it is a new year and with new years come all kinds of possibilities. One of the fun parts when the calendar switches to 2026 is seeing what different projection systems have in store for your favorite
team. I am a big believer in data. That is what the lab is all about. Few things are ever universally true, but one of the general rules in data is that the more information you have the better your data is going to be.
A couple of weeks ago I looked at the differences between the Hall of Fame Index and JAWS. The primary difference was data. The books used three different data sources to one for JAWS. The current formula only uses two. Similarly, going with one projection system seems foolish when there are so many to choose from. There are a couple that have not released their data yet, but most have and we will be going with our rule of three.
What we are doing in this piece is comparing the 2025 Astros with their projections in the coming season. We are looking strictly at OPS for now. Unfortunately, most of the projection systems do not include the data needed to calculate bases per out. So, we will include sources like Steamer/ZIPS, The BAT, and Baseball-reference.com. The idea is that if we get a broad enough cross-section of experts then we will get a good idea of where the players are expected to land and then we can compare that with how they finished in 2025.
I should make a few notes before we really dive into what these numbers mean. First, you will notice that I did not include any of the counting numbers for these players. There are two very good reasons for that. First, the projections for percentage stats like batting average, on base percentage, slugging percentage, and OPS tend to be much more accurate than the counting numbers historically. The second reason is somewhat related, but is particularly acute in the Astros case.
You will notice that we have too many players in the table. On a 26 man roster we can expect 13 position players at any one time. My math skills are sometimes shaky, but I count 15 players in the table. That means that two of them will either spend most of the time in the minors or on another major league team. We can certainly speculate based on rumors and simple observation, but it is best to include all of those players until we know. You will also notice that Victor Caratini is in the table even though he has not officially signed.
Furthermore, when we look at the data we see some positives and some negatives. The players that had poorer seasons generally are expected to improve. However, with the exception of Yordan Alvarez, the players with the best seasons are all expected to regress some. Take out Alvarez and you are looking at a lineup with a ton of players coming in above .700 but below .770. You do have a few bench pieces below that, but on most nights it is a pretty homogenous lineup.
A part of a manager’s job (maybe the most important part of his job) is determining how to marshal his resources. In other words, he gets to choose who gets to play, where they get to play, and where in the lineup they get to hit. The Astros have a roster that has more flexibility than most, so it makes little sense to project individual counting numbers when we have no idea of how that will play out.
In some case, it will impact the numbers above. Jesus Sanchez has one of the most severe lefty/righty platoon splits in baseball. You might as well throw one of us out there with a bat against lefties. 20 to 25 percent of starters in baseball are lefties, so you will severely limit his plate appearances by not playing him, but you will also boost his OPS. It’s a considerable trade off.
Obviously, once we have gotten the pleasantries out of the way, we can move onto the implications of these projections. I have used OPS here because it is easier to digest for most fans and projection systems often lack the data needed to compute are more sophisticated numbers. OPS has some issues as a statistic. The primary one is that it counts batting average twice. Batting average is the core part of on base percentage and slugging percentage. Therefore, a .300 hitter has a natural advantage over a .250 hitter.
While that would seem to be a no brainer, not all .300 averages are created equal. A hitter that hits primarily singles and doesn’t draw a lot of walks will look like a better player with OPS than they actually are. For instance, Luis Arraez of the San Diego Padres has a career .777 OPS which comes out to a 115 OPS+. He is also a .317 career hitter which means that .634 of his .777 is all batting average. His .693 career bases per out would be 102 BPO+ based on the 2025 MLB universe and almost exactly average for first basemen.
For a team like the Astros that seemingly have a group of about a dozen hitters with very similar OPS scores, it would seem like it would not matter where they hit in the order. The simple fact is that not all OPS scores are created equal. A Yainer Diaz or Christian Walker might be much more skewed to slugging percentage where Isaac Paredes and Carlos Correa might be more skewed towards OBP. Certain spots call for certain skills and if Joe Espada and the hitting coaches can maximize that then they might be able to boost the Astros overall run total by as many as 50 runs.
One of the things we will do in the lab this year is explore alternative numbers that explain offensive and defensive production a little more clearly than cruder numbers like OPS or simple ERA. Interestingly enough, ballparks are beginning to display these numbers for fan consumption. It could be the popularity of the metrics themselves or the increased popularity of fantasy sports and gambling sites. Team are always looking for ways to leverage information to their benefit, so we should do the same.
