Predicting Professional Sports Game Outcomes from Intermediate Game Scores
Predicting Professional Sports Game Outcomes from Intermediate Game Scores
复制标题
从中间比赛得分预测职业体育比赛结果
DOI：
10.1080/09332480.1992.10554981
复制
发表时间：
1992
期刊：
影响因子：
--
通讯作者：
F. Mosteller
中科院分区：
文献类型：
--
作者：
H. Cooper;Kristina M. Deneve;F. Mosteller
“今晚你会看篮球比赛吗？”“也许最后一节，这才是最重要的。”这样的交流，篮球迷们已经听过很多次了。民间智慧表明，职业篮球比赛的结果在很大程度上取决于最后 12 分钟的情况，即使不是最后 2 分钟（已知会持续半小时，实时）。相反，棒球迷最近被评估球队（尤其是替补投手）在第七或第八局后保持领先的能力的统计数据所困扰。这些统计数据使人们越来越认识到，大多数棒球比赛在家乡球迷休息的时候就已经结束了。这些看法准确吗？每个都可以通过比较在某个中间点领先的球队与最终结果来进行实证检验。对于这两项运动的预测是相反的，因为预计篮球的相关性很小，而棒球的相关性应该很大。什么构成小关系或大关系？在这种情况下，与大多数情况一样，大和小的定义是相对的判断。因此，我们必须假设球迷使用的比较单位是其他团队运动，例如职业足球和曲棍球。据我们所知，以前没有尝试客观地评估有关篮球比赛的共同信念的有效性。然而，我们确实知道，职业篮球比赛表现出相当大的主场优势，这至少应该部分抵消民间传说。这导致了一个明显的悖论，即在篮球比赛开始之前，球迷可以预测主队可能获胜（通常是这样），但他们也认为结果不能通过第三节得分来（很好）预测。就棒球而言，扑救和“扑救失败”，或者投手在比赛结束时失去领先优势的次数，现在是固定的——
" Are you gonna watch the bas ketball game tonight?"" Maybe the last quarter, that's all that counts." Fans of basketball have heard this exchange many times. Folk wisdom suggests that the out comes of professional basketball games are heavily dependent on what happens in the final 12 min utes, if not the final 2 minutes [which are known to last half an hour, real time). Conversely, base ball fans have recently been del uged by statistics that evaluate the ability of teams, and particularly of relief pitchers, to hold leads af ter the seventh or eighth inning. These statistics have led to a grow ing perception that most baseball contests are functionally over about the time the hometown fans take their stretch. Are these perceptions accurate? Each can be empirically tested by comparing the team leading a game at some intermediate point with the final outcome. The predictions are opposite for the two sports in that the relation for basketball is ex pected to be small, whereas the re lation for baseball should be large. What constitutes a small or large relation? In this circumstance, as in most, the definitions of large and small are relative judgments. So we must assume that the unit of com parison used by fans would be other team sports, say, professional football and hockey.We know of no previous at tempt to objectively assess the va lidity of the common belief about basketball games. We do know, however, that professional basket ball games exhibit a considerable home-court advantage that should at least partly countervail the folk lore. This leads to the apparent paradox that before a basketball game begins, fans can predict that the home team is likely to win (more often than not), yet they also believe that the outcome can not be predicted (well) by the third-quarter score. With regard to baseball, saves and" blown saves," or the number of times a pitcher loses a lead at the end of a game, are now stand-