first clue in 2 player

[Silene]

Overall - if you give a red clue and then the other player immediately plays it (in some cases it does not make sense to clue back as the first turn - trying to make you discard chop) and you intend to use green clue after that, the odds of the player drawing another green or multi is about 40%. This means that the green clue would probably not be possible. You have to use 1. and later a 2 clue anyway.

Drawing green or multi chance is actually 2/7 in my books which is ~28%. Anyway that's not a problem at all. Because IF they don't clue me than it means I don't have 1s (at least none that they want to clue now), so cluing 1 instead of green is perfectly fine.

So it's like giving 1 clue and 2 clue but then you also used red and green for the same effect. Only multi 3 is saved.

The effect is, they know whether and how to clue your 1s or not. So it improves their first action. Some examples are shown by wreckage. Worst case would even be if you have r1 or g1 on chop and no other 1. If they clue that in response after a 1-clue on their r1g1, that would even lose 2 clue-token AND add a bomb. One for clueing it to begin with and one for not getting a clue-token for discarding it when it's bombed instead.

In addition the red-mark on the 2 has the positive effect that a later 2-clue is definitely for the red 2 no matter whether another 2 was drawn or not. And that's even a GAIN that might pay off IF a 2 is drawn in the next 2 cards (after r1 and g1 is played), because it means you get a tag on another 2 for free, which is a nice potential side-effect. And also you can't be screwed by m2-draw which would make it hard to ever get to r2 in your approach. Yes, the chance is low, but still.

[Silene]

And another thing is:
If I happen to have 1s in my hand, then the color-clues actually help their decision.
If I happen to NOT have any 1s, then color clues loses a clue-token compared to your approach, yes. BUT if that is the case, it also often means we have a bad start with very few playables. In such games, you often hit the 8-clue-counter early, what means you have to waste clues anyway. So a lot of the bad effect of losing one clue-token would be cut off at that point.

[Stroom]

Drawing green or multi chance is actually 2/7 in my books which is ~28%. Anyway that's not a problem at all. Because IF they don't clue me than it means I don't have 1s (at least none that they want to clue now), so cluing 1 instead of green is perfectly fine.

From the initial post. 2 green/multi in hand. I don't really know what I have in my hand. So worst case, I don't have green/multis. 50 cards in deck, 12 in hand. 38 cards left in the deck. 10 of each color. 2 in hand. 18/38 = 47%
Basically a 50-50 whether your next clue can be green or not. And even if I do have all greens/multis in hand, that's 12/38 = 32%. Still quite high for me.

The effect is, they know whether and how to clue your 1s or not. So it improves their first action. Some examples are shown by wreckage. Worst case would even be if you have r1 or g1 on chop and no other 1. If they clue that in response after a 1-clue on their r1g1, that would even lose 2 clue-token AND add a bomb. One for clueing it to begin with and one for not getting a clue-token for discarding it when it's bombed instead.

OK, Should I go through all these examples and show that the red-green are actually a worse option? Often it either wastes an extra clue or adds confusion. Or you risk at 32%-47% of wasting the clues anyway.

One bomb is OK. If you clue red to a player who also sees you have a red 1 between multi 1 and yellow 1, they might have to either clue you yellow or 1 and then not play their own red... Or just accept one bomb from the start of the game in order to keep the pace up and lose less clues doing so.

TBH with 1-2 you can actually sometimes save a bomb if the order in which the other player plays his 1s does not match with you. You can see that the next 1 the player is about to play is already played, you can give that color clue and the player would either choose the next 1 or discard it. And that is not a play clue even if it marks some other card.

In addition the red-mark on the 2 has the positive effect that a later 2-clue is definitely for the red 2 no matter whether another 2 was drawn or not. And that's even a GAIN that might pay off IF a 2 is drawn in the next 2 cards (after r1 and g1 is played), because it means you get a tag on another 2 for free, which is a nice potential side-effect. And also you can't be screwed by m2-draw which would make it hard to ever get to r2 in your approach. Yes, the chance is low, but still.

You still don't get how the start-of-game 1-2 clue works. If you use 1 and then 2, the other player can play the leftmost 2 after all the 1s. And in this specific case, you make 2 moves to mark 3 cards. With red-green you will also need a 2 later to mark that. So you are still spending 3 clues for 3 cards.

Cards that come up later... well, you can play around that as you go. But as I explained already, the risk of a future card messing up your play is higher for color clues rather than number clues.

M2 draw is 5%. Smaller risk. AND as I said, 1-2 in the beginning only counts for the initial hand. If you give the clues straight after each other.

In addition, I also mentioned that you can start with 1-red clues instead. In which case the second red clue marks the red next to the 1 red, meaning that that card is also playable. Which is effectively the same as red-green-2 but you use one less clue.

[Stroom]

And another thing is:
If I happen to have 1s in my hand, then the color-clues actually help their decision.
If I happen to NOT have any 1s, then color clues loses a clue-token compared to your approach, yes. BUT if that is the case, it also often means we have a bad start with very few playables. In such games, you often hit the 8-clue-counter early, what means you have to waste clues anyway. So a lot of the bad effect of losing one clue-token would be cut off at that point.

Actually, it isn't often the case. If you mark the red and have red 1 between different 1s, it could get harder for the other player to mark your 1s. So he either gives YOU 1 clue and discards his cards or has to use multiple color clues... which overall spends clues rather than making the game go faster at the expense of 1 bomb.

In 2p you usually don't hit 8-clue-counter. Every discard is a possible double discard waiting to happens so after almost every discard you are basically forced to spend one clue to let the other player clue if you have something important in your chop.

[Silene]

One bomb is OK. I

A bomb is -1 clue token too. Because usually when that card you are bombing finds it's way into discard pile it would generate you +1 clue token. If you bomb it, it doesn't. So it's the bomb plus 1 clue that you're losing when you boms.

You still don't get how the start-of-game 1-2 clue works.

I don't know where you get that from.

Cards that come up later... well, you can play around that as you go. But as I explained already, the risk of a future card messing up your play is higher for color clues rather than number clues.

Yes you can play around a newly drawn 2 in order to get your r2 (unless it's m2). But I don't want to play around it. I want to mark it so we have it tagged. But when I clue 2, it will get played because it is leftmost 2. But when i have red cards and a 2-clue marks a red AND the new card as 2, then it doesn't mean "play left" but it means "play r2". It pains me having to "play around" a card that i would prefer to be tagged.

While many of the examples we gave have only small propability I do believe they add up to overall better outcome.

[Silene]

From the initial post. 2 green/multi in hand. I don't really know what I have in my hand. So worst case, I don't have green/multis. 50 cards in deck, 12 in hand. 38 cards left in the deck. 10 of each color. 2 in hand. 18/38 = 47%
Basically a 50-50 whether your next clue can be green or not. And even if I do have all greens/multis in hand, that's 12/38 = 32%. Still quite high for me.

Ah, I didn't realize it's without black. In that case one of 54 cards that I can't see will turn up, 18 of which are green (because I see 2 of the greens/multis). So chance is actually 33%. Still not 40

I don't understand your calculation where you get to 47%...

[Stroom]

A bomb is not necessarily -1 clue. Sometimes the fact that you have to play around a card could waste 2 or more clues rather than just bombing it.

1 clue then 2 clue in the beginning generally means play left 2 too. This is quite basic after you have played thousands of 2p games. Otherwise you start with 2 and then 1 clues. If you have played a lot, you will see how playing like that helps kickstart the early game. At the same time starting with color clues means each clue is only for one play clue and then you have to get more specific.

The risks I showed were high enough that starting the game with 2 color clues just does not seem worth it. Especially with that kind of hand. And as I already explained, with this hand, there is no need to play around the new 2 if you use 1 clue and then 2 clue. No matter what.

And I also mentioned that you can use 1 clue and then red clue in the beginning. At 10% less risk than red-green. So... Overall, my way mitigates the risk more. And wastes less clues - for all the examples that were given except the last one (there, it was almost the same but in case of red-green, it would probably end with both players having a marked multi 3 in hand.)

[Stroom]

From the initial post. 2 green/multi in hand. I don't really know what I have in my hand. So worst case, I don't have green/multis. 50 cards in deck, 12 in hand. 38 cards left in the deck. 10 of each color. 2 in hand. 18/38 = 47%
Basically a 50-50 whether your next clue can be green or not. And even if I do have all greens/multis in hand, that's 12/38 = 32%. Still quite high for me.

Ah, I didn't realize it's without black. In that case one of 54 cards that I can't see will turn up, 18 of which are green (because I see 2 of the greens/multis). So chance is actually 33%. Still not 40

I don't understand your calculation where you get to 47%...

Ahh I miscalculated indeed. It's 60 cards total not 50. worst case 18/48 = 37.5% and best case 12/48 = 25%
Still higher than playing Russian roulette so I'll pass with these odds.

[Silene]

A bomb is not necessarily -1 clue. Sometimes the fact that you have to play around a card could waste 2 or more clues rather than just bombing it.

A bomb instead of discarding the card is always exactly -1 clue for that card going into discard pile. Of course there are cases where it's worth causing a bomb for other gain or because preventing the bomb cost more than 1 clue. But in general it's -1. You say that's ok - fine. I think so too. But if I can - like in the example in first post - avoid it, I'd still rather spend 1 more clue for the initial setup with colors to guarantee not getting that bomb and also some other gains as mentioned before. If I can't do that, I'll also take the potential bomb as you do.

On a sidenote I don't agree to your "always clue 2s first"-approach because that's slowing down play (but you call it "kickstarting" the game) and potentially wasting clues when the other player has to wait. But on the other hand I do chop-2-saves in 2p game so those don't mean play left.

So overall we just play different convention and decide to "waste" our clues in a different way . I think we can leave it with that. We won't find an answer for what convention is better through discussing it here.

To summarize (I think we can all be happy with this common ground):
When you play an "always clue 2s first before you clue 1s if the left 2 doesn't match your 1s"-style, then apparently in the example of original post, it is better to clue 1 and not color (I can't back that as I don't play that convention but let's take Stroom's word for that).
When you play "clue 1s first and chop-2-saves on non-unique-2s are allowed in general"-convention, then in my opinion (and some other's who posted here), it is better to start with color-clue in the original example.
So which clue is better depends on the set of conventions you play.

[Wreckage]

The calculations don't matter if you start with the red clue . There is nothing you can draw to mess up a clue on the green 1.

The sample of hands I submitted are biased toward giving color first. After Stroom's first post, I just went and took these from his last 10 games. On the 4 other hands I didn't post, a 1 clue was better because there were no 1s at all or there was no conflict. I believe it's a close call which way to go, but I like red clue better. On a couple of these I think it’s about equal.

But look at hand #3 for example. Isn’t the second player going to clue 1s before they play? So if both green 1s get marked, the BEST that can happen is you spend a clue to fix it.

On hand #5…
You clue 1s
they play G1
You clue 2
they play R1
You discard? or maybe clue something?
Then what? They clue yellow? then wait for R1 to be discarded, and then clue 1 to get the Y1 played, 1/3 chance you drew a yellow or multi, yikes! What a mess!! Maybe just let the bomb happen is best.
or
I clue red
they clue 1s
I clue green
We both start playing. All is clear and they know to discard R1.

On hand #6, If the first clue is 1s by both players, I believe this results in a bomb.

Note; If there is a bomb, you do realize you have lost a clue token, right?