Breaking Down the 2018 Red Sox: How to Analyze That Roster Properly
When I started digging into the 2018 Boston Red Sox roster a few years back, I expected it to be straightforward. It wasn't. Everyone knows they won 108 games and the World Series. What most people miss is how unusual their construction actually was, and getting that right requires looking beyond basic stats. Here is the practical approach I landed on. Start with the actual 40-man roster from October 2018, not the active 25-man World Series roster. The depth pieces matter more than you would expect. Then separate your analysis into three buckets: the stars, the role players who outperformed their projections, and the guys who were essentially roster filler but provided value in one specific area. The problem most people make is treating every player on that roster as equally important. Mookie Betts and Chris Sale obviously dominated. But the real story was in the middle-of-the-order guys who hit way above their Expected Batting Average marks. Xander Bogaerts had a career year, but his underlying numbers told a slightly different story than his slash line suggested. That disconnect is where the interesting analysis lives.
The Method
My workflow for breaking down this roster went like this. First, I pulled Statcast data for every position player and pitcher who logged meaningful appearances. Not just the basics. Exit velocity, barrel rate, chase rate, whiff rate, spin rate for pitchers. Then I cross-referenced with FanGraphs WAR and Baseball Reference WAR to see where the two systems disagreed. The disagreements are usually the most informative. For example, the 2018 Red Sox bullpen had a FIP that suggested it was lucky. David Price had a 3.50 ERA but a 4.30+ FIP over his workload that year. When you look at the strikeout-to-walk ratio and the ground ball rate, the regression was predictable. But he still took the field in October and performed. That gap between what the sabermetrics predicted and what actually happened is exactly where this kind of roster analysis gets interesting and frustrating at the same time. I ran into a specific problem when I was trying to evaluate the outfield depth. Jackie Bradley Jr. had lost significant ground speed by 2018. His defensive value had dropped from elite to average, but his OPS was still respectable. Most roto sites and even some baseball analysts were stuck grading him purely on his 2016 Gold Glove season. The workaround I used was to look at his Defensive Runs Saved and Ultimate Zone Rating from 2018 specifically and weight those heavily against his earlier career numbers. It gave me a much more accurate picture of what he actually contributed that year, which was less than most people thought but still useful in a lineup that stacked Betts, Drew Vernoy, and Hanley Ramirez.
Counter-Intuitive Things About That Roster
Most casual fans remember that roster as loaded with All-Stars. In reality, several key contributors were having career years that statistical models would have pegged as outliers. Steven Wright, for instance. The knuckleballer who posted a sub-3.00 ERA over 30 starts was essentially a one-dimensional weapon who could not get batters out with command. His walk rate was always elevated. But because knuckleballers are unpredictable, standard pitching analytics undervalued him before 2018. He was a classic case of a pitcher whose value came from something traditional scouting tools miss entirely. Another thing nobody talks about enough is how much the 2018 Red Sox benefited from park factors. Fenway Park inflates offensive numbers in a way that is difficult to normalize. When you strip out home park effects, some of those impressive regular season numbers become more ordinary. It does not diminish what they did. It just means you should not project that exact production level onto players who move to neutral or pitcher-friendly parks afterward.
Get the Full Details
Where This Approach Breaks Down
There are real limitations to analyzing a historical roster like this. The biggest one is that you are looking at a single season out of career arcs. A player might have had one incredible year and then collapsed the next. Andrew Benintendi is a good example. His 2018 rookie campaign looked phenomenal, but his underlying metrics suggested some luck in his contact quality. When you analyze a roster in isolation, you can overvalue guys who were peaking and undervalue guys who were about to break out. Another limitation is that modern Statcast data, while detailed, does not capture everything. Pitch design, catcher framing, and clubhouse dynamics all influence performance in ways that no publicly available metric fully accounts for. You can build a very thorough statistical profile and still be wrong about why certain players succeeded or failed.
What Actually Matters When You Look at This Roster
If you are doing this analysis for fantasy purposes, project-based modeling, or just general baseball understanding, focus on three players and the bullpen structure. Mookie Betts was the best player in baseball that year and his two-way value cannot be overstated. Chris Sale was dominant when healthy, and his innings were managed carefully. The bullpen, despite the advanced metrics complaining about it, held together better than anyone expected because of Craig Kimbrel and the setup men. Position-wise, the second base hole was real. Dustin Pedroia missed significant time. The Red Sox answered by moving Mookie to second and plugging in players like Eduardo Nunez and Brandon Drury at various points. That flexibility was not something you could have predicted in spring training. It is the kind of thing that shows up in retrospective analysis but completely escapes people watching the season unfold. The takeaways here are straightforward. When you analyze the 2018 Red Sox roster, do not just look at batting average and ERA. Use the advanced metrics. Acknowledge where luck and park factors played a role. And remember that a 108-win team can still have significant structural weaknesses that get papered over by star power and timely production. That is basically how baseball works, and the 2018 Red Sox are a pretty clear example of it.
