I play on another server that has the first option (1.) for map votes as "Any map will do", so I am bringing this as a suggestion to the community. This would decrease the influence of 1 spammers / undecideds / new players who don't know the maps on the vote outcome as their vote essentially would not count. I'm not exactly sure how the algorithm works but as you can see from the photos below, the most voted for map wins the vote despite more people selecting the "Any map" option: A possible alternative form of this is that the "any map" / "no preference" option actually counts, and if a majority vote for this then a random map is selected from the list. This could help to increase the variety of maps that are played on the server. I am considering predominately the jailbreak server but this could be applied to any of the servers.
I don't think this one screenshot is much evidence to suggest this is the case. That map is a popular map on the server and is often selected by regulars and wins votes, so that is most likely why it won. This is ignoring the fact that the sample size is also incredibly small, you're talking about 2 for the 2. option, and 1 for each other map that was voted for.. I don't believe one spammers / undecideds (who this is mainly directed at) would try to subvert the system by spamming 2 instead... I just don't see why that would happen. You can also can see that most of the votes are for the no preference option. --- Double Post Merged, Jun 5, 2018, Original Post Date: Jun 5, 2018 --- I'm not really talking about people who press one accidentally. I mean "1 spammers" as people who press 1 because they either don't care, don't want to make a decision/are undecided, or don't know any of the maps.
Quite an old suggestion that pops up every few years. Many servers have some way of preventing one spamming including removing 1 as an option or have it as random or that the vote doesn't count. One server had a huge annoying sentence in the voting menu every map advising players to carefully choose which map they want to play instead of spamming the vote. All to prevent the one spam. However I don't see one spamming as a problem. If people want to press 1, not minding which map comes next then that's their choice. If they change their mind they can revote. It is also quite easy to influence players especially via mic to sway the votes to a specific map. Not to mention it's not like one spamming is a constant thing. If there is a more popular map on any other number than 1 then that map has more of a chance of being chosen anyway. Don't forget revote. It would happen because they don't care what map comes next or they want to get the menu off from their screen for example. Pretty straightforward.
Then they would select the "no preference" option which is the first option? That serves the purpose of removing the menu icon and is instinctively easier to press being on number 1 / the first option.. Not to mention the most obvious option if you don't care about the next map is to select "any map" / "no preference".. --- Double Post Merged, Jun 6, 2018, Original Post Date: Jun 6, 2018 --- This is precisely one of the reasons for suggesting this change. Adding the option would offer a specific option for people that don't care which map comes next.. And that is absolutely their choice I agree. The point is that their choice to do this should be accurately reflected in the voting system and should not negatively impact players voting that actually have a preferred map they want to play. Two scenarios.. a) (current system) option 1 is a map that nobody specifically wants to play, option 2 is a map that a lot of players on the server want to play. 51% of players hit option one because they don't care about the next map, i.e. are "1 spammers". 49% of the server vote for option 2 because they specifically want to play that map. Option 1 wins, a map that nobody on the server wanted specifically to play and the players that voted for their preferred map lose out. b) (changed system) option 1 is "any preference", option 2 is a map that many on the server want to play. 51% vote "any preference" because they don't care about the next map. 49% vote for option 2 because they specifically want that map. Option 2 wins and is selected. The 51% that voted for any map because they had no preference are happy because they didn't care which map was selected. The 49% are also happy because their specified preferred map was selected. Everybody wins in this scenario. I think the change is just logical. Are there any problems that could occur with the change? In other words, what negative effects would the new system have in place of the old one? I propose that if there aren't any downsides to making this change, then there's no reason not to test implementing it. In my opinion, for the reasons I've highlighted, it could only have positive effects. Admins have already adjusted to the 1 spamming phenomena by making the first option of sm_vote 's as "spam".. Why not apply this to the map votes? --- Double Post Merged, Jun 6, 2018 --- It's worth pointing out as well that this suggestion wouldn't affect the !revote command. It would still remain usable and serve it's function the same as it already does.
I don't believe it negatively impacts other vote options. This is where we disagree. Your two scenarios greatly generalize your argument like one spamming is a constant issue. It isn't. I am not fond of that (you may not have intended an exaggeration but that is how I read it). You don't mention revote in regards to one spamming or players asking other players to vote for another map which, from my experience pretty much happens every round on busier servers. This greatly reduces the probability of map option 1 being selected. But let's just go with your scenario. In your scenario it is 100% reasonable yes and were this the case every time then your suggestion would have been implemented years ago. It hasn't so far because it hasn't been deemed a needed addition. I.e. the one spamming doesn't pose a problem. (As a side note though, if 51% of players are happy then democracy has prevailed anyway...) To spare you a paragraph of writing, I imagine you would then ask why not implement it anyway for your scenario whether it be rare or common as it seemingly has no downsides, it being "logical". Although I can't speak for Nomy on the changes he would have to make to make it happen, I think it just doesn't justify the change. The "negative" impact on a one spammed map over a preferred, more popular map is limited. Maybe some frustration. I doubt a mass D/C from most players is the most likely scenario at that point. They can rtv. I could argue that a less popular map being played at the time sparks conversations about changing the map, getting conversations going between players, making the server more social, vibrant, better even. I could also argue that this is the way it has been since the start and that we have a steady player base that knows how the voting system works and new players will quickly adapt to it because it isn't complicated. Plausible scenarios aren't hard to create.
They are just models to show what can happen with the current system and what the change would solve. I could use more realistic figures like 40% option 1, 35% option 2, 15% option 3, 10% option 4.. but the model is just for demonstrative purposes so that's why I used the extremes, this is a fairly common thing to do in my experience.. It's not necessarily a constant issue, but it is a regular occurrence, and it can be an issue. Scenarios like the one I describe above (40%..35%) are not uncommon, especially when a lot of players don't know the maps but some do and would like to express a preference. And although small, you have to take into account that some players are not aware of the !revote option. One of the problems with the current system that I would argue makes it a relatively common issue is when a lot of players don't know which map to pick. For example if there are several popular maps, or entirely unpopular maps. In the former, everyone will be expressing their preference in chat to undecideds saying "pick this" or "pick that", and when one of the popular maps happens to be on option 1, this wins simply because of it's happenstance position on the map voting system. This is a common occurrence when this scenario arises, and it does skew the results of the vote in favour of the map that happened to be on position 1. In the latter, often some players will know one of the maps and consider it the best and a good unpopular map... but because of the fact that none of them are generally known by the players of the server, the 1 option wins out just because people don't know any of the maps. Sure, in this scenario when people express a preference it's more often considered and people are more likely to !revote.. but people trust what they know often more than they do what people say, and if they don't know any of the maps, they're more than conceivably just going to hit 1 and hope for the best. The outcomes that I've seen playing on the server suggest this does happen. The any map option would give more power to the players that have knowledge and preference here and in my opinion deservedly so. In my experience on the jailbreak server, !revote doesn't very often significantly affect the map outcome in the context of what we're talking about. For starters, I only usually see on average 2-5 !revotes appear in chat in the situations you mention (on a 20+ player capacity) and these only have an impact when there is a very small gap / vote difference between the top voted maps. The thing is, you're saying that !revote here is already solving the problem I'm talking about, but it's incredibly messy even on the odd occasion when it does, and the risk of 1 spammers winning the vote is always there and does happen. I also don't think it's that easy to influence players' vote decision. People go with the map that is actually being voted the most already more than they opt for the one that's being suggested. The exception to this is when a massive amount of the server is spamming "revote!!" in chat and on mic. Again, you can say this already solves the problem, but it's hardly a real solution, and it's incredibly messy and unnecessary. This also applies to "they can rtv"... It's also worth mentioning that not every player has the confidence or will to express their preference to other players, especially over mic. They speak with with their vote and in a lot of scenarios in the current system, their expression doesn't count as much as it should/could. Yes, if there is a very popular map on an option other than 1 this is usually most likely to be selected. But I think I've highlighted enough scenarios where 1 spamming is a problem, and I am sure it's possible there is more.. if I think of them I will add them in. In my experience the more common outcome is people just being frustrated.. "rtv this shit".. "this map is awful".. "why did we vote for this?". People start to argue, the atmosphere becomes negative, some leave.. The outcome you mention does happen, but it's less common, and in my opinion not a huge reason to keep the system as it is. The second variation on the new suggested system would also create the positive scenario you're talking about. It could be argued that the negative impacts are not that limited, but even in the case that they are, why not implement a change that removes these negative impacts...limited or not? I know nothing about coding, so I don't know how difficult or labour intensive it would be to implement this change. All I can say is that is has been done by other servers so it is possible. Maybe Nomy can weigh in on this. My suggestion is also just to implement a testing phase, of say 2 weeks, just to see how it works in practise and how people respond to it in-game. Sorry if I've waffled at all.
it's not just about spamming many times when i rebel this vote comes out and i don't have time to see the maps so i just press the closest button (1) to remove the list even without adding a new blank vote you can just change between 1 and 0 like this: 1) cancel 2) map_1 . . 7)map_6 0)extend map
Cancel would have the same effect in terms of the mechanism but I think "no preference" is a better one to have as it actually gives a sense of choice to people who don't want to pick any specific map. The cancel option would likely just make people vote for a random map instead, which would reduce the likelihood of people voting for their preference getting their preference selected. --- Double Post Merged, Jun 8, 2018, Original Post Date: Jun 8, 2018 --- I'm currently screenshotting as many map votes as I can. It will take me a while to get a decent sample size but I'll post here with the results and maybe get someone who is proficient in maths to make a statistical analysis. I.e. how many 1. maps win votes, how many of them are popular maps, how many are unpopular maps, how many unpopular maps are selected as 1. in place of popular maps in other positions.. and so on.
As one of those "1 spammers", I would just switch over to whichever number becomes number one. I spam 1 because I don't mind whatever map comes next, and as stated before, when I really want a certain map I can just !revote and try to convince others to do so as well. You often hear jailbreakers spam the mic and chat with "revote 2" or whatever. I despise servers that have these long ass lists trying to convince you not to "spam vote" and then it ends up having two pages of votes, which is actually counter-intuitive if you look at it.
I'm not suggesting this. My suggestion is just that "1" be changed to no-preference, so these votes don't count and people with preferences have more of a say. As you can see on the photos in my OP the map list remains short. --- Double Post Merged, Jun 10, 2018, Original Post Date: Jun 10, 2018 --- I think if 1 was removed altogether people would just spam 2 instead. Keeping 1 as a selectable option would direct players to continue to press this rather than trying to press 1, nothing happening, then just pressing 2 instead. I agree with the second part, so far in keeping an eye on map vote results, the "1" option is disproportionately winning, and regardless of what the map options actually are.
I think its pretty fair to have the map nominated 1st to be on the top of the list and so on.. If a map is popular, chances are it will be nominated 1st. e.g. de_dust2. However it doesnt mean that option 2 and 3 wouldn't stand a chance. I have seen many times de_nuke (which i think is a shit map) being voted over de_dust2 when nuke being 3rd or 4th option. I also think because we have such voting system since day 1 of our servers, having to change it will cause a lot of confusion now, Its easy to select 2 if the map is listed 2nd in the list when really its meant to be option 7. More unhappy faces than happy faces imo.
One thing that I noticed when I first started playing this server, that I think is a bit related to this, is that sometimes maps would win with somewhat low percentages (33%, 40%) and it seems like we sometimes play maps that not that many people want to play. What used to be the system on another server I used to play was that if no map was over 50% of the vote, a run-off would start and the two highest voted maps would then have their own vote to ensure that at least half of the voters actually voted for the map in question. While I don't think this would stop the 1-spam that happens, it would at least make it more likely for another map to be voted because if enough people didn't want to play the first map, then only the two most popular maps out of the pool (including extending the current map) would be put up against each other. If, for example, 60% of the server didn't want to play the map in slot 1, but they were split to the point where they couldn't get a map to beat slot 1, then there would at least be some second vote asking for preference between the top two maps.
Alright so I reached a conclusion with the data collecting with help from @Spiraler (who deserves most of the credit). Basically found that 1 spamming does occur, but the data is limited by the sample size. There are a number of variables that influence map votes, but the main ones I can think of are: nominations/order of nominations, player influence on votes (chat+mic), position on the map vote list and map popularity. To measure these independently and draw any kind of conclusion requires a lot of data, and different types of data depending on the variable. We measured the last two variables quantitatively. I collected 101 map votes over a few weeks from demos and live play. Only the demo ones were valid in the end because the live ones I didn't collect properly. So I have 65 map votes. Essentially to break the sample down into sub-samples of significant value, you need hundreds (at least 300) and this would take too long to gather unless everyone was collecting them or there was a server addon that logged detailed map vote results. The process is to screenshot at the start of a map vote for the list of maps and again at the end with the resulting top 3 maps and Server cvar (nextmap). This takes 30 minutes per map vote in live play, and roughly 4-7 minutes per demo map vote (accounting for errors). The process would take much, much longer if data for player influence and nominations were collected. Measuring motivations for 1 spamming would require qualitative data, i.e. a survey: this might be useful to consider whether 1 spamming is mostly thoughtless (undesirable in vote outcomes) or mindful (okay). I considered making a poll thread along with this post but decided against it due to their being too many variables and it being far too subjective, the results wouldn't be of any real value unless it was a well-defined external survey. Spoiler: Results To summarise the results as they are, slot 1 won 70% of map votes and slot 2 won 20%, with the remaining 10% divided up between the other 5 slots. Spiraler did some cool (i.e. all of the) data analysis upon request and made some graphs. Here they are: The results indicate that slot 1 is selected disproportionately but not necessarily that 1 spamming is occurring. 1 spamming is when slot 1 is selected primarily or solely because of it's position. To determine whether this occurs (without data for nominations) you need to divide the sample up, so we looked at map votes where an unpopular map was in slot 1 and won, with a popular map in another slot, versus map votes of the same scenario, but where the popular map won. Popular maps are defined statistically by Gametracker. GT fluctuates slightly, so we just used one state whereby they were spy, summer, minecraftz, avalanch and lego in that order (this doesn't account for the 6th, 7th or 8th popular maps and so on). The scenario occurred 13 times, of which, the unpopular map won 5 and the popular map won 8. If Gametracker's top 5 is an indicator of map popularity and preference among players, the popular map should win nearly all of these votes, so we might assume that the occasions where the unpopular map won, 1 spamming occurred (the problem is the sample size as mentioned). We can also look at how often an unpopular map wins in a slot other than 1. Of the 65 votes, this occurred 7 times, and of these, only 2 showed an unpopular map beating a popular map (28%), this is in contrast to 38% when the unpopular map is in slot 1. Of the 33 map votes where only unpopular maps were available, 28 were won in slot 1. If 1 spamming didn't occur, we'd expect these to be more evenly distributed. 46% of popular map wins are in a slot other than 1, suggesting that players generally 1 spam much more when only unpopular maps are available, compared with when there are popular maps available. When a popular map was in slot 1, it won 100% of map votes (excluding one 'extend' with 6 total votes for maps on that vote) suggesting that a combination of the 1. slot and popularity gives the best chance of success for a map to win. The results suggest that the two dominant variables/factors in map votes are 1 spamming and popularity. Where 1 spamming has a noticeable impact is in the 20% of votes where a popular map is not in slot 1, and loses 40% of the time, in other words 8% of map votes, and in the scenarios where only unpopular maps are available, which is 50% of votes and where 85% result in slot 1 winning, in other words 42% of map votes. For the first scenario, the popular map should win much more than 60% of votes, given how statistically players favour these significantly more than unpopular maps. For the second scenario, if preference was the main factor for map selection rather than 1 spamming, we'd expect votes/wins to be distributed more evenly, suggesting players with a preference might be losing out to 1 spammers here. If we assume that 95% of map votes in the first scenario should go to the popular map, at 60% this is a 35% shortfall, so 7% of all map votes are potentially dictated by 1 spamming. In the second scenario, map wins should ideally be split more evenly amongst equally unpopular maps, so where slot 1 is winning 85% of map votes this should be closer to 20% (100(%) / 6(maps) = 16, rounded up), so 65% of map votes in this scenario, 32% of all map votes are potentially dictated by 1 spamming. We can estimate that 39% of map vote outcomes are dictated by 1 spamming. Depending on whether or not you see the above scenarios as issues, this effect could be considered negative. I won't go into too much detail on my opinion, but I personally feel that player preference should have the biggest impact on map vote outcomes. Although 1 spammers arguably are exercising their preference, those who do it thoughtlessly could easily be controlled for (see below). Regarding '2 spamming', If it does occur, it is on much less of a scale than one spamming (about half) and proportionately affects outcomes much less. 54% of 1. votes result in 70% of slot 1 victories, where 28% of 2 votes results in 20% of victories in that slot. The other numbers/slots show no evidence of spamming. Spoiler: Suggestion After thinking about this for some time and considering the data I collected, my suggestion is to make option 1. "Cancel menu". I think most 1 spammers just want to get rid of the menu and 1 is the easiest to press to do this. The 1 spammers who press 1 mindfully as a way of randomly selecting a map as someone mentioned in this thread can move to pressing 2 instead and there's nothing wrong with this. The point is that players should have a choice, and currently players have no option to close the menu. It would also encourage players to be more mindful in map votes, and potentially increase the influence of player preference on map vote outcomes (in up to 39% of map votes potentially). Players would be able to utilise !revote if they cancelled the menu due to have been actively playing, to then make a decision when there is some time to breath, for example between commands on Jailbreak or when the enemy has been killed on Clan. Players who choose to cancel and don't have time to !revote wouldn't impact the outcome against players who were able to make an informed decision. There's a chance that players might have more influence in chat+mic on the outcome if 1 spamming is removed. Another positive of this suggestion (if another map wasn't added) is that the current 6 map choices on the list are too many. In the midst of a round it's difficult to multi-task and consider with any real thought all 6 maps that are available to select, or to discuss them with other players. Assuming that the mindful 1 spammers would press 2 instead, 2 spamming might become more of a phenomenon. However, it is more than possible that the majority of 1 spamming is thoughtless so would continue, and mindful 1 spammers might include players who in the suggested system would choose to leave the vote up to the other players by not voting at all. Arguably the number of 1 spammers that move from spamming 1 to 2 would be minimal, though this hasn't been quantified at this point. Spoiler: Map votes reference The map votes I used for the study were between March and May 2018, and were selected from demos at random. I included chat about the vote when it appeared near the screenshotting points. Here they are: https://mega.nz/#F!7D5RyIQb!DB-41lMMqvSXES8Zfx9qFw Spoiler: Disclaimer I did this for fun and enjoyed the process. Whatever the outcome is fine with me. Thanks to suRpr1zE for the input encouraging me to be critical.
Is it possible to make everyone see different mapvote lists? like everyone sees a different order? or maybe the nominated maps dont start at 1 to 10, but instead 10 to 1?
Now we need data from a server that has 1 removed to see if players don't spam the first option regardless of what it is labelled.
@Steezy I would be more than willing to collect data on HeLL servers while I play if the suggestion was implemented. It would take me a couple of months to collect enough, unless there's an easier way than my method. A server plugin that logs vote results would be invaluable and give very strong data. --- Double Post Merged, Sep 11, 2018, Original Post Date: Sep 11, 2018 --- I had a similar idea but not quite the same. With this there is no issue with any kind of number spamming. I.e. if there's 30 players on the server and even if 100% spam 1, there should still be 6 different maps voted for all with 5 votes each, in other words a random map will win. I think if you remove the thoughtless 1 spammers and players who just want to close the menu with a redundant 1. option as well, the fewer remaining mindful/random-map 2 spammers should have little impact on the result compared to player preference, and if they do the result is a genuinely random map (they'll be pitched against the other 2 spammers' maps). The downside I imagine is the extra work to implement. The theory to justify it I guess is that if player preference has significantly more impact on map vote outcomes, players should generally prefer more the maps that are being played on the server, which would result in more players joining + staying on the server.
![]HeLL[ Clan](styles/quark/quark/hell_noshadow_by_amin.png)