FanGraphs Library and RAB’s Guide to Stats. On a more personal level, let’s do as the reader requested and compare two players side by side. Instead of rehashing Cabrera/Trout, though, I’ll do two Yankee players side by side.
Before we get to that, though, I’ll simply define those stats–WAR/wRC+ (and by extension, wOBA)/Iso in ways that I’d explain them in person, to someone who asked me about them. We’ll start with Iso. Iso, quite simply, is a measure of ‘pure’ power in statistical form. You arrive at it by subtracting batting average from slugging percentage. This takes out singles and only counts extra base power. The closer the player is to a .200 Iso, the better; think a good SLG (.500) minus a good BA (.300). The Yankees’ leader in Iso (min. 500 PA) this year was Curtis Granderson, who put up and Iso of .260 (.492 SLG-.232 BA). The trailer was Derek Jeter with a .113 Iso (.429 SLG – .316 BA). So while Derek Jeter got a lot more hits than Granderson, Granderson ‘made his hits count,’ so to speak, by having them go for extra bases a lot more.
Next, there’s wOBA. wOBA is most definitely my ‘go to’ stat for hitters. Put simply, it’s an “all encompassing” stat that takes just about every offensive component that goes into a player and spits it out in a simple number on the same scale as on-base percentage. On a more complicated level, it assigns a run value to the different outcomes at the plate (a HR is worth x; a strikeout is worth -y, etc.). For a more detailed primer of wOBA, click here. wOBA’s wide “range” is great, but it isn’t park/league adjusted. That’s where wRC+ comes in. wRC+ is wOBA adjusted for league and park. An average wRC+ is 100; anything higher than that is good, anything lower than that is below average. Because of the park/league adjustment, players could have identical wOBAs, but different wRC+’s. This is great, and like wOBA, it’s a nice and quick way to check a player’s production. What it lacks, though, is an adjustment for position, and that’s where WAR comes in.
WAR, simply, is our best guess at how many wins above a replacement level player (think Ramiro Pena on the hitting side and say…2009 Brett Tomko on the pitching side) a player is. 2.0 WAR is generally considered to be average. WAR adjusts for position and takes into account defense and baserunning to arrive at its output. Lately, I’ve been straying away from WAR; I can’t remember the last time I used it in a post. The biggest reason for this is that the metrics we have to describe defense and baserunning are in their extreme infancy and the numbers are not wholly reliable. We’ve pinned down offense, to an extent, but the exact measurements of defense and baserunning continue to elude us, and probably will for some time.