This is a metric dictionary and a dated policy comparison, not a Poker-Eye feature announcement. We have not verified that Poker-Eye supplies a HUD, captures a particular room's histories, or is approved on any room. For the metric formulas below, the named reference is PokerTracker's PT3 statistical reference; PT4's custom-statistics guide shows how a later installation exposes each stat's own formula and details. Do not assume two products or versions use identical filters just because their abbreviations match.
Data Source Comes Before the Number
A HUD may calculate from hands you played and recorded, a platform's own integrated data, or another permitted source. It may filter by game, stake, seat count, position, date or tournament format. A ratio is meaningful only after you know which hands entered its numerator and denominator. For each stat, record the product and version, game/filter, sample date, total relevant hands, and the actual action opportunities. If a room or software changes its history format, the same label may not be computed from the same records.
An invented display VPIP 24 / PFR 18 / 3Bet 10 is not a personality diagnosis. It is shorthand for three counts with potentially different denominators. It does not reveal all prior actions, board textures, opponent adaptation or whether a player was multitabling. A third-party overlay is also distinct from a room's own integrated display; its permission must be checked separately.
Six Common Metrics and What Each Divides By
The formulas below reproduce the named PT3 reference, not a universal cross-vendor schema. When a denominator is zero, a percentage should be undefined, not silently displayed as zero. When different streets or filters are combined, the product's definition controls. The PT4 guide explicitly tells users to inspect the stat's Details page for its formula.
| Metric | Numerator in named reference | Denominator | What to keep beside the percentage |
|---|---|---|---|
| VPIP or VP$IP | Hands with a voluntary preflop contribution | Hands played in the reference definition | Posted blind alone does not count; inspect walks and import/filter behavior in the actual version. |
| PFR | Hands in which the player raised preflop | Hands played | A raise can also be a VPIP event; the two percentages are not complementary. |
| Preflop 3-bet | Times the player re-raised a preflop raiser | Preflop 3-bet opportunities | Not all dealt hands offer a chance to 3-bet. Specify preflop, not an all-streets “3Bet Total” field. |
| Total AF, aggression factor | Postflop bets plus raises | Postflop calls | A ratio, not a percentage. If calls equal zero, the ratio is undefined or software-specific. |
| WTSD, went to showdown | Times the player reached showdown | Times the player saw the flop | A postflop conditional rate; not showdowns divided by every dealt hand. |
| W$SD, won money at showdown | Showdowns with money won in the named reference | Showdowns reached | The named PT3 reference includes split pots where some money was won; inspect another vendor's version separately. |
The PT3 guide's “3Bet Total” includes multiple streets; this article's preflop row is narrower. Its AF formula counts bets plus raises over calls, whereas AFq, aggression frequency, is a different percentage with bets, raises, calls and folds in the denominator. Collapsing them under “aggression” without the formula can reverse an interpretation. A player who has bet 12 times, raised 4 and called 8 has AF (12+4)/8 = 2.0. Those same raw counts do not determine AFq until the relevant folds are known.
A Transparent Synthetic Stat Line
Here is a teaching ledger, not a real player's record or a Poker-Eye screen. Assume the product's filters have produced 100 played hands, 24 voluntary preflop contributions, 18 hands with a preflop raise, 20 preflop 3-bet opportunities and two actual preflop 3-bets. Suppose it also records 40 flops seen, 12 showdowns, seven of those showdowns as wins under the named metric definition, 12 postflop bets, four postflop raises and eight postflop calls. The event counts are stipulated for formula practice; they have not been reconstructed from 100 actual hand histories.
| Metric | Calculation | Display from this toy ledger |
|---|---|---|
| VPIP | 24 / 100 | 24% |
| PFR | 18 / 100 | 18% |
| Preflop 3-bet | 2 / 20 | 10% |
| WTSD | 12 / 40 | 30% |
| W$SD | 7 / 12 | 58.33% |
| Postflop AF | (12 + 4) / 8 | 2.0 ratio |
If the player had only one 3-bet in ten opportunities, the HUD would still show 10%; the underlying evidence would be half as large. A percentage hides that difference. Conversely, VPIP and PFR each use a broad hand count in this named reference, but position or game-format filters can leave a much smaller subset. Read both the count and the filter.
Why Small Samples Change How You Read the Display
Two raises from twenty opportunities produce 10%, but one different action would produce 5% or 15%. One action moves this sample by five percentage points. Two hundred opportunities with twenty raises also produce 10%; one action then moves the displayed rate by half a percentage point. This is arithmetic sensitivity, not a statistical confidence interval or a threshold for “enough hands.” The true action tendency may also vary with seat, table and prior action rather than being one fixed rate.
Postflop stats can have far fewer relevant events than the headline hand count. A display may show 1,000 imported hands while a particular river action had only eight opportunities. Do not present the eight-event percentage as a stable read solely because the HUD header says “1,000 hands.” For an off-table review, preserve the opportunity count, uncertainty and a few concrete hands that explain what was measured. For VPIP, check which played hands enter the denominator and how the tracker handles a big-blind walk; this page compares several metrics without treating them as interchangeable.
Room Policy: A Stat Definition Is Not a Usage Permit
Operator rules vary by tool, data source and whether the client is running. The following is a dated reading of the linked official tool policies, rechecked on 22 October 2026, not legal advice, a compliance promise or approval of a named HUD. A stat definition and a permission to display it are separate questions.
| Operator | What its official policy says about this use case | What a player must not infer |
|---|---|---|
| PokerStars | Its permitted category includes limited in-game numerical statistics using information accumulated through the player's own play, with qualitative and feature restrictions. Current-state action advice is prohibited, and advanced calculation/reference tools have client-open restrictions. | “PokerStars permits every HUD” or “Poker-Eye has been approved.” Check the exact feature against the current policy. |
| GGPoker | Its Security & Ecology Policy, version displayed 20260313, names HUDs among prohibited auxiliary tools that aid actions or influence gameplay; it also prohibits charts during play. It says tools that do not support or interact with GGPoker during gameplay will not be flagged under that auxiliary-tools passage. | “Any offline analytics product is banned” or “a third-party live HUD is allowed.” The policy is use-context-specific. |
| CoinPoker | Its HUD section prohibits third-party overlays and opponent analysis connected to CoinPoker gameplay, while distinguishing integrated PokerIntel/HUD features. The page displays 25 June 2026. | An integrated display does not approve an external overlay. The HUD study/other-platform wording and broad client-open FAQ do not certify a specific simultaneous-use setup; ask the operator. |
The 2026 CoinPoker terms, pages 4–5 and 21, prohibit HUD use connected to CoinPoker gameplay and include a separate other-platform HUD clause. Read that clause alongside the dedicated policy's second-monitor FAQ, which gives a broader client-open restriction on prohibited tools. Seek written clarification for a specific simultaneous-use setup; this comparison establishes neither Poker-Eye compatibility nor approval.
A Practical Off-Table Reading Sequence
First, identify the stat's vendor, version and filter. Next, read the numerator and opportunity denominator; if the denominator is hidden, open the hand report rather than treating the percent as a fact about skill. Then compare related stats with their separate meanings: VPIP versus PFR describes voluntary entry and raising, not a complete range; 3-bet uses only 3-bet opportunities; AF is a ratio, not a percentage. Our poker terms guide is a vocabulary reference, not a substitute for a vendor's stat definition. Finally, inspect actual hands and context. A large sample of the wrong game or position is not an answer to the hand in front of you.
For a post-session analysis of those hands, use the hand-review process; a HUD percentage is evidence to investigate, not the review itself.
This sequence is for study, not a direction to keep prohibited software open while playing. PokerStars places certain advanced study tools under a client-open restriction; GGPoker and CoinPoker define their own live-use boundaries. When uncertain, ask the operator about the specific tool and feature before using it. Do not infer permission from a generic article or from another room's rules.
Quick Answers
Is VPIP the same as PFR? No. VPIP counts voluntary preflop contribution; PFR counts hands with a preflop raise under the named reference. Raising usually contributes to both, but calling or limping can raise VPIP without raising PFR.
Does a 10% 3-bet stat mean ten of every hundred dealt hands? No. In the preflop definition used here, it is raises divided by 3-bet opportunities. Two of twenty opportunities yield 10% even if the player was dealt many more hands.
Is AF 2.0 a 2% aggression frequency? No. AF is a bets-plus-raises-to-calls ratio. AFq is a separate percentage with a different denominator in the PokerTracker reference.
Can I use a third-party HUD on every site? No. The three official policies above differ. The page teaches metric definitions; it does not certify a tool or grant permission.
Sources and Scope
- PokerTracker 3: Statistical Reference Guide — Historical PT3 v3.10 definitions for VPIP, PFR, 3-bet, AF, WTSD and W$SD; different versions and filters must be checked separately.
- PokerTracker 4: Custom Statistics Guide — Shows where to inspect a statistic's formula and Details in PT4 instead of assuming that a shared abbreviation guarantees the same calculation.
- PokerStars: Prohibited Programs, Tools and Services — The numerical-statistics and client-open categories define this operator's conditions; no particular third-party HUD is certified here.
- GGPoker: Security & Ecology Policy — Version 20260313, section 2 and the auxiliary-tool section, covers external decision assistance, charts and auxiliary tools including HUDs.
- CoinPoker: Third-Party Tools & Software Policies — The HUD, RTA and training-tool sections distinguish gameplay-connected HUD use, integrated features and client-open restrictions. Rechecked 22 October 2026.
- CoinPoker: Terms and Conditions (2026), pp. 4–5, 21 — Pages 4–5 define gameplay-connected HUD restrictions and the other-platform HUD clause; page 21 addresses live RTA/solvers and HUD use. These are distinct from the policy FAQ's client-open wording. Directly rechecked 22 October 2026.
The stat ledger is synthetic. Vendor definitions and operator policy are different kinds of evidence; neither the sample nor this source list establishes Poker-Eye compatibility or approval.
About the Editorial Team
The Poker-Eye Editorial Team researches, checks, and maintains product and educational content. Sources and product boundaries are reviewed before publication.