EASHL Drop-in Redesign Proposal

This post covers issues with the current implementation of Drop-in in NHL 17. Suggestions are made in order to streamline the whole drop-in experience; and a way to implement class-based line-changes is presented to add more depth to the EASHL/Drop-in game modes.

Introduction

I have some thoughts I’d like to share about redesigning the Drop-in experience. I firmly believe the health of the EASHL, this online gaming community, and to some extent the EA NHL franchise could benefit greatly from a further improved Drop-in experience.

1. Drop-in is the first taste of the 6-on-6 online hockey experience for most new players.
2. While I guess everybody agrees that club is better, Drop-in is (or should be) easier to commit to. I’d like to see the numbers behind the scenes, but there has to be hundreds of thousands of Drop-in games played in a year’s span.
3. When no EASHL club teammates are on, a lot of players will use Drop-In to learn a new position, test some player classes, try to improve their game, and/or just relax and have fun.

Considering those points, the goal of this Drop-in redesign suggestion is to deliver:

• Quicker matchmaking
• Better, more consistent 6-on-6 experience
• Better on-ice gameplay allowing for more strategies

Current Workflow

The current Drop-in workflow is the following:

1. Launch Drop-In
2. Select a position group (Forward, Defense or Goalie)
3. Search for a lobby
4. Pick the exact position in the lobby
5. (1-2 minute wait)
6. Load class selection screen
• Pick exact class
• Pick team + jersey by each team’s captain
7. Load & start the game

That makes for a lot of steps before we are actually playing hockey. Many of those steps expose a possible point of failure.

Current Issues

When trying to play as a forward:

• Expect to get matched with other people that want to play forward only, 9 times out of 10. This, more often than not, results in 3-on-3 games if nobody quits before the game starts. While this looks like a bug, it could be a trade-off EA had to make in order to keep a 2 minute “lobby lifecycle”.
• When matched with only 5 other people in a 3-on-3 lobby, some players will understandably leave the lobby or the class-selection screen in the hopes of finding a 6-on-6 match.
• And when that happens, all players are back to square one – meaning it takes a lot of effort to get a 6-on-6 game going in Drop-in as a forward.

When trying to play as a defenseman or as a goalie, it’s actually not as bad — most lobbies will fill with 12 players. There still are some issues though:

• If your Competitive Rating (CR) is high, you will most likely be matched with poor teammates. My CR is roughly 650-700, and I constantly get matched with 350-450 CR players while the opposition mostly have >450 CR players.
• This isn’t necessarily a bad thing, but the matchmaker doesn’t usually create evenly matched teams.
• Even as a D or G, lobbies sometimes fail to fill entirely, especially if one team has a low-level-badge goalie, or if one team looks “stacked” with multiple prestige-II players (which doesn’t mean anything, to be honest). Or the game simply disconnects on load.

All in all, it can take many iterations to get a game going.

Suggested Workflow

Being a software developer, I’ve been hoping for years that the actual drop-in workflow would change to it’s simplest form which, in my opinion, would look like:

1. Launch Drop-In
2. Matchmaking preferences
3. Matchmaking
4. Load & start the game

Step 1 — Matchmaking Preferences

Here are some explanations and reflections about the functionalities:

• Allow the user to select the exact region/server they want to play on
• Display ping time and total player count to every game server
• By default, select the best server based on ping-time
• Allow the user to select multiple positions (“Pick” column above), and a preferred player class for each position
• This is in order to skip the whole class selection step. This could also be used to allow for line-changes in EASHL (more on that later)
• The estimated wait time has to be shown and has to be per position
• This will make people understand why they have to wait longer to play RW than LD or Goalie (there has to be 3x-10x the amount of people that want to play forward versus defense)
• Showing low wait times for defence might act as an incentive to learn a new position (or to play D as a forward, unfortunately)
• If a player has selected many positions, the one with the lowest queue time should be used by the matchmaker – This could mean quicker games for polyvalent players that like to play multiple positions
• Every user should have a different Competitive Rating (CR) for every position

Step 2 — Matchmaking

• The matchmaker should try to balance teams as much as possible, using the positional CR rating of every player.
• By skipping the “Pick exact position in the lobby” screen, the matchmaker would be able to assign which player goes on which team after the 12 players are matched together.
• The matchmaker should try to make sure each team has roughly the CR total. For example: Team-A LW is more skilled than Team-B LW? Then even things up by giving Team-B the stronger RW. This isn’t currently possible because players are assigned to a team before the lobby is filled up, so every player that leaves the lobby throws off any attempt at balancing.
• Also, with the CR being positionally based that would allow a better team skill approximation; being a great forward offers no security regarding the user’s skill as a goalie.
• The matchmaker should only look to match up 6-on-6, requiring at least 12 players searching for a game on a server.

Step 3 — Load & Start The Game

Skipping the lobby will obviously prevent lobby trolling. There is a very large proportion of games that never go through the lobby phase because of human behaviour. For example, the goalie decides to quit during class selection, a player just never picks sides making it uneven, etc. I could name at least 10 different scenarios here, you all know what I’m talking about.

Notice we also skipped the class & team selection screen. That’s right, there should be no team selection, just use a random home/away NHL teams. Alternatively, the league could be randomized and two random teams from the selected league could be picked.

• The random team will allow for more team diversity, and it will showcase all uniforms.
• Randomizing teams would also prevent jersey trolling, or just plain human error when both team captains pick a dark blue jersey with 3 seconds remaining before game start. We’ve all been there, and it sucks.
• Does anyone actually care about the ability to select a team? Only the captain gets to choose anyways.

Gameplay Improvement

Line Changes in EASHL

With the 3 new forward classes added, and the defense now sporting the Puck-Moving Defenseman build, there are even more strategies to be had regarding matchups. You’ve probably noticed the suggested workflow does not allow picking a class to counter the opponent’s selection.

In the NHL, coaches will often react to what the other team sends on the ice during the game. Sure, some teams will decide to dress an enforcer instead of a 4th-liner, but most of the adjustments are made during the game.

In order to give an option to counter an opponent’s game-style, there needs to be a way to use multiple classes during the game, by simply toggling a line-change.

• Toggling a line change should allow you to go to the bench and switch to a player using the predefined class linked to that line, as seen in the mockup above.
• Line changes should be called by each player, and not by the captain.
• Optional: make fatigue have a greater effect to force people to use all their player classes in 1 game. This would make more sense than keeping a single fatigue value across all lines for the user, at the cost of gameplay complexity for new players.
• Optional: allow customization of skaters names on a per class basis, if so desired. This could add some lore to games, particularly in club.
• There are icons/symbols for every player class already in the game, why not use it in-game as an indicator
• Before any face off, briefly show an icon corresponding the the player class under each player
• Alternatively, change the player selection indicator to show the actual player class at all times
• The pause screen should have a tab where every class for the 12 players on the ice is displayed, with their positional CR as well.
• Playing right defense against a skilled 748 CR Sniper? Why not use your defensive defenseman to try to shut him down.
• Getting beat down in the corner by that Enforcer defensemen? Switch to the power-forward you have on line 3.

• Allow for a single goalie change for goaltenders. This could encourage goalies having a rough start to stay in the game and try to get a win using their backup goalie.
• Goalie stats would have to be available distinctively for starters and backup.
• There could even be a backup goalie leaderboard, where only goalie changes after trailing would be taken into account.

Competitive Rating (CR)

The newly introduced Competitive Rating (CR) concept is a great addition. The CR is inspired by the Elo rating system (more info on Elo here, Elo implementations differ from game to game, and is also often called MMR or TrueSkill). Every major eSports use a rating system: CS:Go, League of Legends, DOTA II, Overwatch, etc. This proves EA is taking steps towards bringing the NHL series into the eSport arena, and this is great news!

In my opinion, there could be a lot more done to put emphasis on CR. A couple of improvements to how CR is used could really benefit the game.

• As mentioned before, having a different CR per position would be very informative. A global CR could still be calculated and showcased, by pro-rating positional CRs by games played.
• In drop-in games, the highest skilled player could be assigned Captain, giving him an exclusive, single-use, time-out ability (pause).
• Both team Captains could be presented before or after the goalie, during the game intro. This gives players something to work towards too (“Hey! I want to be captain, too!”)
• Being named a team’s Captain could unlock an achievement, once again promoting team-play as only wins increase the CR.
• There should be a clear in-game explanation of what CR is, how it is calculated, what perks come with a high CR, etc.
• There could eventually be tiers, grouping only players within a certain CR range and using those tiers to match players together in Drop-in & Club.
• I have a strong feeling this is EA’s plan, but either isn’t developed yet, or was deactivated/put on ice because the population segmentation would make matchmaking even more tedious… We need more players first!

Quitting Penalty

• CR should drop like a loss (it currently is a -30 penalty on competitive rating). CR should accurately reflect player skill, so penalizing more than a loss doesn’t make sense if there is already a penalty for quitting.
• If the quitter tries to join another Drop-In game, display “Game still in progress” when attempting to join another game, until the previous game ends
• Alternatively, the wait time could be an arbitrary number of minutes, ie. 5 -> 10 -> 15 -> 30 minutes, scaling with the user DNF %
• If the quitter tries to join a club game, display “Joining dressing room not permitted on quit — Please wait 5 minutes”.
• The 5 minute delay is to make sure league games aren’t penalized if a player was warming up in Drop-in (league teams usually warm-up together in club, so this shouldn’t really be an issue). Most leagues have a 5-10 minute rule before a forfeit can be declared.

Caveats / Pitfalls

• There wouldn’t be anywhere to show-off your player badge. This means all the hard-work EA has put into player badges, which are pretty cool, would be possibly limited to club dressing-room.
• Side note: Player badges have very little correlation with player skill. A player can be Prestige 3 LVL 24 and still be a puck-hugging-one-man-show-troll with a losing record and terrible CR.
• Playing some positions could mean a very long wait time (RW and LW specifically), but I have the firm belief the system would “auto-balance” itself with more players going for defensive
  positions after a while, because they want to play more games.
• It could be harder to implement group-matchmaking in the matchmaking algorithm.
• The quitting penalty could result in even less players looking for a game.
• The line-change system where a different class is mapped to a line might be unintuitive or too complex for new players, especially if the fatigue is shared across all lines.
• It could be impossible to form a 12-player matchup at some times of the day (early morning), or >1 year after the game is released, or whenever there are less than [2-LW + 2-C + 2-RW + 2-LD + 2-RD + 2-G] searching for a game.
• I’ve seen this happen in other games, and it isn’t necessary a deal-breaker: people change server to the highest population one, trading-in a bit of latency. This is another reason why having an East Coast, West Coast and Europe server toggle with population count is very important.
• The servers are all in different time-zone too, so if the East Coast server is empty at 7 in the morning, there might be enough players on the Europe server where it’s past noon.
• EA might want to keep the number of players online private for some reason, meaning that the Step 1 — Matchmaking Preferences screen would be less informative.

Conclusion

Hopefully the ideas listed in this suggestion will trigger some interesting discussion within the community.

The introduction of a streamlined workflow could allow for a painless/quicker experience in Drop-in. The new CR ratings, on a per position basis, allows for better match-ups. A line-change feature adds a lot of depth to the EASHL gameplay. Last but not least, the quitting penalty would responsibilize some of the chronic quitters.