Amateur draft bonus analysis

The Yankees decided to play the draft a little differently than everyone else this year. Their first round pick, Clarke Schmidt, signed for $2,184,300. Then their second round pick, Matt Sauer, signed for $2,497,500. Notice the difference?

While this level of signing bonus variation is far from normal, Major League Baseball isn’t new to bucking overarching trends. In this article, I will walk through the nuances of the MLB Draft to help process why the draft conforms to certain seemingly-arbitrary trends.

If you’re unfamiliar or simply wishing to know more, the draft works as follows: each pick gets an expected bonus slot that’s higher than the subsequent pick. These assigned slot values come from joint decisions between the Commissioner's Office and the MLB Player’s Association in an attempt to guide the draft’s first ten rounds in a way that is fair and fitting according to all thirty teams’ previous season’s final record (also accounting for free agent compensation and competitive balance assignments).

These draft slots are summed together to create each team’s individual bonus pool; i.e. the total amount of money team x can spend on their selections in the first ten rounds. The one exception to this rule is in the case of a team spending more than $125,000 on a player after the tenth round, which would cause the overage amount to apply to the bonus pool. The whole idea of the pool amount is to keep teams in check at the cost of incremental monetary and draft capital penalties if a team goes over the limit.

The draft almost never plays out this way though. A player in the 39th round will sign for $250,000 and a player in the seventh round will sign for $1,000, as was the case this year for the Phillies’ D.J. Stewart and Braves’ Landon Hughes. Of course, leverage matters a lot in these draft negotiations, which is why one player got 250 times more than the other.

Stewart was a graduating high school senior with a college commitment to Eastern Illinois. He could’ve gone to school and re-emerged in three years at a significantly better draft position. Hughes; however, was a college senior who played all four seasons and had few other options if wished to pursue a place in the game. With zero negotiating power, Hughes had to accept whatever number was thrown out there. And since he was taken in the first ten rounds, he was designated a money-saving pick from the get-go. Hughes’ slot value of $213,400 allowed the Braves to use the leftover $212,400 to sign other picks.

The goal of Stewart and Hughes’ story of fork-in-the-road experiences serves two purposes: (1) to demonstrate the inequality of the draft on college seniors and (2) to demonstrate the fact that, on a large-scale, there’s a shaky relationship between talent, draft position, and financial bonus in the draft.

This makes it extremely hard to project bonuses in one all-encompassing motion. The draft is too complex for a single mathematical equation. Instead, the 40 rounds can be broken down into five major phases that have clear distinctions from one another, allowing for greater accuracy in projecting bonuses. To do so, I created a piecewise regression that predicts the expected draft bonus for all 40 rounds.

To see the visualization in its fullest form, head here.

PHASE ONE: ROUNDS 1 - 2A

Phase one is the first day of the draft, from the first overall pick, Royce Lewis, all the way through pick number 75, J.J. Matijevic. This phase sees bonuses slide down exponentially through the end of the night, eventually landing at Matijevic’s $700,000 bonus, less than 10 percent of the highest, Hunter Greene’s $7,230,000.

Because of the large numbers involved, the ranges on the first day of the draft are the most difficult to predict. Players will sign for $1 million under-slot or $1 million over-slot, depending on the circumstances. These giant variations simply aren’t seen in the later rounds because the bonuses are so much smaller.

Circumstances rule why players don’t get the assigned slot values for where they were picked. The selecting team usually has little to do with it. Rather, injuries, age, and - the trickiest to measure - draft expectations, determine the residual, the delta between a player’s expected signing bonus and what they actually signed for.

Clarke Schmidt, the Yankees first-rounder who signed for more than a $1.2 million under-slot, had recently undergone Tommy John surgery to repair a torn UCL. Still, he showcased elite potential beforehand that warranted a selection this high. Therefore, the Yankees decided the additional injury risk, talent level, and ability to manipulate their bonus pool blended into the right drafting scenario at pick 16.

PHASE TWO: ROUNDS 3 - 5

The second phase of the draft is closely related to the first, but begins to see bonus amounts level off. The first few picks are almost always dotted by high-upside high school prospects, as was the case this year with Blayne Enlow ($2 million), Nick Allen ($2 million), and Freddy Tarnok ($1.445 million).

Bonuses then take on a gradual slope descending to about $500,000 through the fifth round. Teams continue to select highly-ranked prospects and upside plays as they amass talent, which differs from the phases to come. In the graph below, the expected bonus figures (from the piecewise regression) are marked in blue, whereas the actual bonus amounts are in red.

The gap between the end of day one and the start of day two is most clearly exemplified by Matijevic and Enlow. The Twins’ third round selection, Enlow, received $2 million, largely because of pre-draft expectations. Now, this concept of expectations basically means the player’s true talent level, as evaluated by the industry as a whole. One way to prove that talent demands higher bonuses than an actual pick position is to use the Baseball America 500. BA’s player rankings correlate more strongly with the bonus a player received (0.80 R^2) than where the player was picked with their bonus (0.65 R^2).

Eventually, the first senior comes off the board. In 2017, it was Wyatt Mills, the Mariners’ third round pick at 93 overall. Mills signed for $125,000 despite his assigned slot value being listed at $579,800. The pick stunned many in the baseball industry because of how early such a player was taken. Barely 12 picks earlier, a top-30 talent went to the Athletics, and here the Mariners were taking a senior reliever?

Turns out he was needed all along. The Mariners’ previous selection of Sam Carlson was costly because he signed for almost $800,000 above what his slot value was in the second round. The money that Seattle saved in picking Mills went directly to Carlson. While this kind of “gaming” is by no means out of the ordinary, the pick number was.

PHASE THREE: ROUNDS 6 - 10

The prevailing characteristic of this phase is exactly why Mills went in the third - because he was cheap. Without any leverage to negotiate with, college seniors are stuck with what they’re given. Oftentimes, this is hundreds of thousands of dollars below slot value.

To demonstrate the rash of money-saving signings, take a look at the number of seniors who received slot in the entire draft (receiving more than $125,000 after the tenth round also qualifies).

Zero of 302.

The discouraging trend has been covered before, as former Cal Poly senior Zack Zehner told Baseball America, “You’re willing to play for a plane ticket. You want to play. For me it’s about finding the team that wants me the most,” Zehner said. “Who will give you that plane ticket? It’s not a money situation anymore.”

The chart below looks at the percentage over or under slot that each class of player received, on average, in the top ten rounds. High schoolers make out with the best percentage of slot, at 111.77% their designated values. Meanwhile, seniors receive the lowest percentage of slot: a mere 12.24%. Scroll back up to the interactive graphic above and notice the gap in the residual graph between what seniors are signing for and what everyone else is signing for in rounds six through ten.

PHASE FOUR: ROUNDS 11 - 20

The first few rounds of day three look like night and day compared to the previous phase. Now that players don’t have slot values (there is no way to save money against the bonus pool), teams aren’t afraid to take backup or “spare” picks. This is why you’re likely to see a sudden uptick in the number of high schoolers being taken and signed in the 11th round - there’s minimal risk if they don’t sign. Whereas if a player doesn’t sign in the first ten rounds, that slot value becomes nullified, thereby decreasing the total bonus pool amount for the team.

Day three’s high upside picks are almost always taken in the first hour. Once a team signs their first ten rounders - if there is still bonus money left over - they’re free to apply it to their later selections. What this creates is a backup system that allows clubs to have emergency plans in place should a high-ranking selection choose not to sign. Even though the slot money would be lost, the lack of a potentially over-slot deal would allow for pre-existing under-slot deals to save unused pool money. Then, it could be applied to this player who would otherwise assuredly be going to college later in the draft.

If that’s confusing, think of the fallout if the Mariners failed to sign Sam Carlson, the aforementioned over-slot second rounder. Seattle would still forfeit that slot value, but would have nowhere to spend the savings from Wyatt Mills and all of their other senior signs in the top ten rounds. Accordingly, they’d be able to go after someone like their 37th round selection, Jesse Franklin, who rated as the 230th best draft prospect.

The easiest way to test this is by comparing the composition of the 10th round against the composition of the 11th round. In the 10th, there were 7 juniors and 23 seniors selected. But in the 11th, there were 11 high schoolers, 3 junior college players, 15 juniors, and 1 player without a school selected.

The chart below pieces apart the regression used to predict draft bonuses by classification. In it, you’re able to see the average residual for a player of each classification that gets taken in each phase. It’s different from the aforementioned slot chart because these figures are derived from the regression values and not the assigned slot values. For example, a high schooler taken in rounds 6-10, on average, makes $227,577 more than their draft position would predict.

Consider the Junior College, Sophomore, and No School classifications with a grain of salt, as there are significantly fewer of those players drafted.

PHASE FIVE: ROUNDS 21 - 40

Despite being the least glamorous portion of the draft, half of all players will be taken in these rounds. As phase one was the most difficult to predict financially, phase five of the draft is the most difficult to predict talent-wise.

At this point, every team’s big board is going to look completely different. Picks are coming in every 15 seconds and clubs need to be prepared to pick at a moment’s notice. Without seeing every team’s board at once, it’s hard to say how varied they are; however, it’s not difficult to envision a hundred or more players that are unique to each team.

Really, there is no signature aspect to this part of the draft besides that it differs from the slightly-more-prospect-laden earlier portion of day three. It’s an endless stream of seniors and juniors looking to take the first step into professional baseball at the cost of a small signing bonus and a chance. The rare high school signee might pop up in the late 30s, likely because a team had spare change in their bonus pool that they could offer on top of the $125,000 soft ceiling.

The average signing amount gradually descends to $17,881 by the final pick in the draft, but there is no clear way to predict a dollar amount because the talent varies so wildly.

*The signing details of 23 players were not reported, and therefore excluded from this study.

You can follow Justin Perline on Twitter at @jperline