Data for nerds

There's room for many interpretations, so I'll try to be as objective as possible and let each take its conclusions from all this information. Now I leave you with the data, have a good read.

Comments

One of the main data points, and the one you have already seen, are comments, both over discord, the website, and Reddit. We have been reading them all, seeing a wide spectrum of reactions, but it's undeniable there was a decent proportion of people against the name change.

The problem with this data point is that it's not representative because it is just a tiny percentage of people, not everyone is comfortable speaking out publicly, and there is a large potential for bias surging from the discussion. It's easier for someone to join the conversation if their point of view goes in the current direction than if they are an outlier. The first discussions we had in the #exclusive channel in the SimpleRockets Chat Discord Server resulted in an excess of arguments in favor of changing the name, and in the more open discussions, the opposite.

Poll

To get more organized and detailed feedback, we decided to make a form, that came with its problems. There was no limit to how many times someone could answer the poll, and there are some obvious duplicated votes. There was a total of 688 votes, with 70 users admitted to having a pirated copy of the game, representing over a 10%. We assume the number is larger, so we appreciate the honesty of those that admitted to doing it. The following graph shows where voters came from.

The main point of this poll was to ask people how much they liked each of the proposed names from 1 to 5. The questions were presented in a random order to everyone. The two following graphs show the distribution of votes for each name (1 left, 5 right), and then the average and the standard deviation. For those not familiar with the concept of deviation, higher values usually represent more polarised results.

When going through the data to filter outliers (obvious duplicates, people saying they played the game for millions of hours...) I noticed that there was a trend in the early votes, with some people voting 5 points in favor of SR2 and 1 for everything else. The meaning of these votes is obvious: Don't change it. But I was interested in how much people liked each name and this was drifting the results a lot.

To better analyze the results I added a flag to those voting exclusively or almost exclusively to SR2 (for those interested, the flag checked if the SR2 vote was more than 2.5 times larger than the geometric mean of all the other options). This resulted in 192 votes being flagged, representing 27.9% of the total votes. The following graphs show the results of the remaining 72.1% votes.

To see how much the perception of this name change drifted over time, here you can see a graph showing the accumulated percentage of flagged votes, to the left there is a lot of variation because the first votes had a larger proportional impact in their accumulated percentage. There is a clear trend towards a smaller percentage of flagged users over time.

Ads

Asking the community could come with certain biases, so we discussed the best approach to get raw impressions from people that don't know the game would be. The best idea we came up with was using Reddit ads to evaluate the click-through rate of each name. We are aware of how biased this approach is too, we don't know if someone misclicked, if they already knew SR2... But, again, is the best we could try without expending a lot of money.

We targetted two groups: The KSP subreddit, for obvious reasons (there is a large overlap in our target audiences), and the Games subreddit, where we expect a lot fewer people to have ever heard of SR2. We have no control over how many times an ad is shown, our best guess is that Reddit shows the successful ones more, so the times they were shown and the total amount of clicks isn't relevant. These were the results: