Empty clues have no logic

[smallman]

I've noticed this is reaching epidemic levels. The clues are valuable, so giving an empty clue is pretty much NEVER ok.

I have seen them used a lot, by people that are forced to clue on 8 clues or a full hand. The clues are too valuable to waste, people will know you are forced to clue, so should not assume you are doing a finesse (if we can fix this epidemic). Find the best clue you can give that looks the least like a finesse, usually a 5, or other number.

Finesse should only be assumed in these situations, if you can see they had the option to do a clearer non finesse clue. Lets keep this game logical, and not accept the mindless sheep type conventions like the empty clue.

[1Steve0]

I totally agree. If you are forced to Clue, it is not a finesse if it is kept simple non-playable.

[UmpireNate]

The point of conventions is to remove ambiguity. Have you ever given a 4 card finesse? I have, and it is awesome. While rare I agree, you remove that possibility with your suggestion. You can certainly "create" your own convention and state that in your table settings when creating the table, and people agree to play that way, but just because you "want" to play a certain way doesn't jive. Also, if you are in midgame with 8 tokens, the "value" of not wasting a token is gone, it's just as frustrating playing when you can't discard (b/c 8 tokens) and nothing to clue as it is when you want to clue and have no tokens.

But back to my original point, you remove all "great clues" (beyond basic finesse/reverse) by saying "ignore every clue given when 8 tokens".

One game, I was clued 5 in mid hand and partner to my left has 2/4 on their most left, so i played off left and sure enough, the 1, then they played 2, i played 2nd left (3), they played their second left (4), then i played the 5. Admittedly it is rare, but your idea removes all possibility of these kinds of clues which are "memorable" situations, the baseball equivalent of the game winning grand slam.

Also, negative clues do give information using good-touch, if I have a colored card in hand I know is a 3 or 4 or 5, you can eliminate one of those with a neg-clue, or you can neg-color telling me my card is not MC, so there is info to be learned.

finally, neg clues can avoid double discard with same chop. If player before me was just saved (5 on chop for example) and then discards, my partner couldn't save both of those cards without giving a play clue, so I will stall. Also stall after a bomb (I Learned this the hard way the other day, partner made a mistake and bombed m2 and I discarded and my chop was m2)

[GunPetal]

The beauty of hanabi is that everyone can come to the same logical conclusion for the best use of clues, based on one simple principal:

Clues are valuable, make sure they get something played, or saved in danger unless you are forced to give a clue.

There is absolutely no logic to wasting valuable clues on doing empty clues. The occasions that missing out on a good layered finesse due to being forced to give a clue would be EXTREMELY rare, especially since you likely can still do it with color. Comparing this to how much information is almost totally wasted with an empty clue, makes this logical arguement pretty obvious.

If someone wants to play with this bad artificial convention of empty clues when forced to clue, then they should state this when hosting. Otherwise we should be able to assume best logic should apply.

[Romain672]

This post is axed about situations with 8 clues.
And yet the 'best logic' is absolutely atrocious.

The number one weakness of bga's convention is the games were multiple turns are played with 8 clues in bank.
I think that because it's rare, player are fine losing those game 'from randomness', and so don't want to change that.

But when you are at 8 clues, there is an high chance you didn't drew many playable early on, and so the game is already hard. And you make the game even harder by sticking to usual conventions which work well when having 1~6 clues, and so lose more great cards (like non visible 2 and 3) which make the game even harder to win.


I often advise on the forum to at least quadruple check when you need to blind play 2 or 3 cards, and when it's 3 to think a lot more about which alternatives you could see for the clue, and if you see alternatives meanings, how risky it is to not play, and which options available you have to try to see if you should play your slot 1 and most likely bomb, or if you should 'play it safe' and do something else.

Often the first time at 8 clues, it can be ok to give an empty clue to minimise the risks and make sure everyone know there is nothing good to clue on that whole board. (and make everyone see we got an 8 clue situation)
The second time you goes to 8 clues, if you clue a card which was just drawn, assuming it is playable still seem good: you want immediate tempo, and not waste more clues.
Now if it's not the case (you clue a card in chop or something else and which is valuable like a 2 or a 3), then everyone logically should get that this card wasn't playable two turns ago and still isn't playable. And everyone logically should not blind play their finesse position (since it wasn't playable two turns ago***), and should just assume a save.
This require everyone to be in sync, that's why it's so hard. If one player out of four don't play like that, that lead to a bomb which can be pretty bad since the game got few bluffs.

*** except if players in purpose didn't clue a card in finesse position to keep alive the bluffs potential which is good in bga's convention. But overall, if you was allowing to save more cards in those cases, that wouldn't be needed anymore since the important cards would be saved anyway.


So that would benefict greatly to have a change. But it's mostly optimisation because adding a complex rule for games already hard doesn't seem worth it overall. I think you prefer to lose game where it goes multiple time to 8 clues, and instead got a more clean games in others cases. That start to be annoying when you reach high elo, because you need to optimise more and more to gain elo, and some rare cases start to have a bigger weight.

But back to my original point, you remove all "great clues" (beyond basic finesse/reverse) by saying "ignore every clue given when 8 tokens".

You shouldn't care about effeciency when you are at 8 clues. You do a finesse and gain one extra clue, that you lose instantly two turns later because you are at 8 clues.

[UmpireNate]

This whole argument is moot. There is a reason this post is even here, bc the vast majority of players play by and agree a “random” clue still has meaning and they react accordingly.

If you want to make your own conventions, have at it, as I said just make a note in your table settings “all clues with 8 tokens should be ignored” and then it’s an agreed upon convention (if ppl bother to read table settings which unfortunately is rare)

But to say “I don’t like it so everyone should abide by my feelings” is not a solution.

And empty clues are not without information, I gave several examples where they can help and are even necessary, and the multi card finesse I said was rare, but ambiguity is the scourge of Hanabi

Just state your desired convention, repeat in chat before first card is played if necessary, and then play on with that agreed upon rule, not a big deal

[Romain672]

I think your post was destined to me.
And yeah, that's why I goes to hgroup and stopped bga, and a long time later stopped Hanabi.

But like I try to said in my complicated last post: everything look a logical consequence of wanting an easy time. And it doesn't look it worth it to change if your goal is to gain elo (until you reach very high value).
And changing it is pretty hard, especially since it require everyone in sync and isn't safe since it create bomb.

But what is done actually is like the layered finesse: you got a conflict between logic (second time at 8 clues = most cards in board are globally known not playable) and simplicity (i don't know how many clues are left and just play normally).
And those cases are problematic since changing it lead to lots of bombs.

And empty clues are not without information, I gave several examples where they can help and are even necessary, and the multi card finesse I said was rare, but ambiguity is the scourge of Hanabi

Those small information gain while useful are much much weaker than saving 2s and 3s, which is really important when having 8 clues because you usually didn't got time to save them early on because the board didn't got much plays, and you discard random cards because you don't choose which player discard
(there was before a rule about 7 clues spot with hgroup because of it, but it was deleted because people were seeing spots at 8 clues, but not those at 7 clues)
[note in bga, you can still at 7 clues waste a clue in purpose, like reclueing a known playable or do an empty clue to manage which player should discard, I remember doing it, but you need to take the clue which seem the less confusing, and you need strong reasons to do it, mostly either seeing the whole board trash (which increase chance for your hand to be very useful), or you got strong sign before your hand is valuable (per example, others let you do most clues)].

Mostly what I disliked the most is saying that it's logical to play normally at 8 clues (without more informations). And something to note is that the 'best logic' is not the same for everyone, which causes this kind of post.


I never said it should change, I gave more arguments for changing because they are less obvious to see/more interesting to talk about.

If you want my own opinion after writing all that, I think the cost of changing look too high. Layered finesse can be done regularly, while 8 clues spots are imo too rare.
And outside of those 8 clues spots, empty clues while rare (but from smallman, they look too commun now), still seem useful mostly to resync the whole team without a risk of misscomunication.

[UmpireNate]

“and you discard random cards because you don't choose which player discard“

You already went to it, but ^ this is inaccurate and is quite often a misplayed strategy. What I mean is, long before there are 8 clues, paying attention to discard management is huge.

IMO, the trash bluff is the most powerful clue in the game ; bc it’s like any bluff/finesse that gains a play but also manages discard, it’s even more powerful in avalanche when it’s a way to mark that pesky m1 that’s 3-4 turns from discard that’s preventing color clues.

The whole skip means play is a fallacy, I don’t ever skip and take the finesse card, but if my left has garbage on chop, I will “give their clue” rather than discarding blind (again, basic chop save or middle hand clue)


Anyways, back to my point, the premise of the OP is to essentially create a new convention; simply write that in the table notes and remind at start of game and have at it

[smallman]

Nate stop repeating the same stuff, you are flat out wrong. Doing what is most logical is not a new convention, its just the best way to play. Doing terrible empty clues is the new awful convention that has been adopted by all the people too lazy to use best logic like the game is designed for, and should be what is stated up front.

You are saying when you have a full hand and down to 1 clue, you should still do an empty clue. This is totally insane, you cant afford to just waste clues like this, and there is no reason to, if everyone applies logic, you get way better results. I played over 10k games, and I am 100% sure about this.

[Zigwardo]

I agree with smallman. Never waste clues.