Last night I was watching Bill James’ fantastic 60 minutes piece when I asked myself a question. Why haven’t more people moved past on base plus slugging? Recently I have been reading “The Book”, which is basically a handbook to baseball written using historical baseball statics and sabermetrics. Early on the authors bring up a very interesting point. When people in the sabermetric community discovered that  the flaws in both OBP and SLG were equivalent to the strengths of the other statistic they decided to add them together. Although this definitely helped it still didn’t do the job perfectly.

The main issue with OBP is that it only takes into consideration weather a player reached base or not. Now this is a very useful tool, however it fails to consider the value of an extra base hit. SLG only takes into consideration total bases per at bat. It’s a very useful statistic however It does not take into consideration how incredibly more valuable anything is than an out. Where OBP will tell you how often a player managed NOT to make an out, SLG will not. A guy who reaches base once and gets 1 home run is not as valuable as a guy who walks 4 times. SLG would value both as equal.

Now combining the statistics to take into account both failure to make and out, and success at reaching multiple bases, should create a perfect statistic right? Wrong. OPS, or on base plus slugging, still fails in one significant way. It values each base as equal. This is entirely wrong. By looking at the run expectancy before, and after, and event on average through major league history we can see that certain base are more valuable than others. For example, the difference between a single and an out is greater than the difference between a single and a double. This is where a new statistic comes in, wOBA (or weighted on base average).

wOBA is a significant improvement of OPS. Instead of using total bases to value each event, for example saying a home run is 4 times as valuable as a walk, wOBA takes the actual run value of each event and then multiplies it by the number of times that event takes place. It then divides the number by plate appearances to get the true value a player contributes to his team each plate appearance. This also takes into account the value of anything over an out as an out lowers your value and anything else raise it. wOBA is the perfect combination of OBP and SLG that we have all been waiting for. Such a statistic is sabermetric gold. The exact formula at the time of the publication of “the Book” is:

wOBA =  HR*1.7 + 3B*1.37 + 2B*1.08 + 1B*.77 + BB*.62 all divided by PA.(Note: this is not the full formula, just a basic one. The full one includes a few other events such as hbp).

Now this can also be league adjusted. Say an American leaguer hits a double in 1960, and another American leaguer hits a double in 2004. By looking at the value of a double in 1960 and 2004 one can see that in fact a double in 1960 would be worth more. So to league adjust instead of using all time run values one could use single season in league run values. That way someone playing in an offensive era would be expected to produce more than someone hitting against, say, Bob Gibson in the 60’s.

So now I pose my question again, why hasent wOBA caught on while OBP, SLG, and OPS have. I believe it is because of 2 things. One reason is that wOBA values hits more than walks, something many (including me) disagree with. A ball put into play isn’t always a single, infact its only a single about a third of the time. A walk is always a walk. It always puts someone on base. So someone who puts the ball into play, but only enough to get a single, should not be considered more value. In fact I could say he was less valuable. But its just how you interpret the numbers and its very difficult to come up with something that doesn’t have anything to do with luck.

The other more important reason I see here is the complexity of the formula. To come up with these numbers you must first calculate run expectancy values for every event during each year and in each league. This requires a knowledge of where to find, and how to manipulate, historical play by play databases. While OBP, SLG, and OPS require no knowledge of sabermetrics or even of statistical analysis in baseball wOBA does. This is where the mainstream fan loosed wOBA.

This formula is being accepted by some baseball insiders. For example, current Pirates GM Neal Huntington, in a quest to use sabermetrics in player evaluation, has said that he does indeed use wOBA as well as stats like VORP and range factor. Progress.

Still I think more people need to become educated in run expectancy and in the wOBA formula, however until these numbers become more available online I just don’t see it happening. I also think the flaws in OBP, SLG, and OPS have to be addresed no matter what even if it doesnt happen through wOBA.

Alex Geshwind is a sabermetric baseball analyst for fantasybullpen.com. He can be contacted at alex.geshwind@fantasybullpen.com.The above picture is a cartoon of Bill James for those of you who do not know.