HUD Basics Guide: 7 Core Metrics, Sample Sizes & Population Baselines

TL;DR: A HUD turns your hand history database into live opponent reads. Seven stats (VPIP, PFR, 3-Bet %, Fold to 3-Bet, Flop C-Bet, Fold to Flop C-Bet, and AF) cover about 80% of what you need. VPIP and PFR become useful after roughly 20-30 hands, while 3-bet and fold-to stats need 200 to 1,000+ hands to be trustworthy. Compare every number to population baselines, and never let a tiny sample talk you out of fundamentals.

How Do Hand History Trackers and HUDs Work?

Every time you play online, your poker client writes a text file for each hand: seats, stacks, actions, bet sizes, and showdowns. A tracker such as PokerTracker or Hold'em Manager imports those files into a database and tags every action. Did this player raise first in? Did he face a 3-bet and fold? Did he c-bet the flop as the preflop aggressor?

The HUD is the overlay that reads that database in real time and prints small numbers beside each player's name. The tracker stores the data, and the HUD shows it at the table.

Every stat is a fraction: times the player did X ÷ times he had the chance to do X. That denominator is the whole sample-size story. A stat built on 6 opportunities is a rumor, and one built on 60 is a read. Keep that in mind through everything below.

The Preflop Trio: VPIP, PFR, and the Gap

VPIP (Voluntarily Put Money in Pot) is the percentage of hands where a player calls or raises preflop. Posting a blind doesn't count, and neither does a free check in the big blind. It measures looseness.

PFR (Preflop Raise) is the percentage of hands where he raises or 3-bets before the flop. It measures preflop aggression.

The two numbers matter more together than apart:

• 22/18 (VPIP/PFR): a standard regular. The gap is small and most hands are played aggressively.
• 40/8: a calling station. He plays lots of hands and rarely raises.
• 12/10: a nit. You can fold to his aggression and steal his blinds.
• 35/28: a loose-aggressive maniac. Expect wide ranges and plenty of pressure.

A gap wider than 6-7 points usually means passive, limp-and-call tendencies. This is the cheapest leak to spot and exploit.

3-Bet % and Fold to 3-Bet: The Pressure Stats

3-Bet % is how often a player re-raises when facing an open. A typical 6-max regular sits around 6-8%. Below 4% you are looking at a value-heavy range, so fold your marginal hands to his 3-bets. Above 10% you can start 4-betting lighter or calling wider in position.

Fold to 3-Bet is how often he folds after opening and facing a re-raise. Population players fold about 58-68%. Anyone above 70% is begging to be 3-bet light, and anyone below 45% should only get your value hands.

These two stats drive preflop exploitation, because they tell you which of your 3-bet bluffs print money and which light up the board. To see why certain hands make better 3-bet bluffs than others, review the range analysis foundations at PokerMath.org.

Flop C-Bet, Fold to Flop C-Bet, and Aggression Factor

Flop C-Bet % is how often the preflop aggressor bets the flop. The healthy range is roughly 55-68% in single-raised pots. Anyone above 80% is firing automatically, so call or raise him wider. Anyone below 40% only bets when strong, so respect his bets and attack his checks.

Fold to Flop C-Bet shows how often a player gives up when you bet. A typical figure is 45-55%. If he folds 65% or more, c-bet nearly any board against him. If he folds 30%, drop the bluffs and bet for value.

Aggression Factor (AF) is (bets + raises) ÷ calls postflop. AFq (aggression frequency) is (bets + raises) ÷ (bets + raises + calls + folds). Most regulars land at AF 2.0-3.0 and AFq 45-55%. An AF under 1.0 means a passive player who calls too much. An AF over 4.0 means a player who bets nearly every street, which often signals bluff-heavy lines worth studying through showdown data analysis.

How Many Hands Do You Need Before a Stat Means Anything?

Beginners ask this, and sharps think about opportunities rather than raw hands. A stat only updates when the player gets the chance to act in that spot, and those spots vary enormously in frequency.

• VPIP and PFR: every hand counts as an opportunity. After 20 hands you can identify the extremes: a 70% VPIP or a 5% VPIP is real. The middle is still fuzzy until about 100 hands, and the numbers become reliable around 300-500.
• 3-Bet %: the opportunity only arises when someone opens in front of him, roughly one hand in six or seven. At 200 hands you have about 30 chances, which means two or three actual 3-bets. Treat it as directional only.
• Flop C-Bet: he has to open, get called, and see a flop heads-up or multiway. In 100 hands you might see only 5-8 chances, so wait for 300+.
• Fold to 3-Bet: he must open and then face a re-raise. That happens about once per 100 hands. You need 1,000+ hands for a solid read. Before that, use the 60% population average as your default.
• AF: it swings wildly early. Don't trust it until 500+ hands.

Most trackers show the sample in small print, such as a "7/9" fraction. Learn to read it. A 100% c-bet over 3 opportunities isn't a read.

What Are the Population Baselines?

A stat only means something relative to the pool you are playing. These are rough 6-max No-Limit cash baselines at micro and small stakes. Adjust them for your own database.

• VPIP: regulars 21-26%, recreational players 28-45%
• PFR: regulars 17-22%, recreational players 8-15%
• 3-Bet %: population average 5-7%, regulars 7-9%
• Fold to 3-Bet: 58-68%
• Flop C-Bet: 55-68%
• Fold to Flop C-Bet: 45-55%
• AF: 2.0-3.0 (AFq 45-55%)

Full-ring games run tighter, so subtract about 3-5 points from VPIP and PFR. Higher stakes compress everything toward the regular range. The deviations tell you where to attack. A player at 15 points above the baseline is a bigger leak than one at 3 points above it, and bet sizing patterns add another layer, as covered in our guide to bet sizing tells.

Color-Coding and Table Layout

Raw numbers are slow to process when you are playing four tables. Color-coding makes the read instant. Here is a simple scheme:

• Red: a leak you can exploit hard (VPIP above 40, fold to 3-bet above 75, fold to c-bet above 65).
• Orange or yellow: mild deviations from baseline.
• Green: a stat in the normal range, or one you are unsure about.
• Blue or purple: an unusually tight or strong profile. Proceed with caution.

Keep the HUD small. Six to eight stats per player is plenty. A typical layout is VPIP/PFR/3-Bet on the first line, then Fold to 3-Bet, C-Bet, Fold to C-Bet and AF below it, with hands played in a corner. Positional stats such as steal percentage and blind defense can wait until you have mastered the basics. Fade the color intensity when the sample is under 30 opportunities so a premature read doesn't mislead you.

Avoiding Small-Sample Paralysis

The failure mode I see most often is players who stare at a HUD for 30 seconds, second-guessing a 12-hand sample. That is paralysis, and it costs more than ignoring the HUD entirely.

Here is how sharps handle it:

1. Start with a default. Assume baseline numbers for every stat until you have data.
2. Use VPIP/PFR first. They converge fastest.
3. Only deviate from fundamentals when the sample supports it. A 20-hand read justifies folding marginal hands against a 6/5 nit. It doesn't justify a 4-bet bluff.
4. Use notes and showdowns. A single observed bluff or tank-fold can outweigh a thin stat.
5. Spend most of your time on your own hand and the board. The HUD informs your decision, but it doesn't make it for you.

Review your database away from the tables as well. Filter by position and stack depth to find your own leaks, then see whether the numbers match your gut feel for opponents.

What Is the Bottom Line?

Seven stats do the heavy lifting: VPIP, PFR, 3-Bet %, Fold to 3-Bet, Flop C-Bet, Fold to Flop C-Bet, and AF. Learn them in pairs, learn the population baselines, and match each stat to the number of opportunities it needs. VPIP and PFR identify extremes in 20 hands, while 3-bet and fold-to-3-bet reads need hundreds or thousands. Use color-coding to read the table quickly, fall back on population defaults when the sample is thin, and put your effort into sound fundamentals before chasing exploits. A HUD is a tool for adjusting your strategy, and it works best once you already have a solid one.