I love the ISO stat. Why do I love the ISO stat as opposed to good old slugging percentage (SLG)? Well, the problem with slugging percentage is that it’s linked to batting average.
Slugging percentage is calculated by adding up the total bases — singles + (doubles x 2) + (triples x 3) + (home runs x 4) — and dividing by at bats. What does this mean? It means that slugging percentage is linked to batting average which isn’t necessarily the best statistic to begin with. This means that slugging percentage can be inflated by a high batting average, whereas a player with a lower batting average who hits for a lot of power can have a seemingly lower slugging percentage.
Examples: Ryan Schimpf plays for the New Hampshire Fisher Cats. He leads the team with a .521 slugging percentage. His teammate is Kevin Pillar who owns a .464 slugging percentage. Both players are everyday players for the Fisher Cats and we can see that according to slugging percentage, Schimpf has more power. This is certainly true. Schimpf leads the Eastern League with 11 HRs, far more than anyone else on the team, but what slugging percentage doesn’t show is how many of a player’s hits go for extra bases because hits (singles) are included in slugging percentage.
ISO (or isolated power) removes those hits from the equation. Schimpf is only hitting .219 while Pillar is hitting .328. Remove that from their slugging percentages and Schimpf is left with an ISO of .301 while Pillar is left with .135.
Fangraphs tells us that a .145 ISO is average while .250 is excellent. .301 is astronomical. Interestingly, Schimpf’s .301 ISO slightly above what it was in his portion of the season in New Hampshire last season in 137 plate appearances (he has 178 so far this season), so it is not a fluke.
Now here’s where we have some fun. Another stat that batting average influences (for good and bad) is on-base percentage (OBP). Let’s use the same two players. They have similar OBPs. Schimpf’s is .360 and Pillar’s is .371. Again, due to the over 100 point different in these players’ batting averages, it means that Schimpf is getting little help from overall hitting and is relying almost entirely on his ability to take walks. Whether Schimpf walks more due to a better eye, or because pitchers are wary of his power is something to be discussed another day.
So, to examine a player’s ability to get on base free from their batting average as well as the ability to assess a player’s ability to get on base via the walk (and other, non-hitting means) as well as a player’s ability to hit for power, I propose two new statistics: ISOp and ISOt. The P is for Patience, and the T is for Total.
I have decided to keep ISOp simple rather than complicate things. ISO is such a great stat because it’s simple subtraction. You can look at a player’s extra-base pop just by subtracting his batting average from his slugging percentage. Unfortunately, the denominators for Batting Average and On Base Percentage are different which complicates things. Batting Average uses at bats to divide hits by. OBP uses a slightly more complicated configuration of at bats plus walks (intentional and unintentional) plus sacrifice flies to divide by.
So instead of recalculating BA, I’m just going to say subtract BA from OBP and you get ISOp. This isn’t a perfect, all-encompassing stat, but what it does is it tells you how good a player is at getting on base via means other than hits.
Why is this important? Well, sometimes you want to look at players and see how their eye is developing. Looking at Schimpf, who has a .141 ISOp, and first baseman Clint Robinson, who has a .397 OBP but a .102 ISOp, tells you that Schimpf doesn’t quite get on base as much, but walks more. ISOp does this without needing to look at BB% or adjusting the raw number of walks based on plate appearances.
Where the real insight comes is in ISOt or Total ISO. The best description that I can come up with for this stat is the “Added Value” that you get from a player. Some people speak of empty batting averages, where hitters get only singles without getting on base much. In 2013, Melky Cabrera is one of these players who, with ISOt, these players are heavily penalized. Right now, Cabrera has a batting average of .288 with an OBP of .323 and a SLG of .384. This makes his ISOt .131. It may be off a point due to rounding that goes into the calculation of the original rate stats.
When we take a player like Edwin Encarnacion, who is hitting only .255, but has an OBP of .335 and SLG of .503, we’re left with an ISOt of .328. This means that the Jays are getting far more “added value” from Encarnacion than they are from Cabrera. Of course, this is reflected somewhat in OPS (On Base plus Slugging). Their OPSs are only 121 points apart because the difference in batting averages — 33 points — is counted TWICE. ISOt elminates this problem and illustrates the difference in added value quite clearly.
Back to Ryan Schimpf. The 25 year old third baseman would demolish even Edwin. He has an ISOp of .141 and add that to his ISO of .301, you get a .442 ISOt or added value rating, whereas someone like Kevin Pillar, who trails Schimpf in OPS by 46 points would have a .179 ISOt.
I’m not saying that Schimpf is a better overall player than Pillar, but by using ISOt, you can see much more clearly the things that he does well, much better even, than Pillar who was written about recently by Marc Hulet of Fangraphs. One writer actually told me that Schimpf will probably not make the major leagues while Pillar will likely have a career as a fourth outfielder. The thing to keep in mind is that ISOt is not a complete stat. It isolates (see what I did there) certain aspects of a player’s production. It is also a descriptive stat and not a predictive stat. It describes and puts a new spin on what has happened on the baseball field. It doesn’t tell us anything about what a player is going to do. That said, ISOp and ISOt can be useful tools to look at specific things like walks and extra-base-hits. It also doesn’t allow batting average to get in the way and cloud a stat like OPS.