Keeping Up with the Quants: Your Guide to Understanding and Using Analytics by Thomas H. Davenport & Jinho Kim

Author:Thomas H. Davenport & Jinho Kim [Davenport, Thomas H.]
Language: eng
Format: mobi
ISBN: 9781422187258
Publisher: Harvard Business Review Press
Published: 2013-05-20T14:00:00+00:00

When Results Don’t Imply Actions

We expect that the results from quantitative analysis will lead to action, but sometimes perfectly good results don’t translate into specific actions, although they are still useful to know. For example, Jinho once collaborated on a study of the value of moving first in the game Go. Go, one of the oldest strategic board games in the world, takes simple elements (lines and black and white lens-shaped pieces, or “stones”) and applies simple rules. Yet despite the apparent simplicity, the game generates subtleties that have enthralled players for thousands of years.12 In Asia, especially in Korea, Japan, and China, tens of millions of people enjoy playing Go, and nearly a thousand professionals make their living competing in tournaments that offer millions of dollars in prize money.

At the beginning of the game the board is empty. One player takes the black stones and the other player the white stones. Black and White alternate turns, with Black moving first. That confers an advantage, but no one had ever studied just how much an advantage. Since 1974, in professional games 5.5 points have been given to White at the start of a game to compensate for Black’s having the first move. This is called komi.

Jinho analyzed data from 577 professional games in Korea to understand whether 5.5 points is the right value of komi. The results suggested that it was too small overall. However, a complication emerged in the analysis. The data suggested that moving first is an advantage only for the player who is skillful enough to use it. One theoretically plausible action that might be taken based on these results is komi discrimination—that is, both (1) maintaining the current level of komi (5.5 points) for those players moving first whose play indicates that the current komi is not likely to be associated with winning a game, and (2) increasing the komi for other players. It seems unlikely that such komi discrimination would be acceptable in professional tournaments. Therefore, a good analysis (that led to a prominent publication for Jinho and his colleague in an academic journal) did not lead to any specific action.13

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.