Five Questions Before Any Prop Bet: The 60-Second Checklist | DMP Learn

Five Questions Before Any Bet

What You'll Learn

This lesson distills the market evaluation framework into a quick five-question checklist you can run through in under 60 seconds before placing any prop bet. If a prop fails on two or more questions, either skip it or significantly reduce your stake.

---

The Checklist

Question 1: Is this a two-way market?

What you're checking: Can you see both sides of the bet (Over/Under) with transparent pricing?

Why it matters: Two-way markets let you calculate the true vig and devig the line to find fair probability. One-way markets (anytime TD, anytime HR, first scorer) hide the vig — typically 20-40% — making it nearly impossible to determine whether you're getting a good price.

Pass: Over/Under props on points, yards, rebounds, assists, strikeouts. You can see both sides and compute the vig.

Fail: Anytime scorer markets where only "Yes" is prominently priced. First/last TD scorer. HR derby-style props.

What to do if it fails: You can still bet it, but know that the hidden vig is your biggest enemy. Your edge estimate needs to be much larger (15%+ rather than 3-5%) to overcome the structural disadvantage. Size your stake accordingly — half-unit or less.

---

Question 2: Does the stat fit a known distribution?

What you're checking: Can you model the probability mathematically, or are you guessing?

Why it matters: When a stat follows the normal distribution or Poisson distribution cleanly, your probability calculation has a mathematical foundation. When it doesn't, your "edge" is based on intuition rather than math — and intuition is what sportsbooks are designed to exploit.

Pass: The stat maps to Normal (passing yards, assists, rebounds) or Poisson (strikeouts, TDs, HRs) with reasonable fit. CV under 35% for continuous stats, VMR between 0.8-1.3 for count stats.

Fail: The player is wildly inconsistent (CV > 50%), heavily overdispersed (VMR > 1.8), or the stat doesn't map to any standard distribution.

What to do if it fails: Reduce confidence in your edge estimate. If you bet, require a much larger apparent edge and use smaller stakes.

---

Question 3: Is the vig transparent and manageable?

What you're checking: Can you quantify the sportsbook's take, and is it small enough that a reasonable edge can overcome it?

Why it matters: Vig is the tax on every bet. On standard -110/-110 lines, total vig is about 4.5% and your break-even edge is roughly 2.3%. On a one-way market with 30% vig, your break-even edge is about 15%. The vig determines the minimum edge you need to be profitable.

How to check:

For two-way markets, calculate the Market Vig:

Market Vig = (IP_Over + IP_Under) - 100%

Where IP = implied probability from the posted odds.

Example: Over -110 / Under -110 → IP = 52.38% + 52.38% = 104.76%. Vig = 4.76%.

Pass: Vig under 6% (standard for two-way prop markets at most books).

Caution: Vig 6-10% (some recreational books juice props more heavily, especially alt lines).

Fail: Vig above 10%, or you can't calculate the vig because it's a one-way market.

---

Question 4: Do I have enough data?

What you're checking: Is your projection based on a meaningful sample, or a handful of games?

Why it matters: A player's 3-game average after returning from injury isn't a reliable projection. Neither is a pitcher's K rate over 2 starts. Small samples produce wildly unstable means and standard deviations, which means your probability estimates are unreliable even if the distribution fits.

Minimum thresholds:

• Individual player projection: At least 10-12 recent games in a similar role. Below this, your mean and SD estimates have large error bars.
• Matchup context: The opponent adjustment should be based on their full-season sample, not just recent games.
• VMR calculation: At least 12 games to get a meaningful VMR reading. Below 10, the VMR fluctuates too much to be diagnostic.

Pass: 12+ games of relevant data for the player, full-season opponent data available.

Fail: Fewer than 8 games of data (new season start, position change, return from long injury), or a totally new context (team trade, scheme change) that makes historical data unreliable.

What to do if it fails: Either wait for more data or acknowledge that your edge estimate is speculative. Size stakes smaller accordingly.

---

Question 5: Is my edge larger than the vig?

What you're checking: After everything above — is the bet actually +EV?

Why it matters: This is the final gate. You can have a great market, a clean distribution fit, transparent vig, and plenty of data — but if the sportsbook's price already reflects the true probability, there's no bet.

How to check:

1. Devig the sportsbook's line to get their implied fair probability
2. Calculate your estimated true probability (using the appropriate distribution)
3. Subtract: Edge = Your Probability - Book's Fair Probability

Pass: Your edge estimate exceeds the half-vig hurdle. On a standard 4.5% vig market, you want at least a 2.5% edge. On a 10% vig market, you want at least 5%.

Fail: Your edge is smaller than half the vig, or it's negative. Pass on the bet.

Borderline: Edge is exactly around the half-vig threshold. If the market is Tier 1 (low variance, high modelability), this can be worth a half-unit bet. If Tier 2 or 3, pass.

---

Putting It Together

Here's how the checklist works in practice:

Example 1: NBA assists Over 7.5 at -108

1. Two-way market? Yes — Over/Under with both sides posted. ✓
2. Known distribution? Yes — assists follow normal distribution cleanly for starting PGs. CV = 22%. ✓
3. Vig transparent? Yes — can devig the -108/-112 line. Total vig ≈ 4.2%. ✓
4. Enough data? Yes — 45 games played this season, stable role. ✓
5. Edge larger than vig? Calculate: adjusted mean = 8.2, SD = 2.1, P(Over 7.5) = 63.1%. Book implies 58.5%. Edge ≈ 4.6%. Yes — exceeds the ~2.1% hurdle. ✓

Result: 5/5. Standard unit bet.

Example 2: Anytime TD scorer at -125

1. Two-way market? No — one-way pricing, No side is +220 with massive embedded vig. ✗
2. Known distribution? Partially — Poisson applies but VMR = 1.45 (overdispersed). Caution.
3. Vig transparent? No — estimated hidden vig is 28%. ✗
4. Enough data? Yes — 14 games this season with consistent usage. ✓
5. Edge larger than vig? Your model says P(TD) = 52%, book implies ~55.6%. You're actually on the wrong side. No. ✗

Result: 2/5. Skip.

Example 3: MLB pitcher strikeouts Over 6.5 at -115

1. Two-way market? Yes — Over -115 / Under -105. ✓
2. Known distribution? Yes — Poisson with VMR = 1.08. ✓
3. Vig transparent? Yes — vig ≈ 4.8%. ✓
4. Enough data? Marginal — only 7 starts this season (early season). Caution.
5. Edge larger than vig? Calculated P(Over 6.5) = 56.2%. Book implies 53.5%. Edge ≈ 2.7%. Exceeds ~2.4% hurdle but data confidence is low.

Result: 4/5 with one caution. Half-unit bet if you trust the small sample, or wait for 2-3 more starts.

---

The Speed Version

After you've internalized this checklist, it becomes almost instantaneous:

1. Two-way? (glance at the pricing structure)
2. Distribution fit? (do I know what math to use?)
3. Vig okay? (standard juice or inflated?)
4. Data? (enough games in this context?)
5. Edge? (does my number beat their number by enough?)

Most bad bets fail on Question 1 (one-way vig) or Question 5 (no actual edge). Training yourself to check these first saves the most time and money.

---

How DMP Handles This

DMP runs a version of this evaluation for every prop in its pipeline. The 14-signal scoring system assesses market structure, statistical reliability, data availability, and edge magnitude before surfacing a play as +EV. When DMP flags a prop as a recommended bet, it's already passed these filters at a quantitative level.

That said, DMP can't evaluate everything — injury news that breaks after lines are set, late lineup changes, or context that the data doesn't capture. Using this checklist alongside DMP's recommendations gives you an extra layer of quality control.

---

Key Takeaways

• Run five questions before any bet: two-way market, distribution fit, transparent vig, enough data, edge exceeds vig.
• If a prop fails two or more questions, either skip it or reduce stakes significantly.
• Most bad bets fail on Question 1 (hidden vig) or Question 5 (no real edge).
• The checklist takes under 60 seconds once internalized and prevents the most common betting mistakes.
• Use this alongside DMP's recommendations as a personal quality filter.

---

Next lesson: Line Shopping: The Easiest Edge in Sports Betting →