Methodology

How Bobby's Picks Work

Every NCAABB pick is driven by data, not gut feel. We combine ESPN BPI predictions, Torvik efficiency ratings, DraftKings market odds, and team-level analytics to surface the games where our models disagree with the market — and that disagreement is your edge.

Our Data Sources

ESPN BPI

Win probability predictions powered by ESPN's Basketball Power Index. BPI rates every team based on performance, strength of schedule, and game-by-game simulations.

Torvik Rankings

Adjusted offensive and defensive efficiency, tempo, strength of schedule, and predictive ratings from Barttorvik — one of the most respected college basketball analytics sources.

DraftKings Odds

Real-time moneyline, spread, and total lines from DraftKings. Market odds reflect the consensus view and serve as the benchmark we measure our edge against.

Team Stats

Season averages, recent form, home/away splits, injury reports, and key player metrics. Context that turns raw numbers into actionable predictions.

The Process

1

Fetch Today's Games & Odds

We pull the full slate of NCAABB games from ESPN and match them against real-time DraftKings odds — moneyline, spread, and totals — to build a complete picture of each matchup.

2

Compare BPI vs Market Probabilities

ESPN BPI assigns each team a win probability. We convert DraftKings moneyline odds into implied probabilities (removing the vig) and compare the two. When BPI and the market disagree, that's where value lives.

3

Analyze Matchup Factors

We break down each game across key dimensions: offensive and defensive efficiency, shooting percentages, rebounding margins, tempo compatibility, and recent performance trends.

4

Generate Confidence Scores

Using Torvik efficiency gaps, BPI edge magnitude, key matchup factors, and market alignment, we assign each pick a confidence grade. The more data sources that agree, the higher the grade.

5

Rank & Present Picks

Picks are ranked by confidence and presented with full reasoning — the data behind every recommendation is transparent so you can make informed decisions.

Understanding Grades

Each pick receives a letter grade reflecting how many data sources converge and the magnitude of the detected edge. Higher grades mean more signals agree and the edge is larger.

A+
Elite

Multiple strong signals align: large BPI edge, favorable Torvik efficiency gap, key matchup advantages, and market undervaluation. Our highest-conviction picks.

A
Strong

Clear edge supported by BPI and at least two additional factors. Strong analytical foundation with meaningful market disagreement.

B+
Above Average

Solid edge with good supporting data. Typically a moderate BPI advantage backed by favorable matchup dynamics or efficiency metrics.

B
Playable

A detectable edge with some supporting factors. The data leans in one direction, but the margin is thinner and fewer signals converge.

C+
Lean

A slight analytical lean. The data points in a direction, but the edge is small and the confidence level is lower. Proceed with caution.

Edge Finder

BPI vs DK Implied Probability

The Edge Finder compares ESPN BPI's predicted win probability for each team against the implied probability derived from DraftKings moneyline odds (with vig removed). When BPI says a team has a significantly higher chance of winning than the market implies, we flag it as a value opportunity.

For example, if BPI gives a team a 65% win probability but DraftKings odds imply only 55%, that's a +10% edge — the market is undervaluing that team relative to our model's assessment.

Disclaimer

Our analysis is for informational purposes only and does not constitute financial or gambling advice. Past model performance does not guarantee future results. Sports betting involves risk, and you should only wager what you can afford to lose. Always gamble responsibly.