Page:CIAdeceptionMaximsFactFolklore 1980.pdf/39

C00036554 * increases as the likelihood of compromise given use increases--i. e., if the asset is less likely to be able to be used again, bigger stakes are required to justify its use, and
 * decreases as the discount factor increases and/or the cost of maintenance increases.

Both of these results are in accord with intuition. What is somewhat unexpected, however, is that, for opportunity distributions that are highly skewed (many opportunities of low value and progressively fewer of high value) as might be expected in practice, the optimal threshold is not highly sensitive to the above factors. Moreover, for highly skewed distributions of future opportunity the analysis shows that it pays to wait for high stakes (big opportunities) despite risks of compromise and/or costs of maintenance. This latter finding is particularly intriquing as, according to Axelrod (46);

Turning the perspective around, one can see that it would be a mistake to evaluate the opponent's resources for surprise by what you have seen when the stakes were low or moderate. He may be rationally waiting for an event with sufficiently large stakes to justify the exploitation of whatever resource for surprise he has.