[ExI] Lotto

[Damien Broderick]

On 7/2/2010 12:19 PM, BillK wrote:

> you were arguing with John about casinos.   ;)

Only by accident. I started talking about Lotto (since I was challenged 
about my book on that topic), then got sidetracked into a more general 
assertion about lotteries, then JKC whipped in casinos. It's obvious 
that the casino is an appalling arena to look for small effect size 
interventions of any kind, including (as you noted) the owners' quick 
way with anyone obviously taking too much dough from them. I'm 
interested in evidence for or against interventions in real life from 
psi, and lotteries (especially lotto) where you get to nominate the 
number/s of your choice are far more relevant than the bear pit of casinos.

In case anyone is interested, here is a sample of the preferences 
factors at work in Tattslotto in the draws I examined. Each cell is the 
raw vote for each number rescaled to a mean expected value of 100%:

  SATURDAY DRAW  997

WINNING NUMBERS  20 22 27 32 37 40
   SUPPLEMENTARY NUMBERS  23 38

TOTAL IS   58,273,422     MEAN IS  1,294,964.933

   1             1162572            89.78
   2             1263090            97.54
   3             1353445           104.52
   4             1318749           101.84
   5             1374645           106.15
   6             1271508            98.19
   7             1574496           121.59
   8             1340856           103.54
   9             1300274           100.41
  10             1485755           114.73
  11             1544334           119.26
  12             1446903           111.73
  13             1564392           120.81
  14             1267285            97.86
  15             1294608            99.97
  16             1215894            93.89
  17             1303089           100.63
  18             1434662           110.79
  19             1569658           121.21
  20             1601798           123.69
  21             1464726           113.11
  22             1346668           103.99
  23             1346843           104.01
  24             1174756            90.72
  25             1264991            97.69
  26             1332128           102.87
  27             1531800           118.29
  28             1487852           114.90
  29             1382858           106.79
  30             1430302           110.45
  31             1227272            94.77
  32             1054380            81.42
  33             1122991            86.72
  34             1194937            92.28
  35             1299758           100.37
  36             1306096           100.86
  37             1248649            96.42
  38             1123961            86.79
  39             1079271            83.34
  40             1201223            92.76
  41              896145            69.20
  42              963186            74.38
  43             1036134            80.01
  44             1007458            77.80
  45             1061024            81.93

You can use this to maximize your payout in those rare instances where 
the least favored numbers are drawn together. But the tax taken off the 
top still makes this unrewarding.

Damien Broderick