Successful = viewers want to watch it and don’t regret watching it.
Developers have most likely seen Tournament viewership has STAGNATED | Town Center ep. 32, ft. @T-West from three days ago. I just skimmed some key points and read the comments. I’m bringing attention to some of these comments, which show the playerbase’s views on these issues, and proposing solutions which may suggest paths for improving the game.
NachoAoE comments that he had a higher peak of 5.5k viewers during TaToH vs Sitaux on his Spanish-language channel while the finals didn’t get close to 4.5k. He suggests that “with many matchups being useless (last Group-Stage games) and knowing how it will end, people are maybe not that prepared to stop doing something else and just go back to confirm that the results they expected happened indeed.” He mentions a higher viewer peak in his anti-meta tournament with 1K USD prizepool and thinks there should be more such tournaments or events like the Battle Royale (that T90 won, over TheViper etc.) and not just standard tournaments.
Many others also mention “predictable results”.
- Hera’s dominance — it’s just boring to watch. I used to organize watch parties with friends for finals, with big screens and sound systems, but there’s no point anymore as we know who will win.
About tournament viewership, I think we should address the elephant in the room: Hera’s dominance.
It’s not his fault, he’s just currently better.
I love Hera but I legitimately think his dominance is bad for the game at this point.
Even Viper wasnt this dominant in his prime.
Its definitely not his fault but nobody else is close to his level and its not as much fun to watch the tournaments when you know the Globetrotters are going to smoke the Generals.
And this is basically why I suggested a handicap system for the next Hidden Cup:
https://old.reddit.com/r/aoe2/comments/1km7jil/lets_idly_discuss_a_possible_hidden_cup_that_we/
If there were some kind of invisible handicap system (the available one for lobby games is visible), then allocating handicaps might work something like this:
All participants in the tournament get 10 handicap points to assign. They take turns assigning one point to another player, with the total handicap points received public knowledge, until everyone has assigned 10 handicap points. Example: with 16 players (160 total points), Hera receives 30 handicap points, DauT receives 5, and the weakest player to qualify receives 0.
If the handicap system is not invisible during play, it gets more complicated, because it might be necessary to partially hide the point totals during the assignment process, or randomize them somewhat, so that players can’t just identify opponents based on their known handicap.
One intelligent commenter responded, “An issue you don’t address is which player would ever play a handicapped tournament for prize money? I doubt you’re going to get Hera to play for 50k if he’s handicapped so that the fact he’s better than everyone doesn’t actually help him.” I think Hera would enjoy this, based on the games he’s done with his community where all players are named GL.Hera so no one knows which one is really him.
The same commenter said,
Also, which tournament organiser wants to organise something where the best players (who pull the biggest audience) are less likely reach the final. I know Hidden Cup is hidden, but when it started you always knew one of the finalists would be Viper.
But we didn’t know that. T90’s original excitement for the first Hidden Cup was that he hoped it would encourage players like Miguel to perform better if they didn’t know they were facing TheViper. He hoped they would beat TheViper and we would still be guessing that TheViper was still in the tournament when he was already defeated.
After a few Hidden Cups, then we started to think it would just be TheViper wins. But it was still fun, especially with the mindgames with player colors: I think it was maybe in the finals of one tournament that a player chose Yellow for the first time and went Elite War Elephants.
The point is that we can restore the fun of the format by changing the rules and the incentives (like larger prizes for successfully tricking opponents and the audience). Even the controversy of the most recent Hidden Cup, where the players were informed of which side of the bracket every player was on, so that they could find practice partners that they wouldn’t encounter before the finals: the tournament doesn’t have to compromise in this way. If players are too afraid to practice because they’ll be revealing strategies, then they can just not practice and perform as they will!
As Nacho said, we can have normal tournaments and unusual tournaments. Many tournaments can lack handicaps, and we expect Hera to win: I will discuss that later. For this specific format of “trying to disguise and guess player identity”, we form the objective of some kind of invisible handicap system that won’t reveal the size of the handicap, and therefore provide too much information about the identity of the player.
Also: how to make practice and match scheduling easier.
How to set a hidden handicap: each player sets their own handicap. Option to hide it during lobby creation, similar to hidden civs. But hidden civs are revealed at game start, while hidden handicap would be obscured even after game finishes. We need a way to verify that tournament participants chose the correct handicap, without revealing each player’s handicap to the opponent: we do this by specifying an admin’s identify in some form, encrypting the handicap information in the file, and providing the admin with the facility to decrypt it. If the game needs this data to play a recorded game, then it’s just “encrypted” in the file format, but we could make it complicated enough (like with salts, etc. if necessary) for an audience to be confident that players didn’t access this data to guess their opponent.
The form of the handicap
Three big classes of mistakes that worse-performing players make: having idle units, either villagers or military; having banked resources; and making poor strategic choices like not countering a unit type.
We honestly cannot force a player to be bad in any of these areas with a handicap, except maybe with an APM limiter. TheViper used a third-party tool for the LimitedViper series, and had problems like “to patrol is two actions (clicking patrol and then the destination), both useless without the other”. An in-game limiter of eAPM would be better: you could change to a control group, change the camera view using the minimap, and then attack-move or patrol to a location, all using just a single effective action, instead of 5+ keyboard and mouse events.
LimitedViper was able to defeat players with much higher eAPM, using about 15 game eAPM, showing APM is not everything. A limiter would be a handicap. But I think it would be better if a player limits their own APM in order to imitate a slower player, without necessarily having a big effect on their chance to win a game.
Dark Age performance can be very similar between players, and so any penalty might make it easy for an audience to guess the handicap unless CaptureAge numbers like total resources collected were hidden (like how the last Hidden Cup hid player APM). Player performance variance is higher later on. So I think the handicap could be that economic and military units just perform worse the more of that type of unit there is, scaled to the population limit, or minimum 75.
So, with a high handicap: 20 villagers work normally. But 100 villagers work like 90 normal villagers, and 150 villagers work like 130 normal villagers (not 90/100*150 = 135). The function is basically defined in terms of the marginal output for one more villager, which decreases but never goes below 0: we integrate the desired marginal output to get the total output, divide by villagers to get average efficiency, and from there get the required workrate reduction. (Note that farm stockpile workrate should be subject to this reduction as well.)
For military units it’s kind of hard: two trebuchets firing at each other, if one of them attacks slower or does less damage, it could be pretty obvious: instead of a treb dying in 3 hits (201-150 = 51 damage per hit), it might die in 4 (49 damage per hit). Another example that could be obvious: Mangonel with 50 HP does 40+12+1x5 to another mangonel, maximum 57. With a significant reduction, might not one-shot; or with a significant increase, might one-shot even uphill.
So I think we take relative handicap between two players and use it to calculate increased damage taken for the stronger player, based on total population. We also make this affect monk conversion times for both players, by affecting the chance to convert on each tick (not the minimum or maximum conversion intervals). Then we also take the military population of each player with a handicap and use it to slightly decrease the damage and attack rate of non-siege units and buildings.
Obfuscation: if all players have a non-zero handicap, we basically subtract the lowest handicap from all players at the game start, to minimize the effect of the handicap. At regular intervals, we could vary a small random component in the strength of handicap, but this mostly just guards against a viewer counting frames to determine attack rate handicap and might not be worth it.
So: a fight between scouts in Dark Age would not be affected by handicap. Total population is low enough that there is no increase in damage taken, and military population is low enough that there’s no change in attack rate or damage inflicted. If the pop limit is 75 or lower, we might go from no damage taken handicap at 25 total pop to full handicap at 65 pop (which is still possible with 25 pop limit through conversions), while with 200 pop limit, we go from no handicap at 25 total pop to full handicap at 170 pop. The specific economic and military handicaps would use different numbers for their respective sub-population numbers.
In general, we can tune the economic handicap to be stronger than the military handicap: partly because the military handicap is easier to detect, and partly to encourage more combat.
Other tournament ideas
Idea: cooperative play, but pro player can control only a limited number of units. The lower-rated player could be one from a team of several, different for each game. Limit could be determined by a player resource from techs, so could be changed with map, but standard could be limit equal to age number, from 1 to 4. A certain method is used to claim a unit, which stays under pro player’s control until the unit is lost, with deletion not allowed. Observers could distinguish these claimed units, but opponent cannot. If the unit is a monk, units it converts to go lower-rated player, not the pro player’s control.
“Mere familiarity”: getting new viewers to watch one tournament that features pro players makes them more likely to be interested in other, more standard tournaments.
Dev team would have to discuss the implementation requirements to make this work and the value of such a tournament, post is too long to get community feedback about this specifically.
Tournament with ‘secret mission cards’? “Cannot click up to Feudal before time X.” “Cannot make an archery range in Feudal.” “Must make 10 fishing ships in Dark Age if water has fish.” Players rate the difficulty of a mission, assigned missions based on handicap points.
Comment:
“T90 altering Hidden Cup scheduling was really harmful. The best tournament of the scene by miles is essentially being held hostage by it’s creator.” ?? https://www.youtube.com/watch?v=Kin-KUfYB7o&lc=UgzXyxzl760b1SjbnOp4AaABAg “It’s such a good format, T90 and Dave are great together. It’s the only tournament with the cache to bring people who aren’t glued to the scene.”
Hidden Cup: “running the event was very difficult.” https://youtu.be/Rs0axHbCCsk?t=689
One possible difficulty: dealing with anonymous accounts for different players, who are not supposed to know who the others are? Dev team could help here, allowing for anonymous lobbies with an invite link, in which standard account names are hidden and players cannot be identified through normal means, whatever those are. Allows players to play with their normal game account. Discuss with Hidden Cup admins.
Scheduling sets, especially when live sets mean all participants must avoid streaming: maybe just avoid live sets? Might require better tools for syncing casters streaming a recorded game, like preventing anyone from skipping ahead. Discuss with Hidden Cup admins, casters and players who stream to identify problems related to scheduling etc. Altering the game for tournament-specific features, unlikely to be used elsewhere: can be worth it. Note: high-ping servers can disguise identify of opponent, may be preferable to lowest-ping.
Improving standard tournaments with uneven finals
“And then it was 4-0, 4-0, 4-0” https://youtu.be/Rs0axHbCCsk?t=423 Maybe solution: more interviews with losers? Show that they weren’t just giving up, because now they have to talk about their loss. Incentivized to praise opponent, suggesting overall level of play was higher than normal, instead of worse than normal from the losing player. Surprised by strategies, etc. Not just the loss, but why the loss.
Hearing someone talk about their own victories is boring. Close set, more losses, more to talk about for winner. But loser can still talk if the set was not close. What might they have done to win? From winner, what might their opponent have done?
Aka, “instead of expecting or forcing Hera to play worse, other players should get better.” There is opportunity for improvement. Examples: five years ago, almost no one used high-health fishing ships to block fire ship attacks. Even now: almost no one walls on amphibious terrain to block enemy ship movement or protect fishing ships from attack. No one walls the gaps in rings of trees, which is basically a half-built wall. Keeping units grouped up in situations like walking past a TC or dodging BBC fire, instead of splitting up to avoid damage. Sun Tzu said, “Do not repeat the tactics which have gained you one victory, but let your methods be regulated by the infinite variety of circumstances.” Playing “meta” is not meta. To quote TheViper and T90, “it depends.”
Tournament: rewards players who innovate and play better, but doesn’t always happen.
Audience vote, “which player will perform above their seeding?” interview when or if they lose.
If loser doesn’t want to talk about games, that’s sort of like not wanting to improve. Should encourage talking about losses.
In typical sports, all teams have some fans, and they don’t really need more exposure by talking about their loss to the main audience. In AoE2, many players have few fans and would benefit significantly if they were not just forgotten about when defeated in a tournament, even if a set was one-sided. It can even just be hearing someone’s voice for the first time (like hearing Villese speak in an interview during the first Red Bull Wololo at a castle).
Interviewing the loser of a set is up to the tournament organizers and casters. I’m just pointing out that it could help with audience interest, even if Hera wins again.