ꯖꯨꯌꯥ ꯁꯥꯟꯅꯔꯤꯕꯒꯤ ꯑꯔꯥꯟꯕ ꯋꯥꯈꯜ
ꯃꯍꯩ ꯑꯃꯥ ꯃꯇꯝ ꯈꯔꯥ ꯊꯣꯛꯇꯔꯕꯅ “ꯍꯧꯖꯤꯛ ꯊꯣꯛꯀꯗꯕꯅꯤ” ꯍꯥꯌꯅ ꯊꯥꯖꯕ꯫

ꯑꯉꯥꯡꯕ ꯇꯔꯥꯒꯤ ꯃꯇꯨꯡꯗꯁꯨ ꯑꯃꯨꯕꯒꯤ ꯊꯣꯛꯄꯒꯤ ꯈꯨꯗꯣꯡꯆꯥꯕ 18/37 ꯅꯤ꯫ ꯆꯛꯔꯅ ꯀꯔꯤꯁꯨ ꯅꯤꯡꯁꯤꯡꯗꯦ ꯑꯃꯁꯨꯡ ꯀꯅꯥꯒꯨꯝꯕ ꯑꯃꯒꯤꯁꯨ ꯁꯦꯜ ꯄꯤꯒꯗꯕ ꯂꯩꯇꯦ꯫
ꯑꯔꯥꯟꯕ ꯋꯥꯈꯜ
If redꯕꯦꯠꯑꯉꯥꯡꯕꯑꯉꯥꯡꯕ ꯃꯁꯤꯡ 18 ꯗ ꯊꯝꯕ ꯆꯞ ꯃꯥꯟꯅꯕ ꯕꯦꯠ꯫ ꯐ꯭ꯔꯦꯟꯆ: ꯔꯨꯖ꯫ꯃꯄꯨꯡ ꯄꯥꯕꯤꯌꯨ → has come up ten times, many players feel that blackꯕꯦꯠꯑꯃꯨꯕꯑꯃꯨꯕ ꯃꯁꯤꯡ 18 ꯗ ꯊꯝꯕ ꯆꯞ ꯃꯥꯟꯅꯕ ꯕꯦꯠ꯫ ꯐ꯭ꯔꯦꯟꯆ: ꯅꯣꯌꯥꯔ꯫ꯃꯄꯨꯡ ꯄꯥꯕꯤꯌꯨ → is “due” and bet accordingly. Others feel red is “hot” and bet that. Both are reasoning from a sequence that is already over to a spin that has not happened. In the fair single-zeroꯆꯛꯔꯖꯤꯔꯣꯆꯞ ꯃꯥꯟꯅꯕ, ꯗꯖꯟ ꯅꯠꯔꯒ ꯀꯂꯝ ꯕꯦꯠ ꯑꯃꯠꯇꯅ ꯀꯨꯞꯇꯕ ꯑꯁꯪꯕ ꯄꯣꯀꯦꯠ ꯑꯃꯈꯛꯇ; ꯍꯥꯎꯁ ꯑꯦꯖꯀꯤ ꯍꯧꯔꯀꯐꯝꯅꯤ꯫ ꯑꯇꯣꯞꯄ ꯃꯃꯤꯡ: ꯁꯤꯡꯒꯜ ꯖꯤꯔꯣ, 0꯫ꯃꯄꯨꯡ ꯄꯥꯕꯤꯌꯨ → model with independent spinsꯃꯦꯊꯃꯦꯇꯤꯛꯁꯃꯔꯤ ꯂꯩꯅꯗꯕ ꯁ꯭ꯄꯤꯟꯁꯤꯡꯍꯥꯟꯅꯒꯤ ꯃꯍꯩ ꯀꯔꯤꯒꯨꯝꯕ ꯑꯣꯏꯔꯕꯁꯨ ꯁ꯭ꯄꯤꯟ ꯈꯨꯗꯤꯡꯃꯛꯀꯤ ꯊꯣꯛꯄꯒꯤ ꯈꯨꯗꯣꯡꯆꯥꯕ ꯃꯥꯟꯅꯩ꯫ ꯋꯦꯕꯁꯥꯏꯠ ꯑꯁꯤꯒꯤ ꯊꯤꯕ ꯈꯨꯗꯤꯡꯃꯛꯀꯤ ꯌꯨꯝꯐꯝꯅꯤ꯫ꯃꯄꯨꯡ ꯄꯥꯕꯤꯌꯨ →, the next-spin probabilityꯃꯦꯊꯃꯦꯇꯤꯛꯁꯊꯣꯛꯄꯒꯤ ꯈꯨꯗꯣꯡꯆꯥꯕꯃꯍꯩ ꯑꯃꯥ ꯊꯣꯛꯄꯒꯤ ꯈꯨꯗꯣꯡꯆꯥꯕ, 0 ꯗꯒꯤ 1 ꯐꯥꯑꯣꯕ ꯅꯠꯔꯒ 0% ꯗꯒꯤ 100% ꯐꯥꯑꯣꯕ꯫ ꯃꯁꯤꯡ ꯑꯃꯥ: 1/37 = 2.70%꯫ꯃꯄꯨꯡ ꯄꯥꯕꯤꯌꯨ → of black remains 18/37.
ꯀꯔꯤꯒꯤ ꯑꯆꯨꯝꯕꯒꯨꯝ ꯐꯥꯑꯣꯕꯒꯦ
Under the independent fair-wheel model, the proportions of red and black each tend toward 18/37 as the sample grows. This does not require a compensating run of black. After ten initial reds and N further spins, expected red and black counts differ by ten, while the contribution of those initial reds to the total proportion becomes smaller. Actual counts still fluctuate.
ꯃꯣꯟꯇꯦ ꯀꯥꯔꯂꯣ, 1913 ꯒꯤ ꯑꯒꯁ꯭ꯠ 18
ꯀꯥꯁꯤꯅꯣ ꯗꯦ ꯃꯣꯟꯇꯦ-ꯀꯥꯔꯂꯣꯒꯤ ꯇꯦꯕꯜ ꯑꯃꯗ ꯃꯈꯥ ꯆꯠꯅꯅ ꯑꯃꯨꯕ ꯑꯃꯨꯛ 26 ꯊꯣꯀꯈꯤ꯫ ꯑꯃꯨꯕ ꯇꯔꯥꯃꯉꯥꯁꯨꯕ ꯃꯔꯛꯇꯒꯤ ꯁꯥꯟꯅꯔꯤꯕꯁꯤꯡꯅ ꯑꯉꯥꯡꯕꯗ ꯌꯥꯝꯅꯥ ꯊꯝꯈꯤ, ꯑꯅꯤ ꯊꯥꯛ ꯑꯃꯁꯨꯡ ꯑꯃꯨꯛ ꯍꯟꯅꯥ ꯑꯅꯤ ꯊꯥꯛ ꯍꯦꯅꯒꯠꯍꯟꯗꯨꯅ, ꯁ꯭ꯄꯤꯟ ꯈꯨꯗꯤꯡꯃꯛ ꯃꯃꯥꯡꯒꯤꯗꯒꯤ “ꯍꯦꯟꯅꯥ ꯁꯣꯏꯗꯅ” ꯐꯥꯑꯣꯈꯤ꯫ ꯀꯦꯁꯤꯅꯣꯅ ꯐ꯭ꯔꯥꯉ꯭ꯛ ꯂꯤꯁꯤꯡ ꯀꯌꯥ ꯃꯥꯏꯄꯥꯀꯈꯤ꯫ ꯑꯍꯥꯅꯕ ꯑꯃꯨꯕꯗꯒꯤ ꯁ꯭ꯇ꯭ꯔꯤꯛ ꯑꯗꯨꯒꯤ ꯈꯨꯗꯣꯡꯆꯥꯕ 137 ꯃꯤꯂꯤꯌꯟꯗꯒꯤ 1 ꯅꯤ, ꯑꯃꯁꯨꯡ ꯃꯥꯂꯦꯃꯒꯤ ꯀꯦꯁꯤꯅꯣꯁꯤꯡꯒꯤ ꯀꯗꯥꯏꯗꯥ ꯑꯃꯗ ꯃꯁꯤ ꯑꯔꯣꯏꯕꯗ ꯊꯣꯛꯀꯗꯕꯒꯤ ꯈꯨꯗꯣꯡꯆꯥꯕꯗꯤ ꯌꯥꯝꯅꯥ ꯅꯛꯅꯥ ꯑꯃꯅꯤ꯫ ꯃꯔꯝ ꯑꯁꯤꯅ ꯖꯨꯌꯥ ꯁꯥꯟꯅꯔꯤꯕꯒꯤ ꯑꯔꯥꯟꯕ ꯋꯥꯈꯜ ꯑꯁꯤꯕꯨ ꯃꯣꯟꯇꯦ ꯀꯥꯔꯂꯣꯒꯤ ꯑꯔꯥꯟꯕ ꯋꯥꯈꯜ ꯍꯥꯌꯅꯁꯨ ꯀꯧꯏ꯫
ꯑꯉꯀꯄ ꯅꯠꯇꯦ꯫ ꯃꯇꯝ 44 ꯗ ꯑꯃꯨꯛ ꯊꯣꯀꯏ꯫
ꯁꯥꯟꯅꯔꯤꯕꯁꯤꯡꯅ ꯈꯉꯂꯛꯂꯤ꯫ ꯑꯉꯥꯡꯕ “ꯍꯧꯖꯤꯛ ꯊꯣꯛꯀꯗꯕꯅꯤ”꯫
ꯇꯦꯕꯜ ꯑꯗꯨ ꯃꯤꯅ ꯊꯟꯊꯔꯦ꯫ ꯃꯥꯏꯊꯤꯕ ꯈꯨꯗꯤꯡꯒꯤ ꯃꯇꯨꯡꯗ ꯑꯉꯥꯡꯕꯗ ꯊꯝꯂꯕ ꯁꯦꯜ ꯑꯅꯤ ꯊꯥꯛ ꯍꯦꯅꯒꯠꯍꯜꯂꯤ꯫
ꯂꯟ-ꯊꯨꯝ ꯌꯥꯝꯅꯥ ꯃꯥꯡꯈꯤ꯫ ꯑꯅꯧꯕ ꯑꯃꯨꯕ ꯈꯨꯗꯤꯡꯅ ꯑꯉꯥꯡꯕ ꯍꯦꯟꯅꯥ ꯁꯣꯏꯗꯅ ꯐꯥꯑꯣꯍꯜꯂꯤ꯫
ꯁ꯭ꯇ꯭ꯔꯤꯛ ꯑꯗꯨ ꯂꯣꯏꯔꯦ꯫ ꯃꯁꯥꯅ ꯁ꯭ꯇ꯭ꯔꯤꯛ ꯑꯃꯅꯤ ꯍꯥꯌꯕ “ꯈꯉꯕ” ꯉꯃꯗꯕ ꯁ꯭ꯇ꯭ꯔꯤꯛ ꯑꯃꯒꯤ ꯃꯥꯌꯣꯛꯇ ꯕꯦꯠ ꯊꯝꯂꯤꯕ ꯃꯤꯁꯤꯡꯗꯒꯤ ꯀꯦꯁꯤꯅꯣꯅ ꯐ꯭ꯔꯥꯉ꯭ꯛ ꯂꯤꯁꯤꯡ ꯀꯌꯥ ꯃꯥꯏꯄꯥꯀꯈ꯭ꯔꯦ꯫
ꯍꯣꯠ ꯍꯦꯟꯗ
ꯃꯆꯨ ꯑꯃꯥ ꯁ꯭ꯇ꯭ꯔꯤꯛꯇ ꯂꯩꯔꯤꯕꯅ ꯃꯗꯨꯗ ꯕꯦꯠ ꯊꯝꯕ ꯑꯁꯤ ꯑꯔꯥꯟꯕ ꯑꯗꯨꯃꯛ ꯑꯣꯟꯊꯣꯛꯂꯕꯅꯤ: ꯁ꯭ꯇ꯭ꯔꯤꯛ ꯑꯗꯨ ꯂꯣꯏꯒꯅꯤ ꯍꯥꯌꯅ ꯅꯠꯇꯅ ꯃꯈꯥ ꯆꯠꯊꯒꯅꯤ ꯍꯥꯌꯅ ꯊꯥꯖꯩ꯫ ꯁꯥꯏꯀꯣꯂꯣꯖꯤꯁ꯭ꯇꯁꯤꯡꯅ ꯃꯁꯤꯕꯨ ꯍꯣꯠ ꯍꯦꯟꯗꯑꯔꯥꯟꯕ ꯊꯥꯖꯕꯍꯣꯠ ꯍꯦꯟꯗꯃꯥꯏꯄꯥꯛꯄꯒꯤ ꯁ꯭ꯇ꯭ꯔꯤꯛ, ꯅꯠꯔꯒ ꯃꯍꯩ ꯑꯃꯒꯤ ꯁ꯭ꯇ꯭ꯔꯤꯛ ꯑꯃꯅ ꯑꯇꯣꯞꯄ ꯑꯃꯥ ꯊꯣꯛꯅꯕ ꯍꯦꯟꯅꯥ ꯌꯥꯍꯜꯂꯤ ꯍꯥꯌꯅ ꯊꯥꯖꯕ꯫ ꯆꯛꯔꯗ ꯃꯁꯤ ꯖꯨꯌꯥ ꯁꯥꯟꯅꯔꯤꯕꯒꯤ ꯑꯔꯥꯟꯕ ꯋꯥꯈꯜ ꯑꯗꯨ ꯑꯣꯟꯊꯣꯛꯂꯕꯅꯤ꯫ꯃꯄꯨꯡ ꯄꯥꯕꯤꯌꯨ → ꯍꯥꯌꯅ ꯀꯧꯏ, ꯒꯤꯂꯣꯚꯤꯆ, ꯚꯦꯂꯣꯟ ꯑꯃꯁꯨꯡ ꯇꯚꯔꯁ꯭ꯀꯤꯅ 1985 ꯗ ꯕꯥꯁ꯭ꯀꯦꯠꯕꯣꯂꯗ ꯕꯣꯜ ꯊꯥꯗꯣꯀꯄꯗ ꯁ꯭ꯇ꯭ꯔꯤꯛ ꯊꯤꯈꯤꯕ ꯑꯃꯁꯨꯡ ꯂꯥꯛꯀꯤ ꯃꯊꯛꯇ ꯑꯃꯠꯇ ꯐꯪꯗꯈꯤꯕ ꯊꯤꯖꯤꯟꯕ ꯑꯃꯒꯤ ꯃꯇꯨꯡ ꯏꯟꯅꯥ꯫ 2018 ꯗ ꯃꯤꯂꯔ ꯑꯃꯁꯨꯡ ꯁꯥꯟꯖꯨꯔꯖꯣꯅ ꯇꯦꯁ꯭ꯠ ꯑꯗꯨ ꯃꯁꯥꯃꯛ ꯌꯦꯟꯊꯔꯝꯃꯤ ꯍꯥꯌꯅ ꯎꯠꯈꯤ: ꯃꯇꯝ ꯀꯨꯞꯄ ꯔꯦꯀꯣꯔꯗ ꯑꯃꯗ ꯊꯣꯀꯄꯥ ꯃꯇꯨꯡꯗ ꯍꯧꯖꯤꯛꯇꯒꯤ ꯊꯣꯀꯄꯒꯤ ꯆꯥꯡ ꯑꯗꯨ ꯑꯦꯚꯔꯦꯖꯗ ꯊꯥꯗꯣꯀꯄ ꯃꯤ ꯑꯗꯨꯒꯤ ꯄꯨꯝꯅꯃꯛꯀꯤ ꯆꯥꯡꯗꯒꯤ ꯅꯦꯝꯅ ꯂꯥꯀꯏ, ꯃꯔꯝ ꯑꯗꯨꯅ ꯍꯣꯠ ꯍꯦꯟꯗ ꯑꯃꯠꯇ ꯂꯩꯇꯕ ꯁꯥꯟꯅꯔꯤꯕ ꯑꯃꯥ ꯈꯔꯥ ꯏꯪꯕꯒꯨꯝ ꯎꯏ꯫ ꯆꯨꯃꯊꯣꯛꯂꯕꯗ ꯗꯦꯇꯥ ꯑꯗꯨꯃꯛꯅ ꯑꯄꯤꯀꯄ ꯍꯣꯠ ꯍꯦꯟꯗ ꯑꯃꯥ ꯎꯠꯂꯤ꯫ ꯃꯤꯑꯣꯏ ꯑꯃꯥ ꯁꯥꯠꯂꯛꯄ ꯌꯥꯏ; ꯆꯛꯔ ꯑꯃꯒꯤ ꯂꯩꯒꯗꯕ ꯐꯤꯚꯝ ꯑꯃꯠꯇ ꯂꯩꯇꯦ꯫ ꯁꯥꯟꯅꯔꯤꯕ ꯈꯔꯅ ꯊꯥꯖꯕ ꯑꯅꯤꯃꯛ ꯃꯇꯝ ꯑꯃꯗ ꯊꯝꯃꯤ, ꯁ꯭ꯇ꯭ꯔꯤꯛ ꯑꯃꯥ “ꯌꯥꯝꯅꯥ ꯁꯥꯡꯂꯛꯠꯔꯤꯉꯩ” ꯐꯥꯑꯣꯕ ꯃꯈꯥ ꯆꯠꯊꯒꯅꯤ ꯑꯃꯁꯨꯡ ꯃꯗꯨꯒꯤ ꯃꯇꯨꯡꯗ ꯂꯣꯏꯒꯅꯤ꯫ ꯆꯨꯝꯃꯤꯟꯅꯕ ꯆꯛꯔ ꯑꯃꯗ ꯃꯈꯜ ꯈꯨꯗꯤꯡꯃꯛꯀꯤ ꯐꯪꯒꯗꯕ ꯃꯃꯜꯃꯦꯊꯃꯦꯇꯤꯛꯁꯐꯪꯒꯗꯕ ꯃꯃꯜꯃꯇꯝ ꯁꯥꯡꯅ ꯌꯦꯡꯂꯕꯗ ꯕꯦꯠ ꯑꯃꯒꯤ ꯑꯦꯚꯔꯦꯖ ꯃꯍꯩ꯫ ꯃꯔꯨꯑꯣꯏꯕ ꯌꯨꯔꯣꯄꯤꯌꯟ ꯕꯦꯠ ꯈꯨꯗꯤꯡꯗ ꯊꯝꯂꯕ ꯁꯦꯜꯒꯤ −1/37꯫ꯃꯄꯨꯡ ꯄꯥꯕꯤꯌꯨ → ꯀꯣꯏꯟ ꯊꯥꯗꯨꯅ ꯃꯆꯨ ꯈꯅꯕꯒꯥ ꯃꯥꯟꯅꯩ: ꯊꯝꯂꯕ ꯁꯦꯜꯒꯤ −2.70%꯫
ꯑꯁꯦꯡꯕ ꯇꯦꯕꯜ ꯑꯃꯗ
Croson and Sundali (2005) watched 18 hours of security video from one double-zero table at a large casino in Reno: 139 players placing 24,131 bets. On the even-money betsꯕꯦꯠꯆꯞ ꯃꯥꯟꯅꯕ ꯕꯦꯠꯑꯉꯥꯡꯕ/ꯑꯃꯨꯕ, ꯕꯤꯖꯣꯔ/ꯖꯣꯔ, ꯑꯋꯥꯡꯕ/ꯑꯅꯦꯝꯕ꯫ ꯃꯁꯤꯡ 18, 1:1 ꯄꯦꯑꯥꯎꯠ꯫ꯃꯄꯨꯡ ꯄꯥꯕꯤꯌꯨ →, players were about evenly divided between following short runs and opposing them. After five of a kind in a row they leaned against the run, and after six or more, 85% of their bets were on the other side. The same players showed the hot hand about themselves: after a win on the numbers they covered 13.6 numbers on the next spin on average, against 9.2 after a loss. The study found no statistically significant evidence of wheel bias. These are observations from one table, and the authors discuss alternative explanations for changes in betting after wins.
ꯀꯦꯁꯤꯅꯣꯒꯤ ꯃꯄꯥꯟꯗ
ꯑꯁꯦꯡꯕ ꯋꯥꯔꯦꯞ ꯂꯧꯕ ꯃꯐꯃꯁꯤꯡꯗꯁꯨ ꯂꯧꯁꯤꯡ ꯑꯁꯤꯃꯛ ꯎꯏ꯫ ꯀ꯭ꯂꯣꯠꯐꯦꯜꯇꯔ ꯑꯃꯁꯨꯡ ꯀꯨꯛ (1993) ꯅ ꯃꯦꯔꯤꯂꯦꯟꯗꯀꯤ ꯅꯨꯃꯤꯠ ꯈꯨꯗꯤꯡꯒꯤ ꯃꯁꯤꯡ ꯒꯦꯝ ꯌꯦꯡꯈꯤ: ꯃꯁꯤꯡ ꯑꯃꯥ ꯊꯣꯀꯄꯒꯥ ꯂꯣꯌꯅꯅ ꯃꯗꯨꯗ ꯊꯝꯕ ꯁꯦꯜ ꯌꯥꯝꯅꯥ ꯍꯟꯊꯔꯀꯈꯤ ꯑꯃꯁꯨꯡ ꯍꯟꯅꯥ ꯐꯪꯅꯕ ꯊꯥ ꯀꯌꯥ ꯆꯪꯈꯤ, ꯗ꯭ꯔꯣ ꯑꯗꯨꯒꯤ ꯅꯤꯡꯁꯤꯡꯕ ꯑꯃꯠꯇ ꯂꯩꯇꯕꯁꯨ꯫ ꯆꯦꯟ, ꯃꯣꯁ꯭ꯀꯣꯋꯤꯠꯖ ꯑꯃꯁꯨꯡ ꯁꯨ (2016) ꯅ ꯄ꯭ꯔꯣꯐꯦꯁꯅꯦꯜꯁꯤꯡꯒꯤ ꯃꯔꯛꯇꯁꯨ ꯃꯁꯤ ꯐꯪꯈꯤ꯫ ꯌꯨꯅꯥꯏꯇꯦꯗ ꯁ꯭ꯇꯦꯇꯁꯀꯤ ꯑꯦꯁꯥꯏꯂꯝ ꯖꯖꯁꯤꯡꯅ ꯃꯃꯥꯡꯒꯤ ꯀꯦꯁ ꯑꯗꯨ ꯌꯥꯔꯕꯗ ꯀꯦꯁ ꯑꯃꯥ ꯌꯥꯗꯕꯒꯤ ꯈꯨꯗꯣꯡꯆꯥꯕ ꯄꯔꯁꯦꯟꯇꯦꯖ ꯄꯣꯏꯟꯠ 3.3 ꯐꯥꯑꯣꯕ ꯍꯦꯟꯈꯤ꯫ ꯕꯦꯁꯕꯣꯜ ꯑꯝꯄꯥꯌꯔꯁꯤꯡꯅ ꯁ꯭ꯇ꯭ꯔꯥꯏꯛ ꯑꯃꯥ ꯀꯧꯔꯕꯥ ꯃꯇꯨꯡꯗ ꯑꯃꯨꯛ ꯍꯟꯅꯥ ꯀꯧꯕꯒꯤ ꯈꯨꯗꯣꯡꯆꯥꯕ ꯄꯣꯏꯟꯠ 1.5 ꯍꯟꯊꯈꯤ, ꯑꯃꯁꯨꯡ ꯈꯟꯊꯕ ꯀꯠꯊꯣꯀꯄ ꯄꯤꯆꯁꯤꯡꯗ ꯃꯗꯨꯒꯤ ꯑꯅꯤ ꯊꯥꯛꯇꯒꯤ ꯍꯦꯟꯅꯥ ꯍꯟꯊꯈꯤ꯫ ꯐ꯭ꯂꯦꯠ ꯐꯤ ꯄꯤꯕ ꯂꯣꯟ ꯑꯣꯐꯤꯁꯔꯁꯤꯡꯒꯥ ꯇꯧꯈꯤꯕ ꯐꯤꯜꯗ ꯑꯦꯛꯁꯄꯦꯔꯤꯃꯦꯟꯠ ꯑꯃꯗ ꯋꯥꯔꯦꯄꯁꯤꯡꯒꯤ 9% ꯐꯥꯑꯣꯕ ꯃꯑꯣꯡ ꯑꯁꯤꯅ ꯑꯣꯟꯊꯣꯀꯈꯤ꯫ ꯀꯦꯁꯁꯤꯡ ꯔꯦꯟꯗꯝ ꯃꯊꯪ ꯃꯊꯪꯗ ꯂꯥꯀꯈꯤ, ꯃꯔꯝ ꯑꯗꯨꯅ ꯑꯔꯣꯏꯕ ꯋꯥꯔꯦꯞ ꯑꯗꯨꯅ ꯃꯊꯪꯒꯤ ꯑꯗꯨꯒꯤ ꯃꯇꯥꯡꯗ ꯀꯔꯤꯁꯨ ꯍꯥꯌꯗꯦ, ꯑꯔꯣꯏꯕ ꯁ꯭ꯄꯤꯟ ꯑꯗꯨꯅ ꯃꯊꯪꯒꯤ ꯁ꯭ꯄꯤꯟꯒꯤ ꯃꯇꯥꯡꯗ ꯀꯔꯤꯁꯨ ꯍꯥꯌꯗꯕꯒꯨꯝꯅ꯫

ꯑꯣꯟꯊꯣꯛꯂꯕ ꯖꯨꯌꯥ ꯁꯥꯟꯅꯔꯤꯕꯒꯤ ꯑꯔꯥꯟꯕ ꯋꯥꯈꯜ
ꯐꯤꯂꯣꯁꯣꯐꯔ ꯏꯌꯥꯟ ꯍꯦꯀꯤꯡ (Ian Hacking, 1987) ꯅ ꯑꯍꯨꯃꯁꯨꯕ ꯃꯈꯜ ꯑꯃꯒꯤ ꯃꯃꯤꯡ ꯊꯣꯅꯈꯤ꯫ ꯀꯅꯥꯒꯨꯝꯕ ꯑꯃꯥ ꯀꯣꯊꯥꯡ ꯑꯃꯗ ꯆꯉꯂꯛꯇꯨꯅ ꯗꯥꯏꯁ ꯑꯅꯤ ꯗꯕꯜ ꯁꯤꯛꯁ ꯊꯣꯀꯄ ꯎꯏ ꯑꯃꯁꯨꯡ ꯃꯈꯣꯌ ꯃꯃꯥꯡꯗ ꯑꯃꯨꯛ ꯀꯌꯥ ꯊꯥꯗꯣꯛꯂꯝꯂꯒꯅꯤ ꯍꯥꯌꯅ ꯋꯥꯔꯦꯞ ꯂꯧꯏ꯫ ꯊꯥꯗꯣꯀꯄ ꯈꯨꯗꯤꯡꯃꯛꯇ ꯗꯕꯜ ꯁꯤꯛꯁ ꯊꯣꯛꯄꯒꯤ ꯈꯨꯗꯣꯡꯆꯥꯕ ꯃꯥꯟꯅꯩ, ꯅꯨꯃꯤꯗꯥꯡꯒꯤ ꯑꯍꯥꯅꯕ ꯑꯣꯏꯔꯕꯁꯨ ꯂꯤꯁꯤꯡꯁꯨꯕ ꯑꯣꯏꯔꯕꯁꯨ, ꯃꯔꯝ ꯑꯗꯨꯅ ꯎꯈꯤꯕ ꯑꯗꯨꯅ ꯁꯥꯟꯅꯕ ꯑꯗꯨ ꯀꯌꯥ ꯁꯥꯡꯅ ꯆꯠꯂꯤꯕꯒꯦ ꯍꯥꯌꯕꯒꯤ ꯃꯇꯥꯡꯗ ꯀꯔꯤꯁꯨ ꯍꯥꯌꯗꯦ꯫ ꯔꯨꯂꯦꯠꯇ ꯃꯁꯤ ꯑꯉꯥꯡꯕ ꯅꯤꯄꯥꯟ ꯎꯠꯂꯤꯕ ꯕꯣꯔꯗ ꯑꯃꯗ ꯂꯥꯀꯄ ꯑꯃꯁꯨꯡ ꯃꯈꯣꯌ ꯊꯣꯛꯅꯕ ꯇꯦꯕꯜ ꯑꯗꯨ ꯄꯨꯡ ꯀꯌꯥ ꯀꯣꯏꯔꯝꯂꯒꯅꯤ ꯍꯥꯌꯅ ꯋꯥꯔꯦꯞ ꯂꯧꯕ ꯁꯥꯟꯅꯔꯤꯕ ꯑꯗꯨꯅꯤ꯫ ꯃꯍꯧꯁꯥꯒꯤ ꯑꯔꯥꯟꯕ ꯋꯥꯈꯂꯅ ꯍꯥꯟꯅꯒꯤ ꯑꯗꯨ ꯃꯊꯪꯒꯤ ꯁ꯭ꯄꯤꯟꯗ ꯄꯥꯔꯤ; ꯃꯁꯤꯅꯗꯤ ꯈꯔꯥ ꯈꯛꯇꯗ ꯎꯕ ꯐꯪꯕ ꯃꯤꯠꯌꯦꯡ ꯑꯃꯥ ꯀꯅꯥꯒꯨꯝꯕ ꯑꯃꯠꯇꯅ ꯎꯈꯤꯗꯕ ꯍꯥꯟꯅꯒꯤ ꯊꯧꯗꯣꯛ ꯑꯃꯗ ꯄꯥꯔꯤ꯫
ꯃꯁꯤ ꯊꯥꯖꯈꯤꯕ ꯁꯥꯏꯀꯣꯂꯣꯖꯤꯁ꯭ꯠ ꯑꯃꯥ
ꯑꯔꯥꯟꯕ ꯋꯥꯈꯜ ꯑꯁꯤꯒꯤ ꯃꯍꯩ ꯇꯃꯈꯤꯕ ꯉꯥꯛꯁꯦꯜꯂꯤꯕꯁꯤꯡ ꯂꯩꯈꯤ꯫ ꯖꯔꯃꯟ ꯁꯥꯏꯀꯣꯂꯣꯖꯤꯁ꯭ꯠ ꯀꯥꯔꯜ ꯃꯥꯔꯕꯦꯅ ꯑꯃꯃꯃꯗ ꯄꯣꯛꯄ 49,152 ꯂꯩꯕ ꯔꯦꯖꯤꯁ꯭ꯇꯔ ꯃꯔꯤ ꯌꯦꯡꯈꯤ, ꯃꯈꯥ ꯆꯠꯅꯅ ꯅꯨꯄꯥꯃꯆꯥ ꯅꯠꯔꯒ ꯅꯨꯄꯤꯃꯆꯥꯒꯤ ꯈ꯭ꯕꯥꯏꯗꯒꯤ ꯁꯥꯡꯕ ꯁ꯭ꯇ꯭ꯔꯤꯛ 17 ꯅꯤ ꯍꯥꯌꯅ ꯐꯪꯈꯤ, ꯑꯃꯁꯨꯡ ꯁꯥꯡꯕ ꯁ꯭ꯇ꯭ꯔꯤꯛꯁꯤꯡ ꯂꯥꯛꯅ ꯌꯥꯍꯜꯂꯤꯕꯗꯒꯤ ꯍꯦꯟꯅꯥ ꯈꯔꯥ ꯈꯛꯇꯗ ꯊꯣꯀꯏ, ꯃꯔꯝ ꯑꯗꯨꯅ ꯁ꯭ꯇ꯭ꯔꯤꯛ ꯑꯃꯅ ꯑꯇꯣꯞꯄ ꯃꯍꯩ ꯑꯗꯨ ꯍꯦꯟꯅꯥ ꯊꯣꯛꯄ ꯌꯥꯍꯜꯂꯤ ꯍꯥꯌꯅ ꯋꯥꯔꯦꯞ ꯂꯧꯈꯤ꯫ ꯂꯥꯛꯈꯛꯇꯅ ꯑꯇꯣꯞꯄ ꯍꯥꯌꯔꯤ: ꯄꯣꯛꯄ 196,608 ꯗ 17 ꯅꯠꯔꯒ ꯍꯦꯟꯕꯒꯤ ꯁ꯭ꯇ꯭ꯔꯤꯛ ꯑꯃꯥ ꯃꯇꯝ ꯃꯉꯥꯗꯒꯤ ꯃꯔꯤ ꯃꯔꯛꯇ ꯊꯣꯀꯏ꯫ ꯃꯥꯔꯕꯦꯅ ꯐꯪꯈꯤꯕ ꯑꯗꯨ ꯔꯦꯟꯗꯝ ꯃꯊꯪ ꯃꯊꯪ ꯑꯃꯒꯤ ꯃꯑꯣꯡ ꯑꯗꯨꯃꯛꯇꯅꯤ꯫ ꯑꯦꯞꯁ꯭ꯇꯥꯏꯟ (2013) ꯅ ꯋꯥꯐꯝ ꯑꯁꯤ ꯑꯃꯨꯛ ꯍꯟꯅꯥ ꯍꯥꯌꯔꯤ; “ꯔꯨꯂꯦꯠꯇ ꯁ꯭ꯇ꯭ꯔꯤꯛ ꯊꯣꯛꯄꯒꯤ ꯈꯨꯗꯣꯡꯆꯥꯕ” ꯂꯃꯥꯏꯅ ꯔꯨꯂꯦꯠꯀꯤ ꯃꯈꯜ ꯄꯤꯔꯤ꯫

ꯑꯔꯥꯟꯕ ꯋꯥꯈꯜ ꯑꯁꯤ ꯆꯠꯂꯣꯏꯗꯕ ꯃꯐꯝ
If the wheel were physically biased, past spins would be informative, because they would be evidence about the pocketsꯈꯨꯠꯂꯥꯏꯄꯣꯀꯦꯠꯔꯣꯇꯔꯗ ꯂꯩꯕ ꯃꯁꯤꯡ ꯏꯔꯕ ꯀꯣꯠ 37 ꯒꯤ ꯃꯅꯨꯡꯗ ꯑꯃꯥ꯫ ꯑꯇꯣꯞꯄ ꯃꯃꯤꯡ: ꯁ꯭ꯂꯣꯠ, ꯀꯞ꯫ꯃꯄꯨꯡ ꯄꯥꯕꯤꯌꯨ → rather than about luck. That is the Lab’s subject. The sample required depends on the size of the bias, the test and the acceptable error rates. A short streak alone does not distinguish bias from chance; comparing explanations also requires assumptions about how plausible each was before the data.
ꯋꯥꯍꯪꯁꯤꯡ
ꯅꯠꯇꯦ꯫ ꯃꯁꯤ ꯃꯔꯤ ꯂꯩꯅꯗꯅ ꯍꯟꯖꯤꯟ ꯍꯟꯖꯤꯟ ꯊꯣꯀꯄ ꯊꯧꯗꯣꯛ ꯈꯨꯗꯤꯡꯃꯛꯇ ꯆꯠꯂꯤ: ꯗꯥꯏꯁ, ꯀꯣꯏꯟ, ꯂꯣꯇꯔꯤꯒꯤ ꯃꯁꯤꯡꯁꯤꯡ꯫
With more independent spins, the proportions tend toward their probabilities; they do not have to be close after any fixed number. Standard single-zero bets have an expected lossꯃꯦꯊꯃꯦꯇꯤꯛꯁꯃꯥꯡꯒꯗꯕ ꯁꯦꯜꯐꯪꯒꯗꯕ ꯃꯃꯂꯕꯨ ꯕꯦꯠ ꯊꯝꯂꯕ ꯁꯦꯜ ꯄꯨꯝꯅꯃꯛꯅ ꯃꯅꯥꯡꯕ; ꯁꯦꯁꯟ ꯅꯠꯔꯒ ꯃꯇꯝ ꯑꯃꯒꯤ ꯃꯅꯨꯡꯗ ꯃꯥꯡꯕ ꯑꯦꯚꯔꯦꯖ ꯁꯦꯜ꯫ꯃꯄꯨꯡ ꯄꯥꯕꯤꯌꯨ → of 2.70% of turnoverꯁꯦꯜꯇꯔꯅꯑꯣꯚꯔꯁꯦꯁꯟ ꯅꯠꯔꯒ ꯃꯇꯝ ꯑꯃꯒꯤ ꯃꯅꯨꯡꯗ ꯕꯦꯠ ꯊꯝꯂꯕ ꯁꯦꯜ ꯄꯨꯝꯅꯃꯛ, ꯑꯃꯨꯛ ꯍꯟꯅꯥ ꯊꯝꯂꯕ ꯃꯥꯏꯄꯥꯛꯇꯨꯅ ꯐꯪꯂꯕ ꯁꯦꯜꯁꯨ ꯌꯥꯎꯅ꯫ ꯑꯇꯣꯞꯄ ꯃꯃꯤꯡ: ꯍꯦꯟꯗꯜ, ꯑꯦꯛꯁꯟ꯫ꯃꯄꯨꯡ ꯄꯥꯕꯤꯌꯨ →, but your actual result can be far from that average.
ꯑꯦꯚꯔꯦꯖꯀꯤ ꯅꯤꯌꯝ ꯂꯩꯇꯦ꯫ ꯑꯆꯧꯕ ꯃꯁꯤꯡꯒꯤ ꯅꯤꯌꯝꯃꯦꯊꯃꯦꯇꯤꯛꯁꯑꯆꯧꯕ ꯃꯁꯤꯡꯒꯤ ꯅꯤꯌꯝꯍꯣꯠꯅꯕꯒꯤ ꯃꯁꯤꯡ ꯍꯦꯅꯒꯠꯂꯀꯄꯒ ꯂꯣꯌꯅꯅ ꯃꯔꯤ ꯂꯩꯅꯗꯕ ꯃꯍꯩꯁꯤꯡꯒꯤ ꯑꯦꯚꯔꯦꯖ ꯑꯗꯨ ꯐꯪꯒꯗꯕ ꯃꯃꯂꯗ ꯅꯛꯂꯛꯏ꯫ꯃꯄꯨꯡ ꯄꯥꯕꯤꯌꯨ → ꯂꯩ, ꯑꯃꯁꯨꯡ ꯃꯁꯤ ꯌꯥꯎꯊꯣꯀꯄꯅ ꯊꯕꯛ ꯇꯧꯏ꯫
ꯃꯦꯊꯃꯦꯇꯤꯛꯁꯀꯥ ꯁꯥꯟꯅꯕ
ꯃꯁꯤꯁꯤꯡ ꯑꯍꯥꯅꯕꯗ ꯅꯍꯥꯛ ꯃꯁꯥꯃꯛꯅ ꯍꯣꯠꯅꯕꯤꯌꯨ꯫ ꯄꯥꯎꯈꯨꯝ ꯈꯨꯗꯤꯡꯃꯛ ꯊꯨꯡꯗꯨꯅ ꯊꯝꯂꯤ; ꯁꯦꯝ ꯁꯥꯔꯕ ꯃꯇꯃꯗ ꯍꯥꯡꯗꯣꯛꯄꯤꯌꯨ꯫
ꯀꯣꯏꯟ ꯑꯃꯥ ꯑꯃꯨꯛ 10 ꯆꯥꯟꯊꯔꯕꯗ ꯍꯦꯗ 7 ꯑꯃꯁꯨꯡ ꯇꯦꯜ 3 ꯎꯠꯂꯤ꯫ ꯑꯃꯨꯛ 1,000 ꯆꯥꯟꯊꯔꯕꯗ ꯍꯦꯗ 517 ꯑꯃꯁꯨꯡ ꯇꯦꯜ 483 ꯎꯠꯂꯤ꯫ ꯇꯦꯂꯁꯤꯡꯅ ꯃꯇꯨꯡ ꯏꯟꯅ ꯌꯧꯔꯛꯂꯕꯔꯥ?
ꯄꯥꯎꯈꯨꯝ ꯎꯠꯄ
ꯄꯥꯎꯈꯨꯝ ꯅꯠꯇꯦ꯫ ꯈꯦꯠꯅꯕ ꯑꯗꯨ 4 ꯗꯒꯤ 34 ꯐꯥꯑꯣꯕ ꯍꯦꯅꯒꯠꯂꯛꯂꯦ꯫ ꯔꯦꯁꯤꯑꯣ ꯑꯗꯨ ꯈꯛꯇꯅ ꯆꯞ ꯃꯥꯟꯅꯕꯒꯤ ꯃꯅꯥꯛꯇ ꯂꯥꯀꯈꯤ꯫
The ratio of heads to tails fell from 7/3, about 2.33, to 517/483, about 1.07. This example illustrates how a proportion can move closer to one-half while the count difference grows.
The gap between them grew, from 4 to 34. With a fair coin a gap of around 30 after 1,000 tosses is ordinary: the typical gap grows roughly with the square root of the number of tosses, while the ratio settles toward 1.
Nothing pulled the tails back. The early surplus of heads was diluted by the next 990 tosses, not corrected.
ꯀꯣꯏꯟ ꯑꯃꯥ ꯑꯃꯨꯛ ꯃꯔꯤ ꯆꯥꯟꯊꯕꯤꯌꯨ ꯑꯃꯁꯨꯡ ꯍꯦꯗ ꯑꯃꯒꯤ ꯃꯇꯨꯡꯗ ꯊꯨꯅ ꯂꯥꯀꯄ ꯆꯥꯟꯊꯕꯁꯤꯡ ꯈꯛꯇ ꯌꯦꯡꯕꯤꯌꯨ꯫ ꯃꯈꯣꯌꯒꯤ ꯁꯔꯨꯛ ꯀꯌꯥ ꯍꯦꯗꯅꯣ? ꯐꯪꯕ ꯌꯥꯕ ꯁꯤꯛꯋꯦꯟꯁ 16 ꯒꯤ ꯈꯨꯗꯤꯡꯃꯛꯀꯤꯗꯃꯛ ꯍꯤꯁꯥꯕ ꯇꯧꯕꯤꯌꯨ ꯑꯃꯁꯨꯡ ꯁꯔꯨꯛꯁꯤꯡꯒꯤ ꯑꯦꯚꯔꯦꯖ ꯂꯧꯕꯤꯌꯨ꯫ ꯃꯗꯨ 50% ꯔꯥ?
ꯄꯥꯎꯈꯨꯝ ꯎꯠꯄ
ꯄꯥꯎꯈꯨꯝ ꯅꯠꯇꯦ꯫ ꯃꯗꯨ 17/42, ꯈꯟꯊꯕꯗ 40.5% ꯅꯤ꯫ ꯃꯑꯣꯡ ꯑꯁꯤꯗ ꯌꯦꯡꯂꯕꯗ, ꯆꯨꯝꯃꯤꯟꯅꯕ ꯀꯣꯏꯟ ꯑꯃꯅ ꯍꯦꯗ ꯑꯃꯒꯤ ꯃꯇꯨꯡꯗ ꯏꯪꯂꯛꯂꯤꯕꯒꯨꯝ ꯎꯏ꯫
ꯁꯤꯛꯋꯦꯟꯁ 16 ꯂꯤꯁ꯭ꯠ ꯇꯧꯕꯤꯌꯨ꯫ ꯃꯈꯣꯌꯒꯤ ꯑꯅꯤ, TTTT ꯑꯃꯁꯨꯡ TTTH, ꯗ ꯑꯍꯥꯅꯕ ꯆꯥꯟꯊꯕ ꯑꯍꯨꯃꯒꯤ ꯃꯅꯨꯡꯗ ꯍꯦꯗ ꯑꯃꯠꯇꯥ ꯂꯩꯇꯦ, ꯃꯔꯝ ꯑꯗꯨꯅ ꯍꯦꯗ ꯑꯃꯒꯤ ꯃꯇꯨꯡꯗ ꯀꯔꯤꯁꯨ ꯂꯥꯀꯄ ꯂꯩꯇꯦ; 14 ꯊꯃꯏ꯫
ꯈꯨꯗꯤꯡꯃꯛꯇ, ꯍꯦꯗ ꯑꯃꯒꯤ ꯃꯇꯨꯡꯗ ꯂꯥꯀꯄ ꯆꯥꯟꯊꯕꯁꯤꯡ ꯂꯧꯕꯤꯌꯨ ꯑꯃꯁꯨꯡ ꯃꯈꯣꯌꯒꯤ ꯁꯔꯨꯛ ꯀꯌꯥ ꯍꯦꯗꯅꯣ ꯊꯤꯕꯤꯌꯨ꯫ HHHH ꯅ 3 ꯒꯤ ꯃꯅꯨꯡꯗꯒꯤ 3 ꯄꯤ꯫ HHHT ꯅ 3 ꯒꯤ ꯃꯅꯨꯡꯗꯒꯤ 2 ꯄꯤ꯫ HTHT ꯅ 2 ꯒꯤ ꯃꯅꯨꯡꯗꯒꯤ 0 ꯄꯤ꯫ 14 ꯄꯨꯝꯅꯃꯛ ꯌꯦꯡꯂꯕꯗ: ꯁꯤꯛꯋꯦꯟꯁ ꯇꯔꯨꯛꯅ 0 ꯄꯤ, ꯃꯔꯤꯅ 1/2 ꯄꯤ, ꯑꯃꯅ 2/3 ꯄꯤ ꯑꯃꯁꯨꯡ ꯑꯍꯨꯃꯅ 1 ꯄꯤ (HHHH, THHH ꯑꯃꯁꯨꯡ TTHH)꯫
ꯁꯔꯨꯛꯁꯤꯡ ꯄꯨꯟꯁꯤꯟꯕꯤꯌꯨ: 4 × 1/2 + 2/3 + 3 = 17/3꯫ 14 ꯅ ꯌꯦꯟꯕꯤꯌꯨ: 17/42 ≈ 40.5%꯫
ꯆꯥꯟꯊꯕ ꯑꯃꯠꯇꯅ ꯇꯦꯜ ꯃꯥꯌꯀꯩ ꯂꯧꯗꯦ; ꯈꯨꯗꯤꯡꯃꯛ 50–50 ꯅꯤ꯫ ꯇꯦꯟꯊꯕ ꯑꯁꯤ ꯁꯤꯛꯋꯦꯟꯁ ꯑꯃꯃꯃꯒꯤ ꯑꯦꯚꯔꯦꯖ ꯂꯧꯕꯗꯒꯤ ꯂꯥꯀꯏ: ꯍꯦꯗꯅ ꯊꯜꯂꯕ ꯁꯤꯛꯋꯦꯟꯁ ꯑꯃꯗ ꯌꯦꯡꯒꯗꯕ ꯆꯥꯟꯊꯕ ꯀꯌꯥ ꯂꯩ ꯑꯗꯨꯕꯨ ꯑꯃꯨꯛ ꯈꯛ ꯊꯤ, ꯍꯦꯗ ꯑꯃꯈꯛ ꯂꯩꯕ ꯁꯤꯛꯋꯦꯟꯁ ꯑꯃꯒꯨꯝꯅ꯫ ꯃꯤꯂꯔ ꯑꯃꯁꯨꯡ ꯁꯥꯟꯖꯨꯔꯖꯣ (2018) ꯅ 1985 ꯒꯤ ꯍꯣꯠ ꯍꯦꯟꯗ ꯊꯤꯖꯤꯟꯕ ꯑꯗꯨꯅ ꯃꯁꯤꯒꯤ ꯕꯥꯁ꯭ꯀꯦꯠꯕꯣꯜ ꯁꯥꯟꯅꯔꯤꯕꯁꯤꯡ ꯃꯑꯣꯡ ꯑꯁꯤꯗ ꯌꯦꯡꯈꯤ ꯍꯥꯌꯅ ꯎꯠꯈꯤ, ꯃꯔꯝ ꯑꯗꯨꯅ ꯈꯨꯗꯣꯡꯆꯥꯕꯗꯒꯤ ꯍꯦꯟꯅꯥ ꯁ꯭ꯇ꯭ꯔꯤꯛ ꯂꯩꯗꯕꯒꯨꯝ ꯎꯔꯝꯕ ꯁꯥꯟꯅꯔꯤꯕꯁꯤꯡ ꯑꯁꯦꯡꯕꯗ ꯈꯔꯥ ꯍꯦꯟꯅꯥ ꯁ꯭ꯇ꯭ꯔꯤꯛ ꯂꯩꯔꯝꯃꯤ꯫
ꯅꯣꯛꯅꯕꯒꯤ ꯀꯃꯤꯛ

ꯌꯥꯎꯊꯣꯀꯄꯅꯤ, ꯆꯨꯃꯊꯣꯀꯄ ꯅꯠꯇꯦ꯫ ꯆꯛꯔꯅ ꯀꯅꯥꯒꯨꯝꯕ ꯑꯃꯒꯤꯁꯨ ꯁꯦꯜ ꯄꯤꯒꯗꯕ ꯂꯩꯇꯦ꯫
ꯍꯧꯔꯛꯐꯃꯁꯤꯡ
- Feller, W. An Introduction to Probability Theory and Its Applications, Vol. 1.
- Schwartz, D. G. Roll the Bones: The History of Gambling.
- Ethier, S. N. The Doctrine of Chances: Probabilistic Aspects of Gambling.
- Epstein, R. A. The Theory of Gambling and Statistical Logic, 2nd ed.
- Croson, R. & Sundali, J. “The gambler’s fallacy and the hot hand: empirical data from casinos.”
- Chen, D. L., Moskowitz, T. J. & Shue, K. “Decision making under the gambler’s fallacy: evidence from asylum judges, loan officers, and baseball umpires.”
- Clotfelter, C. T. & Cook, P. J. “The ‘gambler’s fallacy’ in lottery play.”
- Gilovich, T., Vallone, R. & Tversky, A. “The hot hand in basketball: on the misperception of random sequences.”
- Miller, J. B. & Sanjurjo, A. “Surprised by the hot hand fallacy? A truth in the law of small numbers.”
- Hacking, I. “The inverse gambler’s fallacy: the argument from design. The anthropic principle applied to Wheeler universes.”