Call-Up Probability Engine · Model v0.4 · How It Works

How MLB Prospect
Call-Up Probability
Actually Works

A full explanation of how the MKDC Baseball call-up probability model works: what drives CUP scores, how service time suppression affects timing, what the PPI multiplier means, and how to read the ECD window for any MLB prospect.

MKDC Baseball answers one question: will this minor league player make an MLB appearance within the next 90 days? Not how good they are. Not whether they will succeed. When.

This page explains what every number on the leaderboard means and how the model produces them. The framework is fully public. The specific calibrated parameters are proprietary.

What This Model Is Not
This model does not evaluate swing mechanics, pitch mix, athleticism, or any subjective scouting factor. It does not predict whether a player will succeed at the MLB level. A high CUP score means the player is structurally positioned to be called up, not that they will thrive once there. It is not a scouting system, a talent ranking, or a fantasy baseball tool.

How to Read the MLB Prospect Call-Up Leaderboard

Every player tracked by the model has three core numbers: CUS, CUP, and ECD. Understanding these three values tells you everything about a prospect's call-up probability and expected timing. Three numbers appear on every player row: CUS, CUP, and ECD. Here is exactly what each one means and what different values should tell you.

See live examples on the CUP Leaderboard: 27 prospects from the 2026 Top 100, each with a CUS, CUP, and ECD. The calibration log shows every prediction made publicly before Opening Day.

CUS - Composite Upward Score
0 to 100 · Structural readiness score
70 to 100
Very high structural evidencePlayer is at or near MLB readiness across every measurable factor. On the 40-man, at AAA, performing well, org has a clear positional need. Call-up is structurally supported. Still subject to service time and org decisions.
55 to 69
High structural evidenceMost readiness signals are positive. Likely on or near the 40-man. A realistic call-up candidate within the season. One or two suppressing factors are holding the score back.
40 to 54
Moderate structural evidenceSome positive signals but meaningful barriers remain. Not on the 40-man, or blocked positionally, or at a level where a call-up this season requires best-case performance.
25 to 39
Low structural evidenceMultiple barriers in place. Too young for the level, not on the 40-man, deep org with few openings. A call-up this season would require an unusual set of circumstances.
0 to 24
Below thresholdNo realistic 90-day call-up window. These players are tracked for future batches but do not appear on the scored leaderboard.
CUP - Call-Up Probability
2% to 95% · 90-day binary probability
35% and up
Strong call-up caseReal, near-term call-up probability supported by multiple independent factors. A CUP above 35% is not a certainty, but it is a genuine signal worth tracking closely.
22% to 34%
Active watch windowThese are the leaderboard's scored players. The structural conditions for a call-up exist. Timing depends on service time windows, org decisions, or injury events unlocking the path.
13% to 21%
Summer window possibleA call-up is plausible but requires a longer performance runway or multiple things going right. Pathway C drives most of these. Watch for rapid level advancement as the season progresses.
8% to 12%
Low probabilityStructural barriers are the dominant story. A call-up can still happen if multiple suppressing factors clear at once, but the base case for this 90-day window is no promotion.
Under 8%
IL or below thresholdEither on the injured list with no timeline, or too far from the MLB level for a 90-day window to be realistic. Tracked but not scored.

What CUP Is Not

CUP is a binary probability: the player either makes an MLB appearance within 90 days or they do not. It is not a prediction of how many games they will play, whether they will stick on the roster, or how they will perform. A player with a 38% CUP who gets called up and sent back down after two weeks is still a resolved positive outcome for the model. The question is purely: did they appear on an MLB roster within the window?

Why CUP Is Always Below 95%

The model caps CUP at 95% for any player not yet on an active MLB roster. Even a player with every structural factor pointing toward a call-up can face an injury, a trade, or an org reversal. The cap reflects genuine uncertainty rather than false confidence.

ECD - Estimated Call-Up Date
When, Not If
The ECD is a date range: Early, Mid, and Late. Mid is the central estimate. The width of the range depends on which Pathway applies. High confidence means dates are tightly constrained. Low means there is genuine uncertainty in the timing even if the call-up itself is likely.
ECD Mid: Apr 16 · High +/-7d
Pathway A / B / C
What Drives the Timing
Pathway A means a service time window is driving the date. Calendar-driven and the most predictable. Pathway B means an injury or roster opening is creating the opportunity. Pathway C means performance trajectory is the primary driver, with the widest date range.
A = Service Time · B = Opportunity · C = Performance
PPI Flag
Org Financial Incentive
A PPI-High flag means the player is eligible for a Pre-Promotion Incentive bonus if called up before April 9. This creates a real financial reason for the org to promote early. PPI None means that incentive is gone or was never applicable.
PPI deadline: April 9, 2026

The Four-Stage Call-Up Probability Pipeline

Every player evaluation runs through the same four stages in the same order. No stage is skipped. Outputs from later stages do not feed back into earlier ones.

01
📥
Raw Inputs
12 player-level data points from public sources available at the evaluation date. Each input must be verifiable and consistently defined. Nothing from after the run date is used.
Stage 1
02
⚙️
Feature Scoring
Each raw input converts to a normalized feature score. Scores are additive. Weights are set at design time, not adjusted after results come in.
Stage 2
03
🎯
CUS + CUP
Feature scores sum to the CUS (0 to 100), which maps to a base call-up probability. Three sequential multipliers then produce the final CUP percentage shown on the leaderboard.
Stage 3
04
📅
ECD Engine
CUP converts to an Estimated Call-Up Date through one of three pathways. Each pathway has a confidence tier and a date window that reflects how predictable the timing actually is.
Stage 4

The 12 Model Features That Drive Call-Up Probability

The model uses only inputs that can be sourced from public data at the evaluation date. No subjective assessments, no scout reports, no proprietary metrics.

Readiness Signals
Current minor league level (proximity to MLB)
Age relative to league average at that level
30-day performance stats (K-BB% for pitchers, contact and power for hitters)
Performance trend vs. 60-day baseline
Spring training performance tier
Org and Roster Context
40-man roster status
Org opportunity (IL count, roster churn, competitive window)
Minor league options remaining
Position scarcity at the MLB level in that org
Rule 5 draft exposure flag
Priority Signals
Prospect ranking tier within the org, used as organizational priority proxy only, capped at minimal weight
PPI status: whether the player qualifies for a Pre-Promotion Incentive and whether the deadline window is still open

The Three Multipliers: Service Time, PPI, and Org Coefficient

After the base CUP is set from the CUS score, three multipliers are applied in sequence. All are multiplicative, not additive. The final CUP is clamped between 2% and 95%.

Org Coefficient
Range: 0.85 to 1.02
Every MLB organization has a different historical pattern of promoting prospects. Some promote aggressively. Others hold players back well past the point their readiness would justify a call-up. 30 org-level coefficients capture this behavior.
COL = highest · TBR = lowest
Positional Base Rate
Range: 0.85 to 1.15
Different positions get called up at structurally different rates. Relief pitchers move up frequently. Catchers rarely do. This multiplier adjusts for the base frequency at which each position group historically reaches MLB within 90 days.
RP = highest · C = lowest
Service Time Suppression
0.65x during window
The most impactful multiplier in the model. When a team has financial incentive to delay a player's first MLB appearance to preserve a year of team control, this 0.65x suppressor applies. It reflects a documented force in MLB roster decisions.
First-callup players only

ECD Pathways: How Call-Up Timing Windows Are Calculated

Every player gets exactly one pathway. The pathway determines the width of the date range and the confidence tier on the leaderboard.

A
High +/-7d
Service Time Suppression
Calendar-driven · most predictable
The service time window has a known end date based on the MLB calendar. Once it closes, the org has no financial reason to keep the player in the minors if they are ready. ECD is calculated from Opening Day plus the suppression period. Errors on this pathway are usually within a week.
B
Med +/-21d
Injury / Opportunity
Event-driven · moderate precision
Triggered when the org has a clear positional vacancy, usually from MLB injuries, roster churn, or a competitive situation requiring an upgrade at a specific role. Timing depends on events that are identifiable but not perfectly predictable, so the window is wider.
C
Low +/-45d
Performance Unlock
Default pathway · widest range
When neither a service time window nor a clear vacancy exists, the ECD comes from the player's performance trajectory. The wide range is not a model weakness. It reflects genuinely uncertain timing. A Low confidence ECD is accurate information, not a bad prediction.

Calibration Commitment: Public Track Record for Every Prediction

Public Track Record
Every prediction is logged publicly before outcomes resolve. The Brier score, the standard accuracy metric for binary probability forecasts, is calculated at fixed checkpoints and published permanently. Misses are logged the same as hits.

A model that only shares its wins is not a model. The checkpoints below are when outcomes are formally resolved.
April 9, 2026
PPI Deadline - downgrade flags resolved
May 1, 2026
Week 6 - service time windows unlock
May 29, 2026
Week 10 - majority of windows resolved
June 18, 2026
90-Day Resolution - Brier score published

What CUP Measures and What It Does Not Predict

What CUP estimates
Probability of an MLB appearance within 90 days
Likely timing window (ECD Early, Mid, Late)
The pathway driving the timing estimate
Confidence tier for the date range
Org-level context driving or suppressing promotion
What CUP does not estimate
Player talent, ceiling, or long-term potential
Whether a player will succeed once called up
Fantasy baseball value or roster priority
Award probability or statistical projections
Any outcome beyond 90 days from run date

Version History

No parameter changes within a version. Any change requires a new version number. All outputs are permanently tied to the model version that produced them.

v0.1
Core pipeline. Five features. Static CUS to CUP mapping.
v0.2
Expanded to 12 features. Org coefficients. Positional base rates. Service time suppression.
v0.3
ECD engine. Three pathways. Confidence tiers. Early, Mid, Late date ranges.
v0.4 - Current
First batch run (Sub-Batch A). PPI multiplier layer. Carry-forward handling. Active MLB routing.
v1.0 - 2027
Model trained on labeled outcomes from the 2026 season. Calibrated probability mapping from real historical data.

Frequently Asked Questions

What does CUP mean in baseball prospect rankings?
CUP stands for Call-Up Probability. It is a binary 90-day probability that a minor league player will make at least one MLB appearance within the next 90 days. A CUP of 35% means the model estimates a 35-in-100 chance of an MLB appearance in the 90-day window. CUP does not predict whether the player will succeed or stick on the roster: only whether they will appear.
What is service time suppression and how does it affect prospect call-ups?
Service time suppression refers to the practice of keeping a prospect in the minor leagues for the first two to four weeks of the season to delay their service time clock. A full year of MLB service time is 172 days. By keeping a prospect down through early April, a team gains an extra year of control before arbitration and free agency. The MKDC model applies a 0.65x suppression multiplier to all first-callup players in the first 16 days of the season, which reduces their CUP score to account for this organizational practice.
What is the PPI deadline and why does it matter for prospect call-ups?
PPI stands for Prospect Promotion Incentive. It is a compensatory draft pick awarded to a team if an eligible prospect wins the Rookie of the Year award or finishes as a finalist for the MVP award. To qualify, the player must be promoted by a specific deadline: typically in early April: and must appear on qualifying Top 100 preseason prospect lists from at least two major outlets. The 2026 PPI deadline is April 9. A player not promoted by that date forfeits the incentive for that season. The MKDC model applies a 1.12x PPI multiplier for High-eligible players, 1.06x for Medium-eligible, and 1.0x for ineligible players.
What does ECD mean in the MKDC Baseball model?
ECD stands for Estimated Call-up Date. It is expressed as a three-value range: Early, Mid, and Late. The Mid ECD is the model's central estimate for when a call-up is most likely to occur if it happens. The range width depends on which pathway drives the call-up: service time windows produce narrow ranges (2-3 weeks), opportunity pathways (injury or roster change) produce medium ranges (4-6 weeks), and performance trajectory pathways produce wide ranges (8-12 weeks). ECD is only meaningful for players with a CUP above 8%.
How is MKDC Baseball different from other prospect rankings?
Traditional prospect rankings evaluate talent, tools, and long-term ceiling. MKDC Baseball does not evaluate talent at all. It evaluates one thing: the structural probability that a minor league player appears in an MLB game within the next 90 days. The model uses 12 structural features including organizational depth, service time position, positional base rates, and roster construction: not swing mechanics, fastball velocity, or scouting grades. Every prediction is logged publicly before outcomes are known and scored via Brier score at the 90-day calibration checkpoint.
What is a Brier score and how does the model use it?
A Brier score measures the accuracy of probabilistic predictions. It is calculated as the mean squared error between each predicted probability and the actual binary outcome (1 = call-up happened, 0 = did not). A perfect score is 0. A score equal to 0.25 means the predictions are no better than random guessing. Lower is better. The MKDC model logs all predictions publicly at the run date and calculates the Brier score at the 90-day calibration checkpoint: June 18, 2026 for Sub-Batch A. The full calibration record is public at mkdcbaseball.com/calibration.
Follow the Model
In Real Time.

You understand how the CUP engine works. The weekly update shows the model updating in real time, probability revisions, formal checkpoint notes, and the variables that are actually moving the numbers.

✓ You’re in. First update goes out this week.
Weekly. Free. No spam. Unsubscribe anytime.
Follow the Model
In Real Time.

You understand how the CUP engine works. The weekly update shows the model updating in real time, probability revisions, formal checkpoint notes, and the variables that are actually moving the numbers.

✓ You’re in. First update goes out this week.
Weekly. Free. No spam. Unsubscribe anytime.
Weekly Update · Free
Follow the Model
In Real Time.

You understand how the CUP engine works. The weekly update shows the model updating in real time, probability revisions, formal checkpoint notes, and the variables that are actually moving the numbers.

  • Formal CUP score revisions with the reasoning behind each change
  • Calibration notes when outcomes deviate from model predictions
  • PPI multiplier updates as the April 9 and May 1 checkpoints close
  • Sub-Batch B announcement when the next model run launches
✓ You’re in. First update goes out this week.
Weekly. Free. No spam. Unsubscribe anytime.