Spring Training Standings Aren't Real — But Your Hype Analysis Method Should Be
Most people treating spring training results as predictive data have it wrong. The W-L records you see in Daytona and Phoenix are almost entirely noise. Players are getting reps, coaches are testing lineup combinations, and the standings tab on MLB.com is basically decorative. I've sat in press conferences where managers openly admitted they'd rather their club lose five extra-inning games in March than risk a starter throwing 95 pitches against a roster full of call-ups. That doesn't mean spring training analysis is useless. It means you need a completely different framework than the one you'd use for regular-season standings. The standard approach most analysts fall back on — win percentage, run differential, batting average — collapses under even a casual inspection. A team can go 12-8 while outscoring opponents 45-40 because their rotation consists of four bullpen arms and a guy returning from Tommy John surgery who hasn't thrown more than 40 pitches in any single outing.
Spring Training Standings A Comprehensive Analysis Of The MLBs Early Season Hype
Here's how I actually approach this each year. I don't look at the standings column at all. What I track instead is workload distribution, velocity trends, and contextual performance metrics that survive the noise. The first thing I check is how many pitchers each team has used. If a club has thrown 18 different arms through their first 15 exhibition games, that's not a sign of depth. That's a sign they're managing around injuries or evaluating fringe roster players who won't be here by April. I keep a running spreadsheet of pitch counts per pitcher per game, and any starter logging over 65 pitches in a spring matchup gets flagged as either a closure candidate or someone being rushed back from the IL. The second thing is velocity tracking. This is where most people waste time. You don't need StatCast access to get useful data. PitchersReport.com posts spring velocity charts two days after games, and that's enough to spot trends. If a reliever who sat at 96-98 mph last September is touching 93-94 in March, that's a real signal. If a starter who's been flat all spring suddenly sits 95-97 in his third Grapefruit League start, something changed in the offseason — a new grip, a mechanical tweak, or maybe he's just finally healthy. I once spent three days trying to figure out why the Brewers' top prospect was throwing 99 mph in the minors but sitting 92 in spring. Turned out he'd been nursing a flexor strain and had intentionally depressed his velocity to stay under the 60-pitch threshold each outing. Workaround: I stopped looking at his numbers entirely and started tracking his outing frequency instead. Four starts in 14 days with sub-55 pitches meant he was being held back, not failing. Third, I evaluate position player performance byting plate discipline metrics from result metrics. Batting average on spring balls is meaningless — the balls come out of different cans, the fields have uneven turf, the defensive quality varies wildly. What actually matters is walk rate and strikeout rate. A guy striking out in 35 percent of his plate appearances in spring while walking six percent is going to have an ugly April regardless of how many hits he's getting now. Conversely, a .180 hitter who's drawing walks at a 14 percent clip and cutting his swing-and-miss rate is the kind of player who tends to flip his numbers once he faces regular-season pitching. I've seen this pattern hold up across multiple seasons with middle infielders and bench players.
One counter-intuitive thing that catches people off guard: teams that go poorly in spring training often outperform expectations in April. The reason is simple. Clubs that lose spring games are usually making deliberate adjustments — trying new things, giving veterans extra reps, letting players work through slumps in low-leverage situations. The hype machine loves to take a 3-10 spring record and project a corresponding regular-season slide, but that record almost never correlates with April performance. What actually predicts April success is how many meaningful decisions the coaching staff has made by late March. Are they committed to a starter? Do they know who their closer is? Has the manager settled on a bench role structure? These questions are answerable in spring, and the answers matter far more than any exhibition win. There are real limitations to this approach, and I should be blunt about them. The biggest one is sample size variability. Some teams play 33 spring games. Others play 28. Some play interleague exhibitions against MLB opponents; most play against each other or minor league squads. This makes cross-club comparison inherently messy. My velocity tracking works well for established big leaguers. It falls apart for players who haven't been signed yet or guys working out of extended spring training — you simply don't get public data on them. And the whole framework assumes you have access to pitch count logs and postgame mechanical notes, which most casual fans don't. If you're just reading headlines, you're already behind. Another pitfall is recency bias in spring evaluation. A pitcher throws three shutout springs outings and suddenly he's the front-runner for the rotation. A batter goes 1-for-22 and everyone writes him off. Neither outcome is predictive. Spring baseball is designed to generate exactly these kinds of dramatic mini-narratives because it gives reporters something to talk about during a dead period. The actual signal is buried under layers of exhibition noise, and separating them takes work that most analysis shortcuts skip entirely.
Get the Full Details
The most reliable single metric I've found for predicting early-season performance is bullpen inning consumption rate across the final ten spring games. Teams that give their primary relievers 15 or more combined innings in that stretch tend to enter April with clearer roles and less fatigue. Teams that barely touch their setup men are usually still figuring out who they are, which correlates with slow starts. It's not perfect. I've seen it fail in years where training camp injuries wiped out the very arms it was supposed to predict. But it's the closest thing to a usable predictor I've found in over a decade of tracking this stuff. If you want to actually use spring training data productively, stop looking at the standings. Start building your own tracking sheet with the three categories I mentioned — workload distribution, velocity trends, and plate discipline rates. Cross-reference them against Opening Day roster announcements. The gap between what spring training tells you and what it pretends to tell you is where the actual edge lives.
