Keep a Blackjack Session Log: The Habit That Keeps You Honest

Every serious player I've mentored over the years has eventually had the same uncomfortable moment. They sit down with a pen, try to reconstruct their last three months of blackjack from memory, and announce that they're "roughly breaking even, maybe a little up." Then we pull their deposit and withdrawal history, and the real number is a loss of 60, 80, sometimes 150 units. They weren't lying to me. They were lying to themselves, and they didn't know it.

That gap between what you remember and what actually happened is where bankrolls go to die. The fix isn't more willpower or a better system. It's a session log: a short, disciplined record you complete after every session, without exception. This guide covers exactly what to record, how to compare your actual results against theoretical expectation, and how 50 or more logged sessions will expose leaks you would never find on your own.

Why Your Memory Is the Worst Bookkeeper at the Table

Human memory doesn't store events evenly. Emotionally intense moments are encoded more strongly than routine ones. A session where you split eights against a dealer 10, caught a pair of doubles, and walked away up 35 units produces a jolt of excitement that makes the memory vivid and easy to recall. A session where you slowly bled 20 units over two hours of unremarkable hands produces almost nothing. Six weeks later, the big win is a story you've told three friends. The grind-down loss has dissolved.

Psychologists describe several overlapping effects here. Selective recall means wins surface more easily than losses. Self-serving attribution means wins feel like skill while losses feel like bad luck, a dealer "on a heater," or a bad player at third base, so losses get filed as exceptions rather than data. Rationalization lets you rewrite a loss as "I was actually up 15 at one point," as though the peak counts as the result. Put these together and the average recreational player's mental ledger is biased toward optimism in a way that's both predictable and expensive.

Blackjack makes this worse than most games. Because the house edge on a good game is small, under 1% with correct basic strategy, the outcome of any single session is overwhelmingly driven by variance. Big swings in both directions are normal. That gives your memory plenty of dramatic wins to cherry-pick and plenty of "unlucky" losses to excuse.

The Six Essential Data Fields

You can track dozens of variables, but most players who over-engineer their log abandon it within a month. These six fields are the minimum viable professional journal. Each one earns its place by enabling a specific analysis later.

1. Date, Time and Platform/Casino

Record the date, start time, and exactly where you played, with enough detail to distinguish tables. "Online" is useless. "1win, Evolution live blackjack, table seated 7 players" or "Stake, provably fair RNG blackjack" is useful. Live dealer streams and RNG games play at wildly different speeds and often carry different rules, so lumping them together will corrupt every calculation downstream. Start time matters because time-of-day patterns are among the most common hidden leaks.

2. Table Rules

Write down the rule set every time, even if you always play the same table, because operators change rules and you won't notice unless you're checking. The core variables are number of decks, whether the dealer hits or stands on soft 17 (H17/S17), double after split (DAS), surrender availability, and the blackjack payout (3:2 or 6:5). These rules determine the house edge you'll plug into your expectation formula. If you're unsure how a rule set grades out, run it through our Table Grader tool before you sit down, and log the resulting edge.

3. Base Unit Size and Starting Buy-In

Your base unit is the bet you'd make on a neutral hand. Your buy-in is the amount you allocated to this session. Recording both in dollars and in units lets you compare sessions across different stake levels. If you haven't formalized how many units you bring to a session, read our guide on Session Buy-In Units first; the log assumes you're working from a defined allocation, not "whatever's in the account."

4. Session Duration and Estimated Hands Played

Duration is easy: start time to end time. Hands played is the critical variable for expectation, and most players have no idea what their real number is. Reasonable estimates:

• Live dealer (online or brick-and-mortar): roughly 60–80 hands per hour, lower at full tables or with a slow dealer, higher heads-up.
• RNG/digital blackjack: 200+ hands per hour is common, and fast players exceed 400.

Many online platforms display a hand history. Where they do, use the actual count instead of an estimate. Where they don't, time a sample of 20 hands once and calibrate your per-hour figure for that table type.

5. Final Cash-Out and Net Unit Win/Loss

Cash-out minus buy-in gives your net dollar result. Divide by your base unit to get net units. Units are the language of serious tracking because they normalize results across stake changes: a 20-unit loss means the same thing whether your unit was $5 or $50.

6. Discipline Score (1 to 10)

This is the field that separates a real journal from a scoreboard. Immediately after the session, grade yourself honestly on three questions:

• Did you deviate from basic strategy on any hand, including "feel" plays like standing on 16 against a 10 because the dealer "always has it"?
• Did you stay past your planned time or loss limit?
• Did you chase, meaning raise bets after losses outside your planned spread?

A 10 means you executed your plan perfectly. Deduct points for every violation. The discipline score is valuable precisely because, unlike your net result, it's something you fully control. It should be anchored to the rules you set in your Blackjack Session Plan; without a written plan, there's nothing to score against.

Actual vs Theoretical: Calculating Your Expected Loss

In casino operations, "theoretical hold" is what the house expects to win based on edge and volume, and "actual hold" is what it really won. You should run the same comparison on yourself. The core formula is simple:

Expected Loss = Hands Played × Base Bet × House Edge

Strictly speaking, house edge figures are quoted per initial wager, and your total action is slightly higher because of doubles and splits, but for logging purposes this formula is the standard and it's accurate enough. For quick per-hour figures across different speeds and stakes, use the hourly loss rate calculator on BlackjackMath.org, which handles the arithmetic and lets you compare configurations side by side.

Here's how rules and game speed combine to set your theoretical cost at a $10 base bet, assuming perfect basic strategy:

Two lessons jump out. First, a 6:5 payout roughly quadruples your cost compared with a good 3:2 game. Second, RNG speed multiplies everything. A "cheap" 0.42% game played at 200 hands per hour costs nearly three times as much per hour as the same game live. Your log will make both effects visible in your own numbers, not just in a chart.

A Worked Example

Say you play a 3-hour live session on a typical 8-deck, S17, DAS, no-surrender table with an edge of about 0.46%. At a seven-seat table you average 60 hands per hour, so 180 hands total, at a $10 base bet.

Expected loss = 180 × $10 × 0.0046 = $8.28, less than one unit.

You actually lose $140, or 14 units. Does that mean you played badly? Not necessarily. The standard deviation in blackjack is roughly 1.15 units per hand, so over 180 hands one standard deviation is about 1.15 × √180 ≈ 15 units. A 14-unit loss sits comfortably inside normal variance. The session alone tells you almost nothing about your skill, which is exactly why you need many of them.

What 50+ Sessions Reveal

Variance shrinks relative to expectation as volume grows, but more slowly than most players assume. At 50 sessions of 180 hands, you've played about 9,000 hands. One standard deviation is now about 1.15 × √9,000 ≈ 109 units, while your theoretical loss at 0.46% is around 41 units. Noise still dominates your net result.

That's an important and honest point: a log won't "prove" whether you're a winning or losing player within a few months. What it will do is three things far more valuable.

It Flags Results That Are Too Bad to Be Bad Luck

If your actual losses after 9,000 hands are 350 units worse than theoretical, you're nearly three standard deviations below expectation. That's possible by chance, but unlikely. The more probable explanations are strategy errors, bet sizes creeping above your logged base unit, or rule sets that are worse than you recorded. The log turns a vague feeling of "running bad" into a concrete red flag you can investigate.

It Correlates Discipline With Results

Sort your sessions by discipline score. Because discipline is far less noisy than net results, patterns appear quickly. Most players discover that sessions scoring 6 or below account for a wildly disproportionate share of their total losses, typically because low scores coincide with chasing, which inflates bet size precisely when the bankroll is shrinking.

It Exposes Situational Leaks

Filter and segment your log by the variables you recorded. Common findings include:

• Fatigue thresholds: discipline scores collapse after a specific duration, often around the two-hour mark. That's your natural session cap.
• Time-of-day slots: late-night sessions, especially after a workday, show more deviations and more limit breaches.
• Table conditions: you play noticeably worse at RNG tables because the speed rushes decisions, or at crowded live tables where you get bored and start "making things happen."
• Rule drift: you've gradually been accepting worse games, such as H17 or 6:5, because they were the only seats open.

Tools for Logging: Spreadsheets vs Mobile Notes

The best logging tool is the one you'll actually use every session. That said, the options aren't equal.

Spreadsheets (Google Sheets or Excel)

A spreadsheet is the gold standard because it computes for you and supports filtering, sorting, and charts. A clean layout looks like this:

If hands are in column F, unit in E, and house edge (as a decimal) in D, theoretical loss is simply =F2*E2*D2. Net units is =(H2-G2)/E2, where G is buy-in and H is cash-out. Keep a running total row of cumulative actual and cumulative theoretical, and chart both lines over time. Watching the two lines diverge or converge is the single most educational visual in blackjack.

Mobile Notes

A notes app is fine as a capture tool. Type the raw numbers on your phone the moment you leave the table, then transfer them to the spreadsheet within 24 hours. Don't treat notes as your permanent record; unstructured text can't be filtered or summed, and entries tend to get skipped.

Privacy and Security Practices

Your log is a detailed financial record, so treat it like one:

• Never store account passwords, player card numbers, or payment details in the log.
• Protect the file with a strong password and enable two-factor authentication on the cloud account that hosts it.
• Don't share screenshots publicly with platform usernames or identifiable session times visible.
• If you play in brick-and-mortar casinos, log after you leave the floor, not at the table. Note-taking at the table draws attention you don't want.
• Retain records long-term. Depending on your jurisdiction, gambling wins and losses may have tax implications, so check local requirements with a qualified professional.

Making the Habit Stick

The log only works if it's unbroken. Build it into your routine the same way you build in your buy-in and stop-loss: no session ends until the entry is complete. Keep it short enough to finish in two minutes. Review weekly for discipline trends and monthly for actual vs theoretical. Resist the urge to edit old entries; the value of the record is that it's honest at the moment it was made.

After a few months, something shifts. You stop arguing with yourself about whether you're "really up." You know your true hourly cost, your fatigue limit, and the three hands you keep getting wrong. That clarity is worth more than any betting system ever sold.