GUESSING CONTESTS IN CASES WHERE STATISTICS ARE TO BE CONSIDERED.

Sec. 125. In Hall vs. Cox, 1 Q. B. (1899), page 198, the record shows that the publisher of a paper, called the Rocket, offered a prize of £1,000 to its readers for a correct prediction of the number of male and female births and the number of deaths in London during a specified week. The competitors were required to give their estimates on vouchers published in the paper and they were not limited to one prediction. Consolation prizes were offered, under the same conditions, for correct predictions or approximations to the number of male and female births and deaths respectively. On appeal to the Court of Appeals this scheme was held by Justices Smith, Rigby and Collins not to be a lottery, following Caminada vs. Hulton (1891) 60 L. J. (M. C.) 116 and Stoddard vs. Segar (1895) 2 Q. B. 474. Smith, L. J., said: "The result, no doubt, depends largely on chance but not entirely and the cases show that to constitute a lottery it must be a matter depending entirely upon chance. There is an element of skill in the inquiry in this case dependent on the investigation of the returns of the previous years and the conditions of the increase of the population, the death rate and such like statistical investigations."

Sec. 126. Here, as in the news paper case decided by the Attorney-General and the Toronto and New Brunswick cases, the Court loses sight of the fact that each guesser reaches a point in his estimates where he realizes he is in the field of speculation and chance and especially in this case that the guesser had to give the exact number of female births, male births and deaths before he could receive the £1,000 prize. In such a case the probabilities of losing are greatly increased. Suppose there were only two numbers in the female births, two numbers in the male births and two numbers in the deaths, then the guesser would not have sufficient data to enable him to determine which of the two in either case he ought to name. The chances of naming the correct

number in each class would be even, but the chances of his naming the correct number in two classes would be one in four and in three classes one in eight. Cham. Encyc., Title "Probability, The Mathematical theory of," and Ency. Brit. Same Title.

But the numbers about which the guessers would have doubt would be largely in excess of two and in such case the chances of naming the correct number in all three classes might amount to thousands.

The promoter of this scheme evidently thought the result of the contest depended entirely on chance. This clearly appears from the fact that he offered a prize of £1,000 ($5,000) to the winner and from the further fact that he gave each contestant as many guesses as he saw proper to make. He realized that the probability of an absolutely correct guess as to the number in the three classes was so remote that he could afford to take the risk he did take and the prize was so large the people would be induced to spend large sums of money in the aggregate to win it. Besides the capital prize offered to the person making an absolutely correct prediction he offered consolation prizes (to be determined by him) for approximations to the correct number.

This was unquestionably a gambling transaction and the prize was made dependent on chance in its gambling sense and according to the course of decisions in the United States it was a lottery scheme.

Sec. 127. A newspaper advertised an enterprise or scheme by which it offered to give the sender of the first guess, giving the correct or nearest correct number of votes of the Democratic and of the Republican candidates, respectively, for the office of Secretary of State for the State of Ohio at the election next ensuing, $100 each; and to the sender of the second correct or nearest correct guess (if no correct guesses should be received) of the vote of either candidate, $50 each; and to the sender of the third correct or nearest correct guess (if no correct guesses should be received) for each candidate

$25; and $5 each to the senders of the next fifteen correct or nearest correct guesses (if no correct guesses should be received) on each candidate thus offering to give the sum of $500 to thirty-six persons. The guesses were to be made on blanks, cut from the paper. The Attorney-General of the United States, Hon. W. H. Miller, in passing on this scheme, on Jan. 24, 1890, held that the award of the prizes did not depend either on lot or chance. The General used this language:

* *

"It will hardly be contended that the enterprise under consideration was dependent on lot. Was it dependent upon chance within the meaning of the statute? In a certain sense and in a certain degree, perhaps, any prediction, as to human action, may be said to be dependent upon chance; that is to say, it is in some measure dependent upon circumstances, the happening of which can not be anticipated or foretold with any degree of certainty. * A student of Statistics might know approximately the number of Republican votes and the number of Democratic votes in the State of Ohio; he might approximate the ratio, in which one or the other might increase or decrease in a given year. It is quite likely that his estimates would often be wide of the mark, but it would not be by reason of chance but by reason of causes in regard to which he had formed erroneous estimates. It would hardly do to say that a child or a school boy could form as correct an estimate in the matter, as an experienced politician, who had given weeks and months of steady attention to the consideration of the question. But without further elaboration, I am quite clear that estimates, made upon the probable political action of the people, in a given State, in a pending election, can not be said to be dependent upon chance within the meaning of this statute." It was argued by the Attorney-General, in this case, that a prediction, that a man, who has for a long time, maintained certain political opinions, will continue on the same line, is not dependent on chance. He admitted that a man might utterly change his opinions but such change would not be purely a matter of chance.

This opinion can not be supported on reason or authority and its influence on the administration of the Lottery Act has been deleterious in the extreme. The gamblers never failed to cite this opinion in support of their proposed schemes and many times it had to be conceded that, if the principle,announced therein, was applied, many gift enterprises, which were excluded from the mails, should have been admitted. If the fact, that stakes are laid on the issue of an event, be a test of the presence of the chance element in it then the result of an election unquestionably has in it that element for, it is well known, that there is, probably, more money lost and won on the result of elections than any other one thing, unless it be cards and horse races. It is not claimed that the result of an election depends on chance, any more than the future state of the market in regard to stocks or grain depends on chance, but it is insisted that so far as the adventurers are concerned there is chance, because they do not know and can not know, and can not, in advance, by their knowledge and foresight, predict or determine or control the result. But Attorney-General Miller undertook to show that a person of experience and political knowledge would, in such a contest, have an advantage over one of inexperience and ignorance and he thought a child or school boy could not form as correct an estimate in such a matter as an experienced politician, who had given weeks and months of steady attention to the consideration of the question; and yet he said that while a student of Statistics might know approximately the number of Republican and Democratic votes in the State of Ohio and might approximate the ratio in which one or the other might increase or decrease in a given year, it was "quite likely his estimates would often be wide of the mark," and then he adds that this would not be "by reason of chance but by reason of causes in regard to which he had formed erroneous estimates." The statement that a student of Statistics, who had devoted months to the study of the subjects, would quite likely be wide of the mark in his estimate of the votes, that the Republican and Democratic candidates would receive at a given election, concedes the whole

ground, especially, when it is further stated that the failure to make a guess not "wide of the mark" would be by reason of causes in regard to which the guesser had formed erroneous estimates.

Sec. 128. Do not these statements bring the uncertainty here exactly within the definitions of chance heretofore given? If the most experienced man may quite likely be "wide of the mark," because there are many occult causes, in regard to the operation of which he forms erroneous estimates, what would be the status of the masses of the people to whose gambling spirit such a guessing contest directly appeals? It is a fact that the most experienced politician would, in his estimates of the votes of the respective candidates, reach a point where he would be conscious of having entered a field of mere speculation, of mere luck or chance and in selecting a number his mental condition would be the same as in the selection of a number in a lottery or raffle. In either case he would be conscious of groping in the dark and that if he should select a winning number it would be due wholly to chance. He might be influenced in the selection by sentiment or by whim, such as taking an odd or even number, or by taking what are popularly supposed to be lucky numbers, but at the same time he would feel his knowledge of the subject was of no use to him. These observations apply, of course, with certain limitations. Every intelligent guesser would feel sure of his ground up to a certain point; but what is meant to be asserted here is that he would, in his estimates, finally pass the line between certainty and uncertainty and enter the region of doubt, where choice would rest upon chance and not upon his knowledge, foresight or sagacity.

That the Attorney-General was correct when he said the experienced statistician would quite likely be "wide of the mark" is shown very clearly during every election, when committees of the political parties, after two or three most careful polls of the State or district, give out estimates, which astonish them when the actual count is announced; and two months after the presidential election of 1896 five leading newspapers