Other ways of grouping up species have also been tried, but none gave a significant result.ĭespite not getting a result here, we are still very interested in the distribution of individual species from eggs and we will be running further, more specific, experiments in the future. This test gave a p-value of p=0.1728, which means we did not find a significant difference in the contents of 5km eggs of the different researchers. We ran a chi-squared test similar as before to the counts of Water and non-Water Pokemon from 5km eggs. It's important that we look at 5km eggs only if we looked at all eggs, a difference in species distribution could be caused by the difference in distance distribution we just discovered. One way to do this is to compare the distribution between Water and non-Water Pokemon from 5km eggs. The counts for individual species are too low to do any meaningful testing, so we had to group different species together. Under the assumption that this is true (most importantly: hatching location is not important) we can research the contents of the eggs. It is believed that the contents of an egg are already determined upon receiving it at a PokeStop (largely due to the fact that Pokemon with legacy movesets were hatched after global moveset changes, among other compelling reasons). We continued by looking at the individual species hatched. Further Testing (Species Distribution, etc.): This is lower than the 0.05 threshold for significance, so we have concluded that the hypothesis that distance distribution is independent of PokeStop has been refuted (See Note 2). ![]() We can now calculate the p-value, this is the probability that a random variable following a chi-squared distribution with 25 degrees of freedom is 43.905 or larger. The number of degrees of freedom is calculated as: If the egg distribution were independent of PokeStop, this total would follow a chi-squared distribution with 25 degrees of freedom. We add up the values for all trainers, and come up to a total of 43.905. These values give a measure of how much the expected and observed counts differ from each other. Next we take the square of the difference between the expected and observed count and divide it by the expected count. ![]() Researcher jFarr hatched a total of 169 eggs (not including 10km eggs). Then we will test whether these expected counts differ significantly from the observed (actual) counts. The first step of the chi-squared test is to calculate the expected number of 2km and 5km eggs for each researcher, based on their total number of hatches and the egg distribution of all trainers. This left a total of 1,715 eggs to analyze, 669 2km eggs and 1,046 5km eggs. So we looked only at the distribution between 2km and 5km eggs. We unfortunately obtained very few 10km eggs, so we are unable to include these in the test (See Note 1). This test checks whether it's plausible that the distance distribution is independent of PokeStop. We decided to run a chi-squared test for independence on the data. The distribution for each researcher is given in the following table: One important reason for starting with this is because the distance is already determined at the PokeStop and the hatching location cannot have any influence on it. Now for the fun part: how we arrived at our conclusions! Analysisįirst, we examined the distribution of egg-distances (ie, 2km, 5km, or 10km) for eggs from each PokeStop. Look forward to more research as we continue learning about PokeStops and Eggs. In other words, we witnessed some travelers receiving significantly more 2k eggs and others receiving significantly more 5k eggs from their PokeStop of choice! We can't yet say why, but the fact that not all PokeStops were observed to give out the same egg odds is a big deal. SummaryĪfter analysis, we are confident that not all Pokestops award eggs from the same distribution. ![]() In total 1,841 eggs were hatched, and the contents have been recorded and analyzed. ![]() Nearly all 26 researchers managed to acquire the target 50, and some managed to get up to 180. We still don't know all the answers, but today we get to share a new, significant finding!Ģ6 Silph Researchers undertook a grueling task: each researcher collected and hatched 50 eggs from a single Pokestop, allowing us to analyze whether different Pokestops gave different eggs. We are often asked, "Do some PokeStops give out different egg species?" or "Do some PokeStops give out more 10km or 2km Eggs?"īecause of the difficulty in gathering this data, the world has been in the dark regarding PokeStop egg drop rates and distributions.
0 Comments
Leave a Reply. |