In The Lab: Early Season BPO and ROV

Essentially, the lab is a place where we can test what we see. That is what statistics are ultimately for. The whole goal is to explain mathematically what we are seeing and feeling. Sometimes we can explain it mathematically and sometimes we aren’t really seeing what we think we are seeing. It might seem like we are throwing a lot of pasta at the wall to see what sticks, but I am a firm believer in the idea that the more numbers we get that say the same thing the more likely we are to be accurate in our assumptions.

This brings us to real offensive value (ROV) and bases per out (BPO). Before we dive into these numbers we should define them, how they are calculated, and what they are looking for. Real offensive value is the easiest one to calculate. The idea behind it is the combine elements like batting average with elements that include everything but batting average. It includes two numbers that can easily be found at baseball-reference.com. Simply put, you add batting average to secondary average and divide by two.

Essentially, secondary average calculates everything a player does that does not include batting average. For most of us in the stats game, that would include isolated power, isolated patience, and stolen bases. A league average secondary average tends to mirror the league average in batting average, so a real offensive value can be interpreted the same way as batting average. A .250 ROV is probably close to big league average although early returns might be different this season.

BPO essentially measures the same thing in a different way. It is calculated by adding total bases, walks, hit by pitches, and stolen bases and dividing it by the total number of outs. Unfortunately, baseball-reference.com does not calculate either ROV or BPO, but the components are all there to do it by hand. While seasons vary, the league average tends to range between .650 and .700.

The first thing we are going to do is clean up our disclaimers and particulars. These are the numbers at the conclusion of Monday night’s game. Obviously, no two numbers will ever have a perfect 100 percent correlation. That is particularly true early in the season. ROV and BPO measure most of the same things, but there are some subtle differences. For instance, ROV does not penalize you for grounding into double plays, but bases per out does. On a long enough timeline, those things won’t matter as much, but in the span of 24 games it all matters a great deal.

These numbers confirm a couple of important things and highlight an area where we may have overlooked something. On the first count, Yordan’s numbers are just stupidly good. You’d have to track the best seasons of Barry Bonds, Ted Williams, Babe Ruth, or maybe Lou Gehrig to find numbers that would even approach that for a full season. For reasons that will hopefully be obvious, we won’t do that yet, but if this trend continues we will definitely look at those comparisons.

Secondly, Diaz has been historically awful both offensively and defensively (as Patrick pointed out yesterday). Again, this is about finding as many different ways to show the same thing over and over. It seems repetitive, but it also confirms something we are seeing with our eyes. This is one of those areas where Patrick has done the heavy lifting for me. It would be easy to look at Vazquez’s numbers and Diaz’s numbers and simply assume that Vazquez should be the starter from here on out.

That may turn out to be the case, but nothing is ever that simple. Vazquez’s underlying hitting numbers point to regression. We are looking at things like expected batting average, expected slugging percentage, and exit velocity. Vazquez will not be an all-star catcher when all is said and done. If we are lucky he could be the equivalent of a mediocre regular catcher when all is said and done. The good news is that he doesn’t have to be.

A baseball season is a 162 game marathon, but within that marathon there are multiple parts. Each player has good parts and bad parts. The trick is to maximize the good parts and minimize the bad parts. Vazquez is hot and Diaz is cold. That will likely change within a few weeks, but Joe Espada might as well take advantage of the hot streak now. If he and the Astros are lucky, Diaz will recover at about the same time that Vazquez begins to falter. In the meantime, you milk as much as you can out of Vazquez and minimize the damage Diaz is currently doing.

The surprise in the numbers comes in the fact that Brice Mathews doesn’t look like hot garbage when looking at those numbers. Many of the Astros young hitters are cut from the same cloth. They bring power, patience, and speed to the offense. They also bring a ton of swing and miss. The outfield will provide Espada with opportunities to mix and match. Center field might be an interesting spot for a platoon when Jake Meyers comes back. A lefty/righty platoon between Trammel and Meyers could be intriguing.

Similarly, a lefty righty platoon between Joey Loperfido and Mathews might also be interesting in left field. Obviously, Yordan will play there some as well as they try to fit Isaac Paredes into the every day lineup. The point is that lackluster pitching might give this team some freedom to experiment with guys in a lower pressure situation. If the playoffs are not riding on the outcome, you can use a couple of spots to test younger players to see if they can be a part of the picture moving forward. We will definitely revisit this as the year goes on. Is there anything that surprised you?