{"id":952,"date":"2025-07-02T18:58:11","date_gmt":"2025-07-02T17:58:11","guid":{"rendered":"https:\/\/hhdealer.com\/blog\/?p=952"},"modified":"2025-09-14T03:31:05","modified_gmt":"2025-09-14T02:31:05","slug":"the-law-of-large-numbers-common-pitfalls-misconceptions-and-implications-to-online-poker","status":"publish","type":"post","link":"https:\/\/hhdealer.com\/blog\/the-law-of-large-numbers-common-pitfalls-misconceptions-and-implications-to-online-poker\/","title":{"rendered":"The Law of Large Numbers: Common Pitfalls, Misconceptions and Implications to Online Poker"},"content":{"rendered":"\n

The Law of Large Numbers (LLN<\/strong>) is one of the most important results in probability theory. It explains why, as the number of trials in a random process grows, the results tend to \u201caverage out\u201d and approach the expected value.
While widely applied in fields such as statistics, finance, and gambling, the LLN is often misunderstood. <\/p>\n\n\n\n

Many people incorrectly believe that it predicts short-term balance or that outcomes will \u201ccorrect themselves\u201d quickly.
This article explores the law in detail, explains how it works with intuitive examples, and highlights common pitfalls that lead to false conclusions and poor decisions.<\/p>\n\n\n\n

\n\n\n\n

1. Introduction<\/h2>\n\n\n\n

Online poker provides a unique laboratory for probability, with millions of hands played daily across the globe. The game blends skill<\/strong> and chance<\/strong>, creating short-term uncertainty but long-term patterns that can be studied with statistical tools.<\/p>\n\n\n\n

Among these tools, the Law of Large Numbers<\/strong> stands out. It explains why a skilled player\u2019s edge becomes visible only after many thousands\u2014sometimes millions\u2014of hands. While a beginner might expect immediate results to reflect their true ability, the LLN shows that variance dominates in the short run, and only large samples reveal a player\u2019s expected win rate.<\/p>\n\n\n\n

Yet many online players misinterpret this law, falling victim to myths such as \u201cbad luck will even out soon\u201d or \u201cI\u2019m due for a winning streak.\u201d These misconceptions fuel frustration, bankroll mismanagement, and ultimately, poor decision-making.<\/p>\n\n\n\n

\n\n\n\n

2. What Is the Law of Large Numbers?<\/h2>\n\n\n\n

2.1 The Basic Idea<\/h4>\n\n\n\n

In poker, every decision has an expected value (EV)<\/strong>. For example, if you call an all-in with pocket aces against pocket kings, you expect to win about 82% of the time.<\/p>\n\n\n\n

    \n
  • In 1 hand<\/strong>, you may lose.<\/li>\n\n\n\n
  • In 10 hands<\/strong>, you might win 7 or 8 times.<\/li>\n\n\n\n
  • In 10,000 hands<\/strong>, you\u2019ll be very close to the true 82% win rate.<\/li>\n<\/ul>\n\n\n\n

    The LLN guarantees that the more trials (hands, calls, bets) you accumulate, the closer your actual results will be to your expected results.<\/p>\n\n\n\n

    2.2 Formal Definition<\/h4>\n\n\n\n

    If a random variable \\(X\\) has an expected value \\(\\mu\\), and we conduct \\(n\\) independent trials producing outcomes \\(X_1, X_2, …, X_n\\), then the sample average <\/p>\n\n\n\n

    \n

    $$\\bar{X}_n = \\frac{1}{n} \\sum_{i=1}^{n} X_i$$<\/p>\n<\/blockquote>\n\n\n\n

    converges to \\(\\mu\\) as \\(n \\to \\infty\\).<\/p>\n\n\n\n

    2.3 Applied to Poker<\/h4>\n\n\n\n
    \n

    $$\\text{Long-term win rate} \\approx \\text{Expected Value (EV)} \\newline \\quad \\text{as hands played} \\to \\infty$$<\/p>\n<\/blockquote>\n\n\n\n

    \n\n\n\n

    3. Applications of the LLN in Online Poker<\/h2>\n\n\n\n
      \n
    • Win Rate Measurement:<\/strong> A player\u2019s true skill level (e.g., +5 big blinds per 100 hands) is only observable after large samples. Short-term results may not reflect ability.<\/li>\n\n\n\n
    • Variance Management:<\/strong> Downswings and heaters are normal in the short run; LLN ensures they smooth out over the long run.<\/li>\n\n\n\n
    • Bankroll Planning:<\/strong> Understanding variance and LLN helps players set realistic bankroll requirements for surviving short-term swings.<\/li>\n\n\n\n
    • HUDs, Data Mining and Hand Histories:<\/strong> Tools that aggregate hand histories rely on LLN to provide reliable statistics about opponents. A player\u2019s VPIP (Voluntarily Put $ in Pot) becomes meaningful only after thousands of observed hands.<\/li>\n<\/ul>\n\n\n\n
      \n\n\n\n

      4. Why It Matters<\/h2>\n\n\n\n

      The LLN is not just a mathematical curiosity. Apart from Poker, it underpins many real-world systems:<\/p>\n\n\n\n

        \n
      • Casinos:<\/strong> Even though players win in the short term, casinos win in the long run because the house edge always asserts itself over many games.<\/li>\n\n\n\n
      • Insurance:<\/strong> Companies collect premiums from thousands of customers. While some will file large claims, the average cost stabilizes, allowing companies to set fair and profitable rates.<\/li>\n\n\n\n
      • Finance:<\/strong> Over decades, stock markets show average growth rates that are much more predictable than short-term fluctuations.<\/li>\n\n\n\n
      • Science and Medicine:<\/strong> Large sample sizes make experimental results more reliable because averages become less sensitive to random noise.<\/li>\n<\/ul>\n\n\n\n
        \n\n\n\n

        5. The Most Common Misinterpretation<\/h2>\n\n\n\n

        5.1 The Coin Toss<\/h3>\n\n\n
        \n
        \"\"<\/figure><\/div>\n\n\n

        Consider the following 2-Player coin toss game:<\/p>\n\n\n\n

          \n
        • Heads: You win $100<\/li>\n\n\n\n
        • Tails: I win $100<\/li>\n<\/ul>\n\n\n\n

          The most common misinterpretation is that long term none of the player will win any money, since the probability of Heads and Tails are 50% and the payout is the same for both players.<\/p>\n\n\n\n

          This is a classic point where intuition about percentages<\/em> (relative frequencies) and absolute amounts<\/em> (money won or lost) can pull in different directions. I\u2019ll explain why the Law of Large Numbers (LLN) guarantees the proportion<\/strong> of heads \u2192 50% while the absolute<\/strong> dollar lead can and typically does grow over time \u2014 and how fast it grows.<\/p>\n\n\n\n

          5.2 What LLN predicts (percentages)<\/h3>\n\n\n\n

          If you toss a fair coin nnn times and let \\(H\\) be the number of heads, the LLN says $$\\frac{H}{n} \\longrightarrow 0.5 \\quad\\text{as } n \\to \\infty$$<\/p>\n\n\n\n

          In words: the fraction<\/em> of heads converges to 50%. So the probability<\/strong> of heads on any single flip stays 50%, and the long-run proportion<\/strong> of heads and tails becomes essentially equal.<\/p>\n\n\n\n

          Example<\/strong>: after 10,000 flips you\u2019ll typically see something very close to 50% heads (e.g. 49.5\u201350.5%), and after a million flips you\u2019ll be even closer.<\/p>\n\n\n\n

          \n\n\n\n

          5.3 What happens to absolute dollars<\/h3>\n\n\n\n

          Each flip in your game moves money by $100 from one player to the other. If heads occurs the first player gets +$100 (and the other \u2212$100); if tails the opposite happens. The net<\/strong> money for the \u201cheads\u201d player after \\(n\\) flips is $$\\text{Net} = 100\\cdot(H – (n-H)) = 200\\cdot\\big(H – \\frac{n}{2}\\big)$$<\/p>\n\n\n\n

          So the net depends on how far \\(H\\) is from exactly \\(n\/2\\). Even though \\(H\/n\\) gets closer to 0.5, the difference<\/em> \\(H – n\/2\\) typically scales like \\(\\sqrt{n}\\)\u200b (this is standard from the Central Limit Theorem \/ binomial variability).<\/p>\n\n\n\n

          More concretely:<\/p>\n\n\n\n

            \n
          • The typical size<\/strong> of the deviation \\(H – n\/2\\) is about \\(frac{1}{2}\\sqrt{n}\\)\u200b.<\/li>\n\n\n\n
          • Therefore the typical net money<\/strong> is about $$\\text{Typical net} \\approx 200\\cdot\\frac{\\sqrt{n}}{2} = 100\\sqrt{n}$$<\/li>\n<\/ul>\n\n\n\n

            So the dollar lead grows without bound as \\(n\\) increases, but only like \\(\\sqrt{n}\\)\u200b \u2014 much slower than the linear growth of total money exchanged (which is \\(100n\\) in turnover).<\/p>\n\n\n\n

            \n\n\n\n

            5.4 Numerical examples<\/h3>\n\n\n\n
              \n
            • \\(n=100\\) flips: typical net \u2248 \\(100\\sqrt{100}=100\\cdot 10 = \\$1{,}000\\).
              Proportion error \u2248 \\(1\/(2\\sqrt{100})=5\\%\\).<\/li>\n\n\n\n
            • \\(n=10{,}000\\) typical net \u2248 \\(100\\cdot100=\\$10{,}000\\).
              Proportion error \u2248 \\(1\/(2\\sqrt{10{,}000})=0.5\\%\\).<\/li>\n\n\n\n
            • \\(n=1{,}000{,}000\\): typical net \u2248 \\(100\\cdot1000=\\$100{,}000\\).
              Proportion error \u2248 \\(1\/(2\\sqrt{1{,}000{,}000})=0.05\\%\\).<\/li>\n<\/ul>\n\n\n\n
              \n

              You see the pattern: the proportion error shrinks (from 5% to 0.5% to 0.05%), but the absolute dollar amount<\/em> a typical leader is ahead by grows (from $1k to $10k to $100k).<\/p>\n<\/blockquote>\n\n\n\n

              Another Example:<\/h4>\n\n\n\n
              Number of Flips (n)<\/th>Dollar Lead<\/th>Heads<\/th>Tails<\/th>Proportion (Heads)<\/th>EV-Deviation<\/th><\/tr><\/thead>
              10<\/td>$200<\/strong><\/td>6<\/td>4<\/td>60%<\/td>20%<\/td><\/tr>
              100<\/td>$1,600<\/strong><\/td>58<\/td>42<\/td>58%<\/td>16%<\/td><\/tr>
              1,000<\/td>$6,000<\/strong><\/td>530<\/td>470<\/td>53%<\/td>6%<\/td><\/tr>
              10,000<\/td>$22,000<\/strong><\/td>5,110<\/td>4,890<\/td>51%<\/td>2.2%<\/td><\/tr>
              100,000<\/td>$84,000<\/strong><\/td>50,420<\/td>49,580<\/td>50,42%<\/td>0.0084%<\/strong><\/td><\/tr><\/tbody><\/table>
              Example: <\/strong>The proportion of Heads\/Tails grows closer to 50% as the number of flips grows, but the absolute dollar difference<\/strong> grows larger!<\/figcaption><\/figure>\n\n\n\n
              \n\n\n\n

              5.5 How to reconcile the two statements<\/h3>\n\n\n\n
                \n
              • LLN (proportions)<\/strong>: \\(H\/n \\to 0.5\\). The relative<\/em> difference between heads and tails goes to zero like \\(1\/\\sqrt{n}\\)\u200b.<\/li>\n\n\n\n
              • Absolute dollars<\/strong>: The absolute<\/em> difference in wins (dollars) behaves like \\(\\sqrt{n}\\)\u200b, so it increases without bound even while the percentage<\/em> gap shrinks<\/strong>.<\/li>\n<\/ul>\n\n\n\n
                \n

                So both statements are true simultaneously \u2014 they just measure different things.<\/strong><\/p>\n<\/blockquote>\n\n\n\n

                \n\n\n\n

                5.6 Recap<\/h3>\n\n\n\n
                  \n
                • The expected<\/strong> net for either player after \\(n\\) flips is 0 (the game is fair).<\/li>\n\n\n\n
                • But the standard deviation<\/strong> of the net grows like \\(100\\sqrt{n}\\)\u200b. That means wide dollar swings are not only possible but typical as \\(n\\) increases.<\/li>\n\n\n\n
                • At any finite \\(n\\), one player is more likely to be ahead than the other (roughly 50\/50), and the typical lead size increases with \\(n\\).<\/li>\n<\/ul>\n\n\n\n
                  \n\n\n\n

                  5.7) Practical implications (poker \/ bankroll thinking)<\/h3>\n\n\n\n
                    \n
                  • In fair games or edge-based games, proportion convergence<\/strong> (LLN) is the reason skilled play wins in the long run \u2014 but you need large sample sizes for that to manifest reliably.<\/li>\n\n\n\n
                  • Absolute swings<\/strong> (variance) matter for bankroll: even if your expected value per hand is positive, your bankroll must withstand the \\(\\sqrt{n}\\)\u200b-scale swings that occur before the long-run average shows up.<\/li>\n\n\n\n
                  • Don\u2019t confuse \u201cgetting closer to the long-run percentage\u201d with \u201cyou won\u2019t lose large amounts on the way\u201d \u2014 losses can and typically will grow in absolute terms as you play more, even if they are small relative to total action.<\/li>\n<\/ul>\n\n\n\n

                    5.8) One last intuitive wording<\/h3>\n\n\n\n