PropBetEdgeLearnSports Intelligence Academy
PropBetEdge
PropBetEdgeLearnSports Intelligence Academy

Sport academies

Field & court

MLBNFLNBAWNBANHLSoccer

Individual

Golf

Racket

Tennis

Combat

UFC / MMABoxing

Network

PropBetEdge.ai PBE All Access
Lessons, glossary and tools from Learn · players, teams, tools and stories from PropBetEdge Search↑↓ to move · Enter to open

Model Lab

Calculators that show their work.

7 tools for the math behind every price and model output. Each one explains its result, asks how uncertain your inputs are, and lists its limitations. Nothing here tells you what to bet.

Tool 01

Expected value

The average result of a stake at a given price, if your probability is correct — plus the range of results if your probability is off.

Why it mattersPrice alone never tells you whether a bet is good. EV combines the price with a probability, and the uncertainty band shows whether the edge survives your own doubt.

Inputs

From a model or your own estimate.
For example +280 or −110.
How far off your probability could reasonably be. Few probabilities are better than ± 3 points.

Result

—expected profit per stake, if your probability is right
Enter values to see the result explained.
Limitations
  • EV is an average over many repetitions; a single result tells you almost nothing.
  • The probability you enter is the whole calculation. If it is biased, the EV is too.
  • The price you can actually get may differ from the one displayed, and limits apply.

Tool 02

Odds converter

Converts between American, decimal and fractional odds and implied probability, in any direction.

Why it mattersModel outputs are probabilities and prices are odds. Converting puts them in the same unit so they can be compared.

Inputs — edit any field

+150 pays $150 profit on $100 and needs to win 40.0% of the time to break even.

Quick reference

AmericanDecimalImplied
+3004.0025.0%
+2003.0033.3%
+1502.5040.0%
+1002.0050.0%
−1101.9152.4%
−1501.6760.0%
−2001.5066.7%
Limitations
  • Implied probability still contains the book’s margin — use the break-even tool with both sides to remove it.
  • Displayed prices are rounded, so conversions can differ slightly from a book’s internal price.

Tool 03

Break-even and vig

The win rate a price requires, and — if you add the other side’s price — the market’s real overround and no-vig fair probability.

Why it mattersAny hit rate is meaningless without the prices behind it, and the vig decides how high the bar is.

Inputs

Add it to measure the real vig.

Result

—win rate needed to break even
Enter values to see the result explained.
Limitations
  • Break-even is a long-run figure; short samples scatter widely around it.
  • No-vig estimates assume the margin is spread evenly across both sides. Books often load more margin on long shots.

Tool 04

Kelly sizing

The Kelly stake fraction for a probability and price, fractional Kelly options, and what happens to the stake if your probability is too high.

Why it mattersSizing decides whether you survive variance. Kelly is only optimal when the probability is exactly right, so seeing its sensitivity matters more than the headline number.

Inputs

How far off your probability could reasonably be. Few probabilities are better than ± 3 points.

Result

—suggested stake at the selected fraction
Enter values to see the result explained.
Limitations
  • Kelly assumes your probability is exactly right. It never is, so treat full Kelly as a ceiling, not a target.
  • It assumes one bet at a time and independent outcomes. Simultaneous or correlated bets need smaller stakes.
  • This is arithmetic, not advice on whether to bet.

Tool 05

Closing line value

How your price compares with the closing price: payout difference, implied-probability gap, and — with the other side’s close — no-vig expected ROI.

Why it mattersCLV measures the quality of the price you took without waiting for noisy results.

Inputs

Needed to remove the vig from the close.

Result

—better (or worse) payout than the close
Enter values to see the result explained.
Limitations
  • CLV assumes the close is a good estimate of the truth. In thin prop markets it can be noisy or stale.
  • One bet’s CLV means nothing; hundreds start to mean something.
  • v1 of this tool showed an “annualized ROI” that assumed one bet a day at a constant edge. It was removed because it presented a guess as a result.

Tool 06

Parlay math

Combined probability, price, payout and EV for 2–6 independent legs.

Why it mattersProbabilities and margins both compound across legs. Seeing the combined numbers is the fastest way to understand why.

Legs

2–6 legs

Result

—expected profit per stake, if every leg probability is right
Enter values to see the result explained.
Limitations
  • Assumes the legs are independent. Correlated legs (same game, same player) win together more or less often than the product suggests.
  • Every leg’s probability error compounds.

Tool 07

Monte Carlo bankroll simulator

Hundreds of simulated bankroll paths for a win rate, price, stake size and number of bets, with percentile bands and drawdowns.

Why it mattersEV tells you the average; simulation shows the path — how wide the range of outcomes is and how deep ordinary losing runs get.

Inputs

53%
1.95
500
300
2%
Stake basis

Result

Median path10th–90th percentilePaths above startPaths below startStarting bankroll
Limitations
  • Assumes every bet has the same true probability and price, and that outcomes are independent. Real edges vary and are uncertain.
  • “Ruin” here means finishing below 10% of the starting bankroll.
  • Simulations show the range of outcomes under your assumptions; they cannot tell you whether the assumed win probability is real.