How Much Does the Spring Tell Us About Players in the MLB Draft?

Even though though the 2026 draft just concluded, most major league scouting departments have already turned their attention to next year’s class. Many of the players have been seen before, but they’re in for a much more thorough evaluation now that they’re eligible for selection.1 Over the next few months, scouts will head out to watch college players in summer wood bat leagues and high schoolers at various showcases across the country. Once they’re finished, and the reports are filed, teams will begin putting together their early rankings for 2027.

Of course all of those players have another season of baseball to be played before the draft takes places. That will give teams three to four more months of scouting looks, medical updates, performance data, and a host of other information they’ll incorporate to make picks. But for many of the top players, the foundation of their evaluation is set before that season begins.

Which got me thinking — how much is the extra information helping?

I’d be incredibly surprised if it wasn’t helping draft outcomes. It’s more a matter of where and to what extent. Because we’ve all seen situations where people overreact to small samples or let additional information lead them astray.

To investigate, I compared the earliest public rankings and/or mock drafts to selections and the eventual outcomes.2 The early rankings aren’t available very far back, so I couldn’t look at outcomes in terms of six-year WAR (or any real ML production). Instead, I looked at FanGraphs Future Value two years after the player’s draft year. As we’ll see, it’s not a perfect measure for individuals, but it’s good enough to help inform our opinion.3

As expected, teams’ actual selections significantly improve upon the preseason values. From 2018 to 2024, the Spearman correlation with Future Value is 0.544 for draft order against 0.297 for the opening ranks. If we restrict the analysis to only the players included in the preseason rankings, the gap grows even further.

But that’s both unsurprising and not that informative. To better understand what’s going on, let’s zoom in on the top ten spots in the draft. There little question that players projected to be top-ten picks before the season are well-known and held in high regard (whereas those questions exist for players further down) and the players selected in those spots are expected to deliver real value.

Looking only at the top tens gives us three distinct groups: players who started in the top ten and were selected there, players who dropped out of the top ten, and players who jumped into the top ten during the spring.

The best outcomes are for the players who stayed in the top ten from start to finish, but among players who changed groups, teams seem to do reasonably well deciding who should move up or down.

If we look at the magnitude of moves, we see a similar pattern.

We can see that the average outcomes for players who move up are better than the players who fall during the spring. And more importantly, the chances of finding a star or avoiding a bust are better in those cases as well.

What’s interesting to me is that the table suggests the preseason priors are rather useful. The stable group is by far the best performing and the players who go from unranked to the top ten have busted more often than they’ve turned into stars.4 And the group of small fallers includes JJ Weatherholt (1→7, 55 FV), Chase Dollander (2→9, 55 FV), Chase DeLauter (8-16, 55 FV), and Corbin Carroll (8→16, 60 FV) who are all likely underrated in this analysis given their current value.

In the cases where it was incorrect to move players down, I see two potential biases.5 The fist is that teams may overreact to draft year injuries. Weatherholt, DeLauter, and Aidan Miller all had non-threatening injuries during their draft year that seemed to push them down the board. Even in the cases of more serious injuries like Dylan Lesko, who had Tommy John during his draft year, the injury may have affected his draft position more than his future. It’s uncomfortable to take someone who has less recent information than others, but it isn’t necessarily wrong.

The second is that there seem to be some players who are famous for so long that fatigue sets in. This applies particularly to big time high school players like Carroll, Brice Turang, and Pete Crow-Armstrong who are on the summer circuit for more than one year. The amount of time causes teams to nitpick them more than other prospects and hold the public praise against them.

It might be exactly what happened with James Wood. Wood went eighth in the first Baseball America mock in 2021, but he lasted until the 62nd selection in the actual draft. Two years later he was a top ten prospect and today he’s one of the most valuable players in all of baseball.

The risks with Wood were already known when that mock was released. But teams had the whole spring to dissect and pick him apart. And while some may correctly point out that Wood signed for the equivalent value of a late first rounder, every team had the opportunity to give him that money multiple times before the Padres finally did.

Whether the spring season adds value was never really in question, but our simple analysis confirmed that it does. It also implies that the priors teams start with are quite strong.

Said another way, teams can and should change their minds, but only when they have reason to. The best forecasters constantly update update their evaluations and — this is the critical part — do so in proportion to the new information.6 A single game look, a week’s worth of stats, or a trivial injury probably aren’t enough to change the opinion on a player you have extensive history with. A few months, on the other hand, may be more than enough to draw an altogether different conclusion. It’s important to know when to change your mind. It’s equally as important to know when not to.