No replies to this topic

[Alphaeus]

The Problem: Unequal Opportunity to Play Each Map

I'm the process of collecting information on the maps/modes made available for voting when I play. My intent is to collect data on 100 map/mode voting sessions to ensure I have a sizeable sample. Currently my sample size of voting sessions is 38, and preliminary analysis indicates that the map/mode voting system prioritizes some maps over others. Here are some examples:

• In 38 voting sessions, Alpine Peaks and Viridian Bog have each been an option 4 times, Mining Collective has been an option 3 times, and Canyon Network has been an option 1 time.
  
• By contrast, Frozen City has been an option 23 times, River City has been an option 21 times, and Forest Colony and Polar Highlands have each been an option 17 times.

The Data

Here's the number of times, out of 38 voting sessions, that each map has been an option, so far:

Alpine Peaks: 4
Canyon Network: 1
Crimson Strait: 9
Caustic Valley: 9
Forest Colony: 17
Frozen City: 23
Grim Plexus: 9
HPG Manifold: 6
Mining Collective: 3
Polar Highlands: 17
River City: 21
Rubellite Oasis: 16
Tourmaline Desert: 7
Terra Therma: 6
Viridian Bog: 4

Note that Frozen City and River City are an option more than twenty times as frequently as Canyon Network (a map I happen to enjoy very much, whereas I'm bored as hell with FC and RC), and the fact that players tend to vote for cooler maps means that they're selected more frequently--especially Frozen City and Polar Highlands--simply because they appear more frequently.

This is interesting. Here's the number of times a map was actually picked, out of the number of times it was an option:

Alpine Peaks: 2/4
Canyon Network: 1/1
Crimson Strait: 3/9
Caustic Valley: 0/9
Forest Colony: 3/17
Frozen City: 3/23
Grim Plexus: 3/9
HPG Manifold: 5/6
Mining Collective: 2/3
Polar Highlands: 5/17
River City: 6/21
Rubellite Oasis: 2/16
Tourmaline Desert: 1/7
Terra Therma: 1/6
Viridian Bog: 0/4

Note that the maps that received 0 votes appeared fewer than 10 times out of 38 voting sessions, and that some of the least frequently-seen maps were selected much more often (Canyon Network, HPG Manifold, Alpine Peaks). It's also worth pointing out that according to the data I've collected thus far (and I'll post the entire set when it's done), Frozen City and Polar Highlands appeared on the same voting session 11 times, and of those 11 co-appearances one of these two maps was selected only 2 times. In other words, by appearing together they tend to cancel out the "I'm voting for the cold map" effect because the cold-map voters look to be further divided into LRM boat (Polar Highlands) voters and high-heat energy or ballistics build (any cold map, but maybe not Polar Highlands if we can help it because I hate getting LRMed to death) voters.

Why All This Matters
PGI is spending a lot of person-hours on developing new maps, but to many players it's obvious that we already have 15 playable options. Forum comments suggest that player dissatisfaction tends to result from boxed-in thinking:

• "My 'mech runs too hot to play on a hot map." THEN REBUILD YOUR 'MECH. In the Equipment menu on the Loadout screen there are these things called "heat sinks"...
  
• "I'm bored with this map." You have access to it through the Testing Grounds function; scout it out and think about how to play it differently. Brute force never won over good strategy... or have you never read The Iliad... ?
  
• "This map doesn't fit my play style." Then learn a new play style. ADAPT. A-frikkin'-DAPT.

The net effect is that PGI is constantly facing demand for new maps from players who won't adapt to what's already available and realize its full potential (which, based on my experienced, is far from having been fully realized). The voting system incentivizes this to a large extent. Here are some changes that could be made to prompt players to continue to evolve their play styles and loadouts:

• If possible, create a voting system that offers each map an equal number of times before restarting the offerings.
  
• Every so many voting sessions, make the system offer a group of "hot" or "cold" maps, with one non-"cold" oddball included with a group of 3 "cold" maps.
  
• The "cold" map voters tend to pile their votes onto a single "cold" map with a group of 3 "hot" maps, so create an options system that avoids this.
  
• Consider the effect that nostalgia might be playing on voting. Since older maps (e.g. Canyon Network, Alpine Peaks, HPG Manifold, Mining Collective, Viridian Bog, Caustic Valley, Crimson Strait, Tourmaline Desert) are available only infrequently, the scarcest of these are probably getting a disproportionately high percentage of votes because we don't see them as often--AND WE MISS THEM. (Yes, even Caustic Valley, which, by the way, my fellow players, is actually COOLER than Tourmaline Desert and Rubellite Oasis.) Basically, one of them can show up and all normal voter thinking is suspended by the presence of a rare option.
  
• Offer more events that incentivize play on seldom-selected maps (I'm assuming you have records of how often each map gets selected, PGI), similar to the event that launched when Rubellite Oasis was released. (That's a great map, too, by the way. I love it.)
  
• Incentivize the rethinking of loadouts. (The skill trees went a ways in doing this; I wonder what it would look like to offer an event that gave rewards for so many matches without overheating, or for matches played with a 'mech containing X number or more of heat sinks.)
  
• Run a 'Blaze of Glory' event in which, for a week or so, the system offers no cold maps and offers hot maps much more frequently. Do this quarterly to shake up the loadouts and remind players that they can't pack eleven ER PPCs and no heat sinks. (Sorry, TheB33f--you're an inspiration to us all; but some players, who are not you, may have gotten the wrong idea from that video...)
  
• Develop the stats pages so that we can track Wins/Losses not only by map but by map/mode combo. I might lose on Caustic Valley in Domination, but I might win quite a bit in Skirmish. It'd be good to know this because, as a player, I might be avoiding Caustic Valley because I (mistakenly) attribute my lack of success on that map to the map when the problem could be the game mode.

Conclusion

As I mentioned, my survey is still under way and I'll post the data as soon as it's been collected and I have time to offer an analysis. (Game mode numbers show other interesting trends; but I'll save that for later.)

At present the voting system tends to incentivize boxed-in thinking about game play and 'mech loadout. As players continue to demand new environments that fit their already-established habits, they defer the cost of the game to PGI. To a point, this is a good thing because it creates demand for your developers' labor; but if it gets out of control it won't be sustainable. On the other hand, if needy players adapted their loadouts and worked to learn new play styles they would use more C-Bills, XP, and MC, taking on some of the cost of the game themselves. (Funny how relevant this is to LRM boating versus sharing armor--some of us pay for things; others of us stand off at a financial distance but still gripe about needing more features.)

I like that MWO is a free-to-play title, and I spend some cash on it every month or so out of gratitude to PGI's developers for making and maintaining a good game. (Though I wish you hadn't nerfed my Timbies and a lot of the weapons I tend to use, and then buffed the Shadowhawk, the clunkiest dumb-bucket in the game, which was already overpowered. But I get it: MW5.) As a player on a limited budget, for me this is a delicate balance. But a large part of the joy in this crunchy game is all of the data available to track and aid my development as a player and I think that data is saying that the voting system needs to be less game-able for boneheads who want the world to bend to their overheated LRM boating whims.