All-in adjusted results: what the graph omits

An all-in adjustment replaces a defined result with an equity-based accounting entry. Read the definition before interpreting the distance between two lines. Which hands qualify? When is equity measured? What pot and contribution are counted? A graph can answer a useful question about recorded runouts while leaving earlier decisions and other hands unexamined.

This lesson uses a fixed fictional ledger. The equity shares are supplied, the units are chips, and no rake or fees are deducted. You can inspect every replacement, retain ordinary results and expose unresolved records. The ledger is an independent teaching model, not a tracker replica, hand-equity calculator or measure of player skill.

Indira: “Where does this calculation begin?” Ace: “At the chosen all-in point.” Indira: “Mark that point before looking at the ending. It determines which uncertainty this entry replaces.”

Define the statistic · Read the receipt · Expose missing records · Audit the ledger · Report the limits

Start with the statistic and its coverage

PokerTracker 4’s published All-in Equity guide describes an actual-card calculation at the all-in point. It subtracts the player’s total contribution from the equity share of the pot and retains ordinary results for non-all-in hands. Its guide also describes unknown-card exclusions and exclusions for unequal-stack multiway situations. Those are statements about that documented profile, not a promise that every tracker uses identical rules.

For any other report, check the current definition, supported hand histories and filters for that product. A line called “EV” may be displaying a narrower statistic than the word suggests. Do not infer the calculation from its color or from a label on someone else’s screenshot. First establish what a single point on the line represents.

The teaching ledger below has its own explicit rule. Two eligible heads-up records receive supplied-equity replacements. A non-all-in record passes through at its realized result. Two other records remain unresolved because this model cannot support their replacements. The output labels the selection and missing entries rather than trying to mimic a software database.

Keep the measurement point and the accounting inputs together

For an eligible record in this model, let P be the final contested pot, I be your total contribution to that pot and e be your supplied equity share at the chosen point. The adjusted net entry is e × P − I. The first term is an expected gross receipt; subtracting the contribution converts it to a net result.

Here, qualifying means a fictional heads-up all-in with the relevant commitments fixed, a known contested pot and a supported supplied equity share. No further wager can change this pot, and there are no side pots, uncalled excess, rake or fees in the eligible examples. These boundaries keep the arithmetic inspectable. They are not universal eligibility rules for tracking software.

I includes your earlier contributions to the represented pot. It is not just the final call you made. P includes the matched money in that same pot, including your contribution. Keep the two fields under different labels. Subtracting the final call alone from a whole-pot receipt would leave earlier investments out of the net entry.

Equity share describes a share of a specified pot under a specified information set. The exercise supplies 60% and 25% rather than estimating either value from cards. No pictured cards, character scene or result table establishes those percentages. In a real audit, retain the source of the equity figure and the point in the hand to which it belongs.

Read the gross receipt and net entry separately

Record A has a final pot of 200, your total contribution of 100 and supplied equity of 60%. The equity share of the pot is 0.60 × 200 = 120. The adjusted net entry is 120 − 100 = +20. A gross receipt of 120 is not a profit of 120 because it includes the return of invested chips.

The fixed ledger records an actual loss for A, so the realized net result is −100. Define the gap consistently as realized minus adjusted. Here it is −100 − 20 = −120. If you reverse the subtraction, you reverse the sign. Always state which direction a number uses before describing a line as above or below another line.

If the same modeled record instead ended with a win, its realized net result would be +100 and its adjusted entry would still be +20. The gap would become +80. Those two gaps have different sizes because the supplied share is 60%, not 50%. The altered ending does not alter the already specified pot, contribution or equity input.

There is no promise that a future win will repay that −120 gap. The entry describes this recorded comparison. It does not create a balance that later cards owe you. If you want to study the decision that led to the all-in, return to the action and alternatives available at that decision rather than treating the gap as an instruction to keep playing.

Add the other records before describing the total

Record B has a final pot of 100, contribution of 50 and supplied equity of 25%. Its gross share is 25 and adjusted net is 25 − 50 = −25. The fixed outcome is a win, giving realized net +50. Across A and B, realized net is −50 while adjusted net is −5. One selected runout finished below its replacement and the other above it.

Record C is a non-all-in hand with realized net +30. Our teaching rule retains that +30 unchanged. Across A, B and C, realized net is −100 + 50 + 30 = −20. The adjusted total is 20 − 25 + 30 = +25. Realized minus adjusted is −45. All three selected records have a defined entry in this complete ledger.

C changes both totals even though no equity replacement was calculated for it. Passing an ordinary result through an accounting rule does not evaluate the choices in that hand. A profitable bluff can be included as an ordinary win without the ledger ever estimating the range facing it or the expected value of its bet size.

Ace: “The subtotal is neat. Can I call it the session?” Indira: “Only if it covers the records you say it covers. Count what is included and keep the unresolved entries visible.”

Missing replacements must stay different from zero

The incomplete set adds D and E. D has realized net −40 but missing opposing cards. E has realized net −60 and an unequal-stack multiway structure outside this simple model. Their outcomes are recorded, but no adjusted replacements are supplied. That is a coverage limit in the worksheet, not an equity estimate of zero.

Across A through E, realized net is −20 − 40 − 60 = −120. The known adjusted subtotal remains +25 from A, B and C. There are two unresolved records, so the five-row adjusted total is unavailable. Subtracting +25 from −120 would compare a five-row actual total with a three-row adjusted subtotal and conceal the mismatch.

Do not substitute an unresolved replacement of zero, and do not silently copy its ordinary result into the adjusted column. Either action creates a treatment that the model never supplied. A product may have a documented convention for such records; check that convention separately. This worksheet leaves its unknown entries visible and declines to print a complete comparison.

Filtering out D and E produces a supported total for the remaining A–C sample. It does not recover the absent replacements or turn the filtered sample into a complete account of all five records. State both the filter and the count. “Known subtotal +25, two unresolved” tells the reader much more than a single unlabeled positive number.

Later information belongs to a later measurement

Suppose commitments became fixed before a community card, but the equity figure you copied uses that later card. The new figure describes a different information set. Replacing the earlier all-in entry with it changes the question. A later-street analysis can be useful, but it should carry its own timestamp and explanation.

Keep the complete action order beside the measurement. Record when the relevant money became fixed, which cards were available at that point and where the equity figure came from. A final-board screenshot and an unlabeled percentage cannot establish all of that history. Do not move the measurement backward merely because the final cards make an earlier decision easier to explain.

An all-in on the river has no remaining board runout in an ordinary single-board hand. It therefore offers no later community-card uncertainty to replace. That observation does not evaluate how the player reached the river. A correctly described final snapshot can still leave a questionable earlier call or missed bet entirely outside its analysis.

Earlier decisions and non-all-in uncertainty still need review

The model starts from the recorded all-in state. It does not compare every earlier choice with checking, folding or using another size. Reaching a favorable all-in equity share after an unsupported earlier call does not retroactively justify that call. You need the earlier decision point, the price and a defensible range model to study it.

Likewise, the result of a hand that never became all-in can contain uncertainty the ledger does not replace. C’s +30 might come from many possible sequences, none specified here. We cannot infer the quality of a bluff, value bet or fold from that number. The ordinary-result label describes an accounting treatment rather than a verdict.

Write the question before opening a graph. “How do supported all-in runouts compare with their recorded equity shares?” fits this kind of adjustment. “Was my turn raise better than calling?” needs a decision review. Keeping those questions separate prevents the clarity of an accounting total from lending false certainty to an unexamined strategy claim.

Reconcile rake, fees and units before comparing reports

The toy pot is fully distributable because the model deducts no rake or fees. A real report may define a pot or net result differently. Check whether a cost is already removed and whether both columns use the same convention. Do not subtract a charge twice simply because one screen labels it separately.

We have not supplied any fee amounts, so the exercise cannot estimate a fee-adjusted result. You would need the applicable costs and a clear rule for assigning them to the represented records. Keep session-level expenses separate until you have that rule. A missing expense field is another missing input, not evidence that the expense was zero.

All five records use the same fictional chip unit. Do not add tournament chips directly to cash amounts, or mix currencies without an explicit conversion convention. A tournament prize-equity report is also a different model from this chip ledger. It needs its own definitions and data; the controls here do not calculate prizes, ICM, buy-ins or an actual tournament result.

Audit the fixed ledger with a visible selection

Start with Complete and All records, then audit the selection. The output shows −20 realized, +25 known adjusted and +25 complete adjusted. Now choose All-ins with a replacement. You see just A and B: realized −50 and adjusted −5. Switching the filter changes the sample rather than changing the original records.

Choose Incomplete and All records to expose the coverage limit. The realized subtotal becomes −120, the known adjusted subtotal stays +25, and the complete adjusted total and comparable gap are unavailable. The selected-row table identifies D and E. Unresolved records alone show realized −100, known adjusted subtotal zero and no complete adjusted total.

Try No records as well. Its subtotals are empty sums, not evidence of a break-even session. Changing either selector clears stale output. Reset ledger restores Complete and All records, clears the result and returns focus to the record selector. The full fixed accounting remains readable without JavaScript. There is no tracker connection, upload, random simulation or saved personal record.

Keep the finding within the records that support it

A defensible note for the complete set is “Across A–C, realized net is −20 and adjusted net is +25 under the supplied no-fee model. Two all-ins received replacements and one non-all-in retained its result.” That statement identifies the sample and the transformation without claiming that every decision was evaluated.

For the incomplete set, say “A–E realized net is −120; the known adjusted subtotal is +25 with two unresolved replacements.” Stop there until the missing information or a supported accounting convention is available. The absence of a complete adjusted total is part of the finding. Removing the label would make the report look more certain without supplying another fact.

Ace: “What should I study next?” Indira: “Recover the missing record, check the definition or review a specific decision. Pick the task that answers your question. The gap itself cannot do that work.”

Keep the source hands, selected filters, measurement point and cost convention with the result. To examine an earlier decision, use the separate hand history review lesson. For other accounting exercises, return to Info & Tools. A useful graph can organize a review while leaving its unresolved questions visible.

Back to top