You are forced to choose. And that's great. From your votes we can see that there are 5 new games that you enjoyed, but not enough to level them with old classics.
Making someone choose is neither great nor bad, it's just an analysis bias. In every survey, the type of question you ask determines the type of result you will get. This is the basis of research methodology.
Placing a reduced number of votes/points/whatever changes the weight and scope of each vote. But the 25 value, regardless of whether it's the same as the previous poll or not, is only arbitrary - it's not a choice that comes from a deliberate analysis or one that was consciously determined on the basis of the number of existing games. I'm not saying with that, and I say that again, that the results of the poll will be good or bad, but that they will only reflect the data according to a very specific bias. There are several scenarios that are not contemplated by this situation.
For example, someone who considers that there are more than 25 incredible games. Taking any of these games off the list would be a random decision, and by not citing these games, the poll will factually be distorting the importance that the games have for those who answered the question and therefore will not correspond with reality. Another example, and this would be my case, the person picks some games that he considers as incredible games, but does not believe that all are in the same category. So he would be forced to either ignore the difference in score to cite all games (distorting the weight of each game) or remove some games from the list (distorting the representativeness of these games). My case is the second, I literally threw dice to pick that games would enter my list.
For example, let's use this same methodology to analyze "the fastest runners of all time". Imagine that each poster here would be the equivalent of a specific race, where only the first 25 runners would be listed as a result. In a given race, the slowest runner listed completed the race in 10 minutes and 6 seconds. (This could be analogous to how much the poster enjoyed some game.) However, the 26th placed runner completed the race in 10 minutes and 7 seconds. Technically, both are almost tied, but one will be on the list and one will not. So we got another race (poster). His top 20 finishers made incredible times, with less than 7 minutes. However, the last 5 places out of the 25 mentioned had times above 12 minutes. These last 5 runners will enter the list of "best runners of all time", but the runner who performed above them will not appear on the list because of the rules of analysis. I bet you can see the problem here.
Again, this is just one of the problems with the methodology used here. Do I mean to say that the results are useless? Not at all, but they will undoubtedly show an incomplete view of reality. It is very possible that games considered universally as good may have a low representation in this list, precisely because of the method chosen. They are games that the consensus would say are definitely among the best, but by the points system used here, may end up not appearing or appearing in positions that do not correspond to the real value that posters would give to it.
But hey, I wouldn't have the patience to do another parallel poll, so a new poll is better than no poll at all.
