wishing a bomb out of your partner

[--Q--]

The Mahjongg is a very powerful card, yet used the wrong way pretty often. The cases where you just go for a random number are way rarer than people believe. With many hands you want to call a specific card.

Example: You have 1 22333 66QQQ K A DR.
You want to call that hand and the only correct call here is King (unless you gave your partner King). You will mostly lose your hand if QQQ falls, so taking out another King protects your Queens. If Queens get hit by Aces, your King is good now, so thats not too much of a problem.

Example: You have a hand that relies on a straight from 4-9 to not get hit. You should wish 10 here so you might kill a straight from the next player that hits your straight. If he wants to hit your straight, he definetly needs a 10, so time to get rid of that. This wish is best if you have 101010 on your own because you can be sure there is no more straight to 10 out there, unless the phenix is used which is mostly awesome.

Example: You have a mediocre hand with the knowledge of trades that your team lacks highcards. You might want to wish an Ace because you will hit your partner less likely.

Example: The opponent next to you called Grand and you have phenix but no aces. You might want to call for a 6 to kill a small straight since he cant compensate that without phenix.

In many cases going for a random number is wrong anyways, so the point "lets see whose bomb we gonna hit" is entertaining and fun, but mostly not the correct play.

[f__c]

The Mahjongg is a very powerful card, yet used the wrong way pretty often. The cases where you just go for a random number are way rarer than people believe. With many hands you want to call a specific card.

...
This was an excellent post, very much eye opening for me. Your reasoning for wishes are warranted, you are protecting your call. That reason alone is sufficient enough for a blind wish, as I have acknowledged also previously.
I have however thought long and hard about the statistics that players refer to when justifying blind wishes and are not protecting a call. If finding the bomb is just one element of it, but destroying your opponent's low straight should have more merit, then the math behind it isn't as much on your side as you perhaps thought.
If you are not buying into my philosophy, that your actions, the wish phase included, are either beneficial for your team or they are not, so a 50/50 split, then we should have a look at the 67/33 argument.
Say you are missing a number from the trade, meaning you have not seen, e.g., a 3 and you wish for this card hoping to ruin some bombs or better yet, someone's low straight. The number of possible combinations of distribution of the card among the three remaining players is 15. (400,040,004,310,301,130,103,013,031,220,202,022,211,121,112). The percentage that you at all are hitting the next player is indeed 10/15 = 66% (The chance of him having that card in any quantity greater than 0). But your arguments for blind wishing isn't about just hitting a player, because for that you could just be wishing for the passed card and you'd be guaranteed to hit him. Your arguments are about ruining bombs and straights. So for that to happen, the opponent must have all 4 of that value or exactly 1. And there are only six such situations out of 15, where you are hitting your target. (two bomb cases and 4 cases with the next player only having 1 card of that value). 6/15 is 40%.
And this is all done totally ignoring the fact, that your passed card, which you refuse to wish for, may very well have formed a bomb or a straight, that you so eagerly set out to find.

So if --Q--'s arguments did not convince you, please consider, that blind wishes don't achieve what you think they are achieving, and are just adding to the randomness of the game, which you should be eliminating, not seeking, if you think you're the better player/pair.

[NeanderthalMan]

Yes, good players avoid 50/50 chance of luck elements. But 67/33 they gladly take in a game that has severe luck influence anyway.

Agreed. If you're wishing blind and there is in fact a quad bomb corresponding to a value you don't have, it's twice as likely to be with the opponents than your partner. Players are frequently calling GT and T with equivalent (67%) or lower probabilities of success. Hitting a flush is bit different because a prior play may satisfy the wish. So, in most typically situations, blind wishing should be seen as a reasonable bet on favorable odds - though there are certainly times when one should not. Just as a high probability GT/T will fail a large minority of the time, blind wishing will inevitably hit your partner on occasion. However, the net effect over time should be in your favor.

[f__c]

Yes, good players avoid 50/50 chance of luck elements. But 67/33 they gladly take in a game that has severe luck influence anyway.

Agreed. If you're wishing blind and there is in fact a quad bomb corresponding to a value you don't have, it's twice as likely to be with the opponents than your partner. Players are frequently calling GT and T with equivalent (67%) or lower probabilities of success. Hitting a flush is bit different because a prior play may satisfy the wish. So, in most typically situations, blind wishing should be seen as a reasonable bet on favorable odds - though there are certainly times when one should not. Just as a high probability GT/T will fail a large minority of the time, blind wishing will inevitably hit your partner on occasion. However, the net effect over time should be in your favor.

thanks for your input but you are agreeing to something, that has been refuted later in the thread, more specifically, in the post before yours. A brief recap: 66/33 rule works, if you know for a fact, that there is a bomb somewhere and you hope to wish it out. But you can't know if there is a bomb, so you are effectively just shooting in the dark and hoping for something good to happen. If you want to inflict real damage, you want to ruin a straight or a bomb. Splitting pairs or triplets usually has no to very insignificant impact on whether to call Tichu or whether you can defend from a Tichu call. And so, if you read my previous post, you will see that you only are doing meaningful damage 6/15 times, so 40%. If you are all about reasonable bets on favorable odds, then you want to re-think your stance on blind wishes. Yes, sometimes it works, sometimes it does nothing, and sometimes it ruins your partner's intention to call T. The passing phase should give you some information to make your wish at least an educated guess. Blind wish is the last resort. But please, play your odds however you like. My main problem is with players writing Sry and then repeat the same next hand. And then next game, and then the game after.

[NeanderthalMan]

thanks for your input but you are agreeing to something, that has been refuted later in the thread

please consider, that blind wishes don't achieve what you think they are achieving, and are just adding to the randomness of the game, which you should be eliminating, not seeking, if you think you're the better player/pair.

Your post doesn't refute anything. The blind wish premise is, as stated, if there's a bomb, it's twice as likely to be with the opponents. If there's not, that knowledge itself has value—which you overlook—just like the other types of info we recall, infer, and track while playing.

Foreclosing the chance of being bombed can allow callers to play differently and not hold Dragon for last, perhaps even setting up easier double victories by ending on a combo the partner can win. Whether successful or not, knowing a given bomb isn't possible reduces overall randomness and butt clinching when you play Dragon before Dog.

Bombs may be the single largest contributor to busted T/GTs and opponent DVs, so forcing early play or precluding the possibility is an understandable goal. This isn't to say that blind wishing is always the best or should be the main approach. However, it may be reasonable if not the most rational on any given hand.

The blind wish itself and the chosen value are not random, as in having no "specific pattern, purpose, or objective" (though the outcome is). The objective is usually to draw/rule out a hypothetical bomb, typically low quads. It's motivated by the intuition that not having 2/3/4s will often mean someone has four.

In fact, this tactic frustrates you because your partners have successfully identified this likelihood, wished accordingly, and hit you. If they're doing it as frequently as you imply, the results will regress to the mean over time and reflect the 67:33 ratio.

And this is all done totally ignoring the fact, that your passed card, which you refuse to wish for, may very well have formed a bomb or a straight, that you so eagerly set out to find.

This is a straw man. Typically, someone makes a blind wish because they know what they passed next couldn't have made a bomb and that it's likely a junk single which they'd prefer the opponent be stuck with by forcing a higher card or bomb play.

Q's examples are intriguing uses but most of your hands won't offer such perfect scenarios and the tactic is just another educated guess that will damage your partner some minority of the time. The goal of targeting a hypothetical bomb, sniping a hypothetical over-combo, or wishing for an ace, is the same, damage the opponents by targeting a possible card/combo that poses a threat and maximize your team's opportunity for points.

It seems frustration is leading you to reverse engineer an argument that blind wishing is always the wrong option, outcome notwithstanding. Having your bomb pulled out by your partner is annoying, but so is getting busted when you called T/GT. You wouldn't argue against calling Tichu with favorable odds simply because some outcomes were bad. Likewise, the blind wish shouldn't be precluded simply because the partner gets hit a minority of the time. Wishing for what you passed next or targeting hypothetical over-combo or suspected quads are all valid and rationally justifiable tactics, context dependent.

[f__c]

Your post doesn't refute anything. The blind wish premise is, as stated, if there's a bomb, it's twice as likely to be with the opponents. If there's not, that knowledge itself has value—which you overlook—just like the other types of info we recall, infer, and track while playing.

The only value I am seeing, is that you see 1 card out of 52 (56) be played. Which rules out only a quad bomb of that face value. There might be a 1 in a 1000 games when this info alone is all you need to call. Great, do it. But generally speaking, any wish for a card you don't have yourself will give you the same information. Is the value of that information adequate to risking accidentally hitting your partner, while doing no damage to the opponents? I am just not convinced.

Foreclosing the chance of being bombed can allow callers to play differently and not hold Dragon for last, perhaps even setting up easier double victories by ending on a combo the partner can win. Whether successful or not, knowing a given bomb isn't possible reduces overall randomness and butt clinching when you play Dragon before Dog.

Simple observation of what numbers players skip when playing on low singles can tell you a better story of what they might be holding. If a 3 could have been played several times, but instead was skipped, then your suspicion alert should go off. If you had blindly wished for the 3, then you would have ruled out a 3 bomb, but then you would perhaps have had no intel on 2s. (the card you actually have wished instead). The uncertainty of what will transpire in any hand will always be there, even if you are tracking every card and every suit to eliminate flush bombs as well. Therefore, a blind wish does not provide any added value compared to a wish for the passed card. It only bears extra risk of "swing and a miss". Make no mistake, only wishing always for the passed card isn't the right play either. Evaluate the outcome of the trade, and make informed decisions. With your eyes open, not blind.

This isn't to say that blind wishing is always the best or should be the main approach. However, it may be reasonable if not the most rational on any given hand.

How can something be reasonable, if not outright the most rational approach on any given hand, and at the same time not be the best or recommended? When will you want to go with anything else other than the most rational option?

The blind wish itself and the chosen value are not random, as in having no "specific pattern, purpose, or objective" (though the outcome is). The objective is usually to draw/rule out a hypothetical bomb, typically low quads. It's motivated by the intuition that not having 2/3/4s will often mean someone has four.

You not having a certain card, means that the probability of someone else having all four of them is only 3/15 (20%), of which only 2/15 (13%) are opponent's hands. (15 being the number of possible variations of distributions of that card among 3 players)
If 13% of the time is, as you say, "often", i will leave up to the readers to decide.
If your wish serves a specific purpose or objective, then it is not a blind wish. Seeing a card only to rule out 1 quad bomb out of 13 possible, is not a specific purpose, because any wished card you don't have gives you the same value. And yet, you don't often see J or Q or K wishes.

In fact, this tactic frustrates you because your partners have successfully identified this likelihood, wished accordingly, and hit you. If they're doing it as frequently as you imply, the results will regress to the mean over time and reflect the 67:33 ratio.

My partners were until recently from 99% random players of very broad ranges of ratings. Some of them i would not vouch for having successfully identified anything about the game. But if you're a blind wisher, if that's your thing, however i disagree with you, fine, be it and stand by it. Don't write sorry, only to do it again at the next opportunity.

This is a straw man. Typically, someone makes a blind wish because they know what they passed next couldn't have made a bomb and that it's likely a junk single which they'd prefer the opponent be stuck with by forcing a higher card or bomb play.

I wonder how you arrived at your conclusion, that your passed low card must be trash for your opponent.
Example: You pass a 2, don't receive it back, and so you wish 3 or higher, because you want the next player to be stuck with a useless 2? Having got it back or not, isn't wishing for a 2 helping your partner shed a loose low, e.g. a 3? Maybe your opponent only needs to get rid of the 2 to call. Maybe all your partner needs is to shed that loose 3. The chances of either are pretty even, so they cancel each other out. So wishing a 2 can be just as bad as it can be good. Wishing a 3, the next player not having it and playing 2, may upset your partner having to play 3 which he needed.

It seems frustration is leading you to reverse engineer an argument that blind wishing is always the wrong option, outcome notwithstanding.

I don't see how it is ever a good option.

Having your bomb pulled out by your partner is annoying, but so is getting busted when you called T/GT. You wouldn't argue against calling Tichu with favorable odds simply because some outcomes were bad. Likewise, the blind wish shouldn't be precluded simply because the partner gets hit a minority of the time. Wishing for what you passed next or targeting hypothetical over-combo or suspected quads are all valid and rationally justifiable tactics, context dependent.

Unless you can see other player's cards, sniffing out a bomb is 100% luck. If my partner wished for a passed card (that also could have been a bomb), or wished for the lowest single (to allow me to shed my lowest single) but missed a bomb, and that unwished bomb ruins my call, then so be it. We have played the right way and only lost to elements of the game that are beyond our control. (I almost wrote bad luck, but then you'd argue, that accidentally hitting your own constitutes bad luck too. And so I rephrased it, as unlike failing to detect a bomb on your first move, your mahjong wish is very much in your control, and so a blind bad wish is entirely your fault, not a bad luck.) If you consider wishing blind viable just to see a mystery card be played and thus cease being a bomb threat, then we just have to agree to disagree.

[NeanderthalMan]

How can something be reasonable, if not outright the most rational approach on any given hand, and at the same time not be the best or recommended?

Follow the logic of what you quoted. Essentially, blind wishing is not always best but may be in a given situation. Also, having imperfect information means you'll sometimes be deliberating between multiple reasonable options.

If 13% of the time is, as you say, "often", i will leave up to the readers to decide.

As you know, the location of 2/3/4s aren't solely the result of the deal, they're the cards most frequently passed to opponents, so when you don't get one, the odds of someone having four are significantly higher than by card distribution alone.

If your wish serves a specific purpose or objective, then it is not a blind wish. Seeing a card only to rule out 1 quad bomb out of 13 possible, is not a specific purpose...And yet, you don't often see J or Q or K wishes.

This is a bad faith argument. You've used "blind wish" for when someone wishes for a card they didn't pass next. You declared this random, 100% luck, and with no specific purpose. As demonstrated, a blind wish isn't random. No one rolls a die and targets whatever number appears; it's an informed decision based on intuitive probabilities of quads being dealt/created and the 67% chance it would be with the opponents. And if there's no bomb, the next player is still more likely to be forced to break a combo to play the wished card as a single than the partner would be.

Furthermore, there are never 13 possible quads because you're holding 14 cards and you may have passed an unrepresented card to your partner. Of those still hypothetically possible, the lower ranked quads are far more likely to result from opposing passes than higher, hence you don't see J/Q/K blind wishes to hit quads.

I wonder how you arrived at your conclusion, that your passed low card must be trash for your opponent. Example: You pass a 2, don't receive it back, and so you wish 3 or higher, because you want the next player to be stuck with a useless 2?

Your example is misleading. That's exactly the occasion someone comfortable blind wishing would wish for what they passed next. Conversely, if you passed 2 next but get one back and haven't seen 3s, then wishing for 3 is reasonable, all things considered.

Risk assessment is a major component of playing Tichu. When you make a call, you're leveraging points, i.e., placing a bet to get out first. My argument is simply that the risk associated with blind wishing (hitting your partner) is comparatively low and not marginally different from that taken on by calling with your average T/GT hand. If you're not risk adverse in those high leverage bets with similar odds of success, there's no reason to be so with a blind wishing, depending on the situation. There are merits to most of the arguments for/against on this thread but you can't make a rationally compelling argument that blind wishing should never be done because the probabilities of damaging your opponents are double that of hitting your partner.

[PeterEbert]

I read your discussion and will for sure think more about what I wish in future. I want to add some thoughts I miss in ur posts:
- The partner of the person with the Mahjong always can call tichu before the wish to 100% make sure the wish doesn't damage the hand. Of course sometimes you only can call at a later point in the game when you have information about some cards.
- I read the part about getting information out of what cards (especially singles) the opponents skip. Shouldn't that be a reason to usually not play the mahjong as first card? Sure, you need to win a trick in order to be able to play it later, but if that happens, you could maybe use the wish to more precisely destroy something in ur opponents hand

[NeanderthalMan]

The partner of the person with the Mahjong always can call tichu before the wish to 100% make sure the wish doesn't damage the hand.

I neglected to mention this but it's a good point. If I want to blind wish but can't call, I'll play mahjong and hesitate to give p a chance to call premeptively. If they do, I'll wish for what I passed next. This doesn't help when the partner doesn't have a tichu worthy hand but is sitting there with 2222.

Shouldn't that be a reason to usually not play the mahjong as first card?

I do this sometimes when possible, e.g., when on a GT where I might have Mahjong, Dog, 2. See what low singles get played and if quads can be ruled out, target something else strategic.

[Silene]

- The partner of the person with the Mahjong always can call tichu before the wish to 100% make sure the wish doesn't damage the hand. Of course sometimes you only can call at a later point in the game when you have information about some cards.

I actually said this in my earlier post. And also that I would wait for that a little when i was about to play the mahjong.

Also the card received from my partner give some counter-indication to blind-wishing. I won't wich blind if my partner gave me
- dog
- mahjong
- something very low (8 or lower will usually do it for me)
Because imo these cards tell me thy might have some plans but maybe can't call immediately.