泥中蟠龍's Game愛歌 [A love song for games of the dragon waiting for an opportunity] Survivorship bias
泥中蟠龍's Game愛歌
[A love song for games of the dragon waiting for an opportunity]
Survivorship bias
December. 14. 2017.
During World War II, United States Armed Forces conducted a study of the damage done to aircraft that had returned from missions to minimize bomber losses. The study showed a tendency by the returning aircraft, in particular, to be hit on the wings and tail with a bullet and had recommended that armor be added to the areas that showed the most damage. However, questioning the analysis result, a researcher proposed that the Navy reinforce areas where the survived military aircraft were unharmed: cockpit and engine. The mathematician commented that the study didn't consider bombers that had been shot down in the engine or cockpit in their missions, while the majority of aircraft with holes on wings and tail returned home safely. If he hadn't detected the fallacy of the data analysis, the U.S. Army may have wasted a lot of budget due to unnecessary schemes and more fighter pilots could be killed. It's a form of selection bias and generally called "Survivorship bias" or "Survival bias".
While talking about commercially successful video games with developers, they used to say that they can succeed in attracting users if only they copy the elements which the top-ranked games have. However, failed imitative video games are more numerous than ones well receiving in markets. Of course, I don't mean that those key elements that make a game successful are bad or wrong. But if developers make the mistake of hasty generalization that some elements determine the success of the game, biased data is more likely to lead them toward the wrong directions. There are dozens of reasons that a successful game is loved by customers, whereas that could be the only way that a failed game is shunned by users.
Mentioning successful mobile-version games—Lineage series, TERA, Yulgang, and Fate/stay the night, etc.—based on well-known originals, quite a lot of people say that a mobile game built on the famous original must have achieved popularity and high profit. However, despite a notable original, I've seen a messed-up game with the poor outcome a hundred times. Game users' love of the latest successful games is not because they are constructed on well-made original works but because it just so happens that they are based on the originals.
Let's go back to the beginning of this article. "Survival bias" is a mistake that anybody can make. We need to look at the data from multiple points of view not to go astray. Game makers should see the market in a different direction while lots of uniformly similar-trend video games are released. The only way to reduce the probability of failure is that those game developers take a closer look at whether they have missed an element of success among excluded factors or they haven't found it yet. I hope that as many as game developers can return safely from the fierce battlefield for success.
※ This is from Kyunghyang Games column by 泥中蟠龍 since September 2013.
(http://www.khgames.co.kr)
Translation by Kim Ki-hui
Comments
Post a Comment