Why guideline Tetris is broken

[Rosti_LFC]

I don't think FFA targeted garbage is that much of an issue.

It adds strategy of waiting for the target to line up with a player who is struggling so you can send the 4 lines to them to knock them out, rather than just slightly hindering a player who is doing OK. Or maybe you don't take the choice of waiting because you'd rather just get your stack down.

On the assumption that you can't be targeted by multiple players at once (which I think is important), then so long as the garbage target doesn't change too quickly, there's really not as much excuse for "I lost because bullshit luck, not because they outplayed me". If you send tons of garbage and still lose then it's usually because you didn't downstack efficiently and prioritise properly. If the targeting system is consistent and cycles in a non-random fashion then there's a lot less room for luck so long as you're actually paying attention to the other players.

If you wanted to win in 4p Tetris DS then you had to really concentrate your efforts on staying as low as possible and surviving until it was 1v1 and you could breathe a bit more. Obviously this creates a somewhat cagey and defensive game, but it's not inherently a bad one - it's just different. Nor is it necessarily dominated by luck over skill. You still can attack, it's just that your goal is to send as many lines as possible whilst consistently staying low, rather than sending as many lines as possible whilst making sure you don't get too high. If sequentially you get hit with spikes from different players as the targets rotate then that's unlucky, but increasing the duration of the targeting cycle can help reduce that and at least limit it to two players' worth of garbage in a short space of time.

OK, it does give the opportunity for you to be fucked over if the stars align against you, but it's not like anyone just plays a best of 1, and those things will work for and against you over 10 games or so. I feel what it brings to the party in strategy and just livening things up a bit is enough to make up for the fact that every now and again you'll lose a game where you were the better player. RNG isn't a bad thing if it doesn't completely change the victor in a Bo9. I'd say part of the reason that the Tetris competitive scene is unappealing to a lot of people and boring as fuck is that it really doesn't take much of a skill gap for any Bo19 to be a very predictable 10-0 whitewash in one direction. Does it really destroy the integrity of the game if some of those are 10-3 instead? Especially when any proper tournament is going to be 1v1 and so FFA games are rarely going to be anything other than fun (and maybe with a slight dash of RNG they could actually be fun instead of being a clinical black-and-white snoozefest).

[caffeine]

I'm really glad you brought up these points.

It adds strategy of waiting for the target to line up with a player who is struggling so you can send the 4 lines to them to knock them out, rather than just slightly hindering a player who is doing OK. Or maybe you don't take the choice of waiting because you'd rather just get your stack down.

"Adds strategy" is something I often hear players use when they want to defend a feature that they've already spent a lot of time learning. The thing is that there are millions of ways to "add strategy," but that doesn't automatically mean it makes for a better game. Allspins, cascade, sticky, "hurdle" clears, items, 4x4 squares... The list could go on and on for possible ways to "add strategy. In my post, I was showing how certain elements can detract from flow, despite whether they added strategy.

On the assumption that you can't be targeted by multiple players at once (which I think is important), then so long as the garbage target doesn't change too quickly, there's really not as much excuse for "I lost because bullshit luck, not because they outplayed me".

You don't have to be targeted at the same time in order to receive way more total garbage than other players. The target just needs to line up on you when your opponent happens to attack.

If you wanted to win in 4p Tetris DS then you had to really concentrate your efforts on staying as low as possible and surviving until it was 1v1 and you could breathe a bit more.

That's exactly right. The targeting system makes attacking play somewhat pointless early on, especially in clean garbage games. Part of my point was that this hurts flow by obfuscated the goal of multiplayer Tetris. And no, I don't believe this is a meaningful "strategy," but rather serves as a hoop players must jump through that only serves to confuse and frustrate.

Especially when any proper tournament is going to be 1v1 and so FFA games are rarely going to be anything other than fun (and maybe with a slight dash of RNG they could actually be fun instead of being a clinical black-and-white snoozefest).

And it was my point that this luck-element hurts flow by weakening the feedback a player receives about his performance and the in-game results. Stronger feedback would make FFA more fun. If games are too one-sided, then we need better matchmaking, or at least a decent handicap system. Tetris is a game of skill. Let's not turn it into video bingo in a misguided attempt to make it seem more exciting.

[Rosti_LFC]

Part of my point was that this hurts flow by You don't have to be targeted at the same time in order to receive way more total garbage than other players. The target just needs to line up on you when your opponent happens to attack.

Total garbage is across an entire game though, which can be quite a long period of time (especially in FFA and especially if players are closely matched) and therefore it's more likely for things to balance out a bit. In the same way that even with a pure randomiser, across 700 pieces you'll still largely get even-ish amounts. The important thing are the spikes and droughts, which in FFA is at least smoothed out a lot if you can't get hit from multiple directions in quick succession because the targets cycle slowly and are cyclical mapping.

If games are too one-sided, then we need better matchmaking

That's a great point, but for that to even be in contention you need a game which can guarantee that regardless of your skill level, there'll be enough people online whenever you want to play for close games to actually be possible. There aren't really any Tetris games out there which I say do that across the whole range from noobs to the likes of the top players on here.

Part of my point was that this hurts flow by obfuscated the goal of multiplayer Tetris. And no, I don't believe this is a meaningful "strategy," but rather serves as a hoop players must jump through that only serves to confuse and frustrate.

I completely disagree with your idea that having to play differently at different points in the game is confusing, frustrating, or that it fundamentally makes the game worse. So what if it weakens feedback? Part of the learning curve of a game is the learning process of figuring out which strategies work and which ones don't, and I would say for a lot of the games the learning curve is the most interesting and entertaining part.

Your video bingo comment is massively strawman - FFA targeted garbage still very much allows skill to influence results. Even for Tetris DS in the higher ends where the matchmaking was a bit dodgy, the targeting system wasn't ideal, and there was very much an unwritten rule of "lets make sure we all target the highest rated player first", you'd still see players like NOAH and Samaf winning the vast majority of games because they were better.

RNG is bad if it's introduced in a way that has a huge influence in the result. But if it's introduced in a more subtle sense then it goes a long way to making games more interesting. How you can adapt and handle adverse random events is a skill in the game itself, and so long as they're not completely overpowered to the extent where they play a big part in who wins, it helps spice things up. Guideline Tetris is inherently very boring because there's very little depth, very little strategy (combos, t-spins and all clears, oh my!) and there are no random elements to keep it interesting - even the randomiser is incredibly predictable.

It's shitty game design to take a game that isn't particularly exciting and to just chuck RNG in there to make it so, and I'd never advocate that, but at the same time I think that the Tetris community has arrived where it has now because of this overbearing "a player who plays better should win every single game every time" aspect that has stripped a lot of the potential for experimentation and for *fun* out of the gamemodes. Items are bad. Interesting garbage is bad. FFA has to send-to-all because anything else is bad.

What matters is that the impact of RNG is relatively small, and that the frequency of RNG events is high enough to average out across a sensible timeframe, not that RNG doesn't exist at all. So long as RNG elements don't swing things so far that players lose more than a small minority of games where they were the better player and objectively should have won, it becomes something that compliments the gameplay and adds flavour to it. RNG plays a pretty big part in games like LoL, DotA and CS:GO and they don't seem to have any issue holding tournaments with millions of dollars at stake - why should we be so scared or arrogant to suggest that it could only be bad Tetris?

[caffeine]

Total garbage is across an entire game though, which can be quite a long period of time (especially in FFA and especially if players are closely matched) and therefore it's more likely for things to balance out a bit.

Obviously it depends on the number of lines and the game length, but this is besides the point. Receiving more garbage by way of blind luck will put you at a disadvantage. It feels silly to even have to say that.

I completely disagree with your idea that having to play differently at different points in the game is confusing, frustrating, or that it fundamentally makes the game worse. So what if it weakens feedback? Part of the learning curve of a game is the learning process of figuring out which strategies work and which ones don't, and I would say for a lot of the games the learning curve is the most interesting and entertaining part.

Because it's not just that you have to "play differently at different points of the game." It's the lacking of feedback for playing well. You can play the best you've ever played, or you could literally do nothing for several seconds on end. At the beginning of a FFA, your performance has little impact on the result of the game. "So what if it weakens feedback?" Well, my whole point is that feedback improves flow, and that makes the game better. It's the subject of this thread, actually.

Your video bingo comment is massively strawman - FFA targeted garbage still very much allows skill to influence results.

Your argument was how a luck element could be good, and that's exactly what I was attacking. Please forgive my use of hyperbole, but that doesn't make it a straw man. Obviously skill will always affect the players ability to win. This is in a similar vain as before when you attempted to sugarcoat the impact of total lines sent. What is important is if this weakens feedback and hurts flow unnecessary.

I'm not arguing for a game perfectly devoid of luck. I'm saying it's preferable to avoid it, except when absolutely necessary. I believe most skill games at a tournament level aim for that. I believe last time we had this argument, the topic of poker came up. It's a good example of a skill game with some necessary luck elements. My position would be like saying that the dealer chip should cycle to each player in a fair manner (as the rules currently dictate) opposed to introducing an unnecessary luck element by randomly selecting the dealer after every hand.

[Rosti_LFC]

I believe last time we had this argument, the topic of poker came up. It's a good example of a skill game with some necessary luck elements. My position would be like saying that the dealer chip should cycle to each player in a fair manner (as the rules currently dictate) opposed to introducing an unnecessary luck element by randomly selecting the dealer after every hand.

To be honest, I don't think poker is a good example of anything when the discussion is on Tetris - you're completely comparing apples and oranges in terms of how the game works, the speed the game progresses, and what aspects of the game actually provide the excitement. Heck, in Poker it is the luck element and the act of overcoming it (or correctly exploiting it) that is exactly why the game is exciting and fun to play.

And just because a luck element is unnecessary doesn't mean it's bad. Randomly selecting the dealer after each hand is a dumb idea not merely because it adds luck, but because it does so without increasing the excitement or entertainment value of the game.

DotA and LoL are games of skill with some unnecessary luck elements and they prove to be immensely popular games. Things like crits and bashes don't need to be in the game for the games to function, and they can directly alter the outcome of individual kills and events - sometimes you get lucky and win an engagement you probably should have lost. That said, it's up to the players to take the probabilities and likelihoods of those outcomes into account when deciding what they're going to do, and it's also very rare that RNG is the deciding factor in the outcome of a whole game.

If luck events destroy flow, then how are CS:GO, DotA and LoL three of the most popular PC games right now in terms of number of players and total hours played - all three of those games have RNG embedded in some of the key aspects of the game mechanics. I can s understand that subtle but high-impact luck will break feedback, which breaks flow, but I don't think that luck alone inherently makes it non-obvious to the player what's going on.

Coming from Tetris and SC2 I originally really didn't like the idea of RNG in a game but if it's executed well then in practice it really doesn't provide much opportunity for lesser skilled players to beat higher skilled ones. And as long as it's properly presented it also doesn't necessarily confuse the player as to what actually happened and break the feedback aspect either. And as a plus it can help provide elements of excitement in the exact same way that dice rolls can make tabletop games exciting, or that card reveals are exciting in poker.

[caffeine]

I haven't played any of those games, so I can't say much about them. In fact, the argument has become really abstract, and I think it would be helpful to refocus on exactly what we're discussing.

Correct me if I'm wrong, but here's what I think the break down is:

• Targeting adds a luck element.

Rosti_LFC's argument:

• The luck allows weaker players more success in the short run (but not long run). This makes the game more entertaining.

caffeine's argument:

• The luck weakens feedback, which hurts flow. This makes the game less enjoyable.

Are we on the same page?

[Rosti_LFC]

I haven't played any of those games, so I can't say much about them. In fact, the argument has become really abstract, and I think it would be helpful to refocus on exactly what we're discussing.

Correct me if I'm wrong, but here's what I think the break down is:

• Targeting adds a luck element.

Rosti_LFC's argument:

• The luck allows weaker players more success in the short run (but not long run). This makes the game more entertaining.

caffeine's argument:

• The luck weakens feedback, which hurts flow. This makes the game less enjoyable.

Are we on the same page?

Not really. My argument isn't to add luck for the sake of adding luck because I believe luck is good. My argument is that features which can make the game more entertaining and interesting shouldn't be ruled out purely on the basis that they add luck, because if it's done properly that shouldn't inherently be an issue. I'm not saying that adding luck is definitely good, just that it doesn't have to be bad.

I also wouldn't say that the main benefit of luck itself would be that it allows weaker players more success in the short run (though this may happen), more that it just adds an element of unpredictability. Two "identical" games on the face of it have slightly different outcomes because RNG can potentially disrupt that. And so long as it's well implemented it will:

- Be obvious to the player that the RNG exists and isn't just some mythical behind-the-scenes bullshit deciding who wins. This means that feedback remains.
- Not have such a massive impact that it can overcome a large skillgap and is overpowered (for example if there was a 1% chance that occasionally you just send 20 lines for a Tetris instead of 4). This means that it doesn't feel cheap.
- Happen often enough so that players would expect to receive it for and against their favour in roughly equal measures across the course of a sensible length of time (say 10 games). This means that it doesn't effect meaningful outcomes of matches.

This means that it just becomes an aspect for players to factor into their decision-making and account for in the way they play, rather than something that detracts negatively from the flow and the enjoyment of the game.

To choose a game example I presume you can relate to, in poker you can be in a position where, from your perspective, you statistically have a 95% chance of winning the hand. It's a strong position, but there's always that potential for you to lose - and sometimes you will. But astute players will recognise that when they went for a big pot in that situation and lost, it's because they were unlucky, not because they made a bad choice going for a big pot. And that actually they made the right call and if they were to face that situation again it'd most likely go their way the next time. With poker usually being a game of more than one hand, the fact that they went the wrong way on that particular one also wouldn't typically be something you'd expect would cost them the entire game either.

Sorry it took me a while to properly figure out and express what I was really trying to say. I've been pretty exhausted the last 48 hours or so.


Also, going back slightly, I don't think targeting is an amazing feature and should totally be in all FFA, but at the same time I do think that if it's well-executed it has certain advantages in FFA games over send-to-all, in particular if you want to try and have games with like 8 people at once. I mostly just think features should be designed on their own advantages and disadvantages rather than just "Do they add unnecessary luck, y/n?".

The best example I can give of a feature I think has unfairly been shoved out of all consideration in competitive play would be items. An item mechanic which was well worked, where the items were balanced, and where they complimented existing play rather than being a core factor in who wins, is something I feel could potentially add a huge amount of entertainment and challenge to a game at both a casual and hardcore level. The fact that most games don't have balanced items (Star item was OP in Tetris DS, Hard Block and Death Block are OP in TGM, etc) doesn't mean that it's not possible to do it right.

[DormantAmbition]

I don't think FFA targeted garbage is that much of an issue.

I believe empirical evidence is appropriate here - whether an aspect of a game is entertaining or not is dependent on the opinions of the players, and not of one person. In this case, caffeine is correct - both experienced players and novice players dislike the targeting system. In other words, no one likes to lose due to an external factor such as luck.

[caffeine]

To summarize, and I hope I'm understanding you correctly, you're saying that we should evaluate whether a feature is good by whether it makes the game more entertaining, not whether it creates luck. You're also saying that an element of unpredictability can be good, when it can be factored into the player's game plan.

I definitely agree with that. Similar to the poker example, there's an aspect of Tetris decision-making that gets washed away by multiple previews, bag randomizer, and hold. In old school Tetris games, it becomes more important to build surfaces that work for any piece. Of course that's not always possible, so the player must face a decision. "This surface will be good, except if I get x-piece. Is it worth the risk?" The player must evaluate the risk and reward due to the chance element. Your decision might not have worked in your favor this time, but that doesn't mean it's not the right move. Over the long run, it could succeed more than it fails. This is an challenging and interesting aspect of old school Tetris.

So I definitely agree that there is a place for probability and chance in games. I believe that the points at where we must disagree are:

• The merits of targeting for its fun value, irrespective of the luck element it introduces.
• Whether this luck element adds a meaningful decision-making aspect of where players must weigh probability, risks, and rewards.
• Whether the luck element weakens feedback and flow.
• Whether the decision-making aspect is worth lessening feedback and vice versa.

In my opinion, targeting doesn't add much if any meaningful decision-making in regards to probability. From my experience, it's more like "there's a chance I could get screwed out of nowhere, so I shouldn't take any risks and stay as low as possible until more players get eliminated." I do believe this weakens feedback and a feeling of progress, which are important to flow.

The best thing it's got going for it is the game theory of choosing who you attack. The second-best player would ideally want everyone to attack the best player. Knowing that, the third best player might want to attack his closest rival, the second-best, instead. Although while interesting in theory, in practice many players simply ignore it in order to avoid slowing down. And even if they don't, it really only serves as a convoluted handicap system at best. The decision-making is centered around punishing players for being better at the actual game. I, for one, would like to see players try to win by improving their Tetris play, not by getting a perk for being second-best.

[nostalgianinja]

Some very interesting arguments here about different ways of garbage and how different types of attack work.

while I may still be very novice in my attempts to improve on Tetris (and I use this term loosely) One of my main qualms of Competitive Multiplayer is that the Hurryup system does not determine how good a person is, or how fast, lucky and/or how one plays.  The Hurryup system could be scrapped and there would be an infinite allowance to play (it doesn't likely happen, but it has happened on one occasion that I've played against one of my friends for +/-10 minutes on Tetris Axis.) but that's not what is wanted:  What is wanted is how to get a balanced way of playing against someone while proving difficulty and power.

I think that once I become a regular with playing Tetris Online (which there isn't a stable and populated game yet that I own) I may ask for help improving

Blitz, based on your issues maybe you should try playing Puyo Puyo.

it's been 24 years but no one's found a spammable formula for that yet! lol

Puyo Puyo hasn't changed for over 21 years, S2.  Many people still rely on the Tsuu rule because it works.  If Tetris were to incorporate its hurryup system it might change the dynamic as necessary.  Sure, the gimmick online modes in 15th/7/20th might be helpful to change the game a little, but most of the game is centered on the Tsuu rule, it's Puyo Puyo's standard of Guideline there (unless Fever standards also apply as in Puyo 7).  Unlike Tetris, Puyo Puyo only stays with Sega, whereas Tetris is broken into and changed every iteration by every developer who has their hands on it.

It's a shame though that Guideline Standards break the game to a point where it can't be fixed.  Does this mean that Guideline itself has to be changed?

[Rosti_LFC]

caffeine: glad we're on a similar page now at least.

I would repeat that I don't think that targeted garbage is perfect. I think that it requires careful implementation to not just screw over players with the possibility of getting hit by 16 lines from multiple players in one go. I think allowing players to freely choose who they attack is inbalanced (because the optimal strategy is just to knock out players higher than you) whilst at the same time with zero choice (you are randomly allocated a player to attack and do so until they die) then it doesn't really offer much.

For the decision-making element, I feel it's largely dependent on how fast rotation is (so how long you're likely to have to wait to attack one player) and also increases when you're using an MMR/Elo system that gives benefit to a higher placed position rather than merely fighting to come first,

That said, I don't think send-to-all is perfect either, especially given that most games these days have garbage cancelling which potentially makes it a bad thing to be the fastest player to open the attack - if you've used up your opening clear first, then you've got nothing left in your stack to potentially cancel out a wave of six other players sending garbage at once. And for lobbies of more than 3-4 players you have to do some kind of scaling so the amount of garbage doesn't get ridiculous early on with so many people, which just doesn't feel elegant to me as a mechanic.

To be honest, I think to really settle the argument there has to be a reasonable amount of testing of different ways of working the system, rather than just the handful of current embodiments (which carry the rest of the other flaws of the game platform). And it's not impossible to create a hybrid system of the two where lines get distributed to certain players, but not everyone (so sending 4 lines might send 1 line to 4 players in an 8 player game, for example).