Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Critiquing a Real Double-deck Pinochle Hand
#1
I want to discuss Hand #7 of the default hand history currently pre-loaded in the Hand Animator.  In case I replace it one day, I'll offer the notation below.  For the time being, you won't need to copy/paste it.  Just load the Hand Animator, click on Submit, and change the Hand Control value to 7.

Now I realize there are some organic and inorganic players in this game, but I don't know which ones they are.
As I Step through each "frame" of the hand, I see multiple "mistakes".
Before I offer my analysis, I'll just ask the question.
What could have been done better in this hand?  Please support your assertions.

Code:
% Format "PPN 1.0"
% Site "worldofcardgames.com"
% Date "2016.02.09"
% ID "14550684069590"
% Players "N E S W"
% tableid "53938614-1671-4854-a53d-4a99b6128f49"

[GameScores "218:380"]
[Deal "E:CTTQJJDAAATKJSTTKQQHATKQ CATKDTTKJJSAATKJJHAAKJJJ CATKKQQJJDQQSAKQJHTTKKQJ CAAKQDATKKQQJSATKQJHATQQ"]
[Auction "S:60 Pass Pass Pass"]
[Contract "S 60 C"]
[Melds "8:CDJSKQHKQ 8:CKDKSKHK 29:CATKKQQJDQSKQHKQ 30:CAKQDAKKQQJSAKQHAQ"]
[Play "S"]
AS JS KS JS S2
JC AC JC KC W2
AS QS AS JS W2
AH QH KH JH W2
AD JD KD QD W2
AC JC TC JC W2
QC TC AC QC E2
AS QS KS QS E2
AH QH TH KH E3
AH KH QH TH E3
JD QD KD TD N2
AD JD KC JD S2
TC KC QC JH S2
AC QD TC JH S2
KS TS TS KS W4
QD KD TD QC S2
KH QH AH JH N2
TS JS KC QS S2
TH KD AD TS S4
TH TD AD TD S6
[MeldScores "37:38"]
[PlayScores "26:24"]
[Result "SAVED"]
[HandScores "63:62"]
It's unbelievable how much you don't know about the game you've been playing all your life. -- Mickey Mantle
Reply
#2
I'm not 100% sure, but I believe that I was North, my human partner, who shall remain unnamed, was South, and the bots were East and West.

Before going further, I will say that I have an excuse for any poor actions, in that I was really tired that night (also I'm just not that good lol). Wink

That being said, I ran through the replay and only noticed one problem. At "play03", East plays the Ace of Spades on their partner's Spades trick. This signals to p that they (probably) have the last remaining Ace of Spades. So when West plays a low card to exit the lead, at "play07", it might have been wiser to exit with a Spade, rather than a Club (which was trump). Provided that no one trumped the trick, control would have gone over to their partner.

So I'm curious to know what else went wrong.
Play Pinochle at World of Card Games!
Reply
#3
South made a fantastic opening bid. Their maximum bid was 57 and rounding up 60 was a push. It cut off meld bids and intimidated the bots and won the bid.
However, at 2 aces and 8 trump that is just enough horsepower to pull 20 and save. That is not strong enough to jump bid to 60. Worked this time but a good percentage of time would get set with that power, max bid, and push to 60.

On other hand, West had all the power and meld needed to jump bid if they were leading this hand (had South dealt) with 5 aces and 7 trump, and their max bid was 75. W should have bid 65 over S 60, and had S bid again at 70 overbid with a 75. W should have won the bid and called Diamonds trump.

Analyzing the play that did take place, Trump caller should exit ("punt") with a Q of trump to force out a counter should partner take it. In this case W took it with an A anyway so no harm done. [Edit: forgot that opponent left (W) declared Aces Around. South J of trump was correct lead for West Aces Around. Good punt.]

W led with an A Spades. Spades had already been played. Led Aces should be on fewest played suits and distributed among suits to lessen chance of getting cut before you get all your aces out. Also shortest suits with fewest aces played. So the A lead distribution for W should have been AH, AD, AS.

With that last lead, p correctly gave an A leadback (either has the other A or is cutting) and W would then exit with a KS to p.

W should not have led A trump. Had two Aces with backers. There are advanced scenarios to lead A of opponents trump to try to disrupt being cut badly but this is not it. In addition, p gave A leadback and may have been out of the suit and that lead may have taken the only trump card they had. So it was doubly wrong lead.

W did luckily get to p with the trump punt with no damage done, but odds are not good with what was done as explained above.

E led a good A Spades. There's no reason whatsoever to lead a good ace. The longer you hold on to it, the more chance that someone who might be cutting it runs out of trump. It's good in end game. Keep good aces.

The punt of E should be in suit with most cards played to give p a chance to cut. The lead was in least cards played (Diamonds) which is the lead if opponent left has led and p has not led, in other words, maximizing putting your partner in the lead. In this case as expected the player that hasn't led yet gets the lead instead.

N came back with good AD. Not the play to make. Also has good AH. The play to make is KS. Opponents had indicated they were good in Spades so you force their high card out and try to give a chance to cut it or make E cut a K instead of one of your good Aces.

That's probably enough for this hand. A nice exercise, I know my bots in trouble giving marya all these pointers. Smile
Reply
#4
(03-21-2016, 08:00 PM)mickmackusa Wrote:  For the time being, you won't need to copy/paste it.  Just load the Hand Animator, click on Submit, and change the Hand Control value to 7.

Mick, BTW - this is cool, thx! - Marya
Play Pinochle at World of Card Games!
Reply
#5
(03-22-2016, 12:33 AM)rdwrites Wrote:  On other hand, West had all the power and meld needed to jump bid if they were leading this hand (had South dealt) with 5 aces and 7 trump, and their max bid was 75. W should have bid 65 over S 60, and had S bid again at 70 overbid with a 75. W should have won the bid and called Diamonds trump.

FWIW, this game was played Feb 9. Since then I've made some changes to the bot code based on user feedback about their bidding. I just tested for this case with the current code, and the West bot now responds by bidding 65. If South bids 70, the bot code passes, however. Maybe they are still too cautious.

(03-22-2016, 12:33 AM)rdwrites Wrote:  E led a good A Spades. There's no reason whatsoever to lead a good ace. The longer you hold on to it, the more chance that someone who might be cutting it runs out of trump. It's good in end game. Keep good aces.

What is a "good Ace"? Is this term something that belongs in the glossary?
Play Pinochle at World of Card Games!
Reply
#6
A "good Ace" would be a safe Ace, versus an Ace that is under threat of being caught by another Ace or beaten by a trump.
This assessment of an Ace requires some thoughtfulness regarding:
  • Length of suit in your hand.
  • How many tricks have been played in the suit.
  • The ranks/information in the tricks played. e.g. did anyone signal? or are there any indications of another player becoming void in the suit?
  • Seat position and team objectives.
  • Revealed meld cards, in some cases.
Basically you just need to gauge risk of loss.  You do yourself a disservice when you lead an Ace with questionable probabilities of winning a trick.

Generally, the first seat after the Declarer (LHO, Left Hand Opponent, 1st Defensive Player, whatever) is desperately seeking to win a trick and become Trick Leader.  The LHO not getting in the lead means that the Declarer's Partner (Dummy, 2nd Offensive Player, whatever) will be able to cash (lead, table, throw, play) all of his/her Aces while they are still "good" (low risk).  The advantage to getting your Aces in early is safety.  When that safety is reduced, it is often a good strategy to conserve your Aces until conditions are more favorable.

More favorable conditions might mean:
  1. Trump has been mostly or completely run out (used up, tabled, thrown,  played, etc.) and the suit has been "set up" to be low risk again.
  2. Late in the Play phase, other players are leading their weaker ranks, resulting in tricks with higher point values due to the "heading" rule (all players must beat the highest card if possible).
There are more concepts that advanced players employ to maximize their Play phase results.  Some of which have been posted in other threads -- it would be worth digging through the Offense & Defense subforums (I haven't done so before posting this).  Mindful trump and Ace management (card counting and selecting) are crucial to playing the best possible Pinochle.

Just a final note, Sandbagging Aces has its place.  Search the forum for "sandbagging" if you want to know more.  There are good and bad situations for sandbagging.  Some people I have played with, do it purely to confuse their opposition.  Smart tactics need sound justifications.  Pinochle bots must be smart before they dare to become random.

Ask questions if you have some.

p.s. I purposely used lots of synonyms in this post to benefit our visitors, not because Marya doesn't understand.
It's unbelievable how much you don't know about the game you've been playing all your life. -- Mickey Mantle
Reply
#7
(03-22-2016, 09:50 PM)mickmackusa Wrote:  p.s. I purposely used lots of synonyms in this post to benefit our visitors, not because Marya doesn't understand.

A quick comment: Isn't there somewhere that these jargon terms could be spelled out? I think the glossary helps, but it's not enough. Maybe more example phrases could be added there. Your post has a very good explanation of numerous jargon terms, but it will get buried and become hard to find.
Play Pinochle at World of Card Games!
Reply
#8
Good Aces are Aces in a players hand with all other Aces in the suit having been played. Until one player remains with Ace(s) in a suit, it is a race to see who can get their Aces led first, thus the importance of getting partner into the lead before one or both opponents. A lot of strategy I talk about have to do with that.

If you have three Aces of Hearts let's say, and Hearts isn't trump, if someone else leads an Ace of Hearts before you do, your three Aces are now "good". No one else has an Ace of Hearts to lead. You now use these good Aces to get into the lead, and do what is good for your team. Your partner may be cutting Diamonds, and everytime a Heart is led you get into the lead and lead a counter (if possible) Diamond for your partner to cut and take a trick. You may rarely still have an Ace and get into lead near end of hand and take last trick. It is a weapon.

That's great that your bot will now bid 65 on that. That would be about what average player would do. I am guessing you are adding 20 to figure out how high to bid which is what average to below average players do, not having experience to estimate how many points hand is likely to win.

There are a couple of problems with that. One is that rule says 20 must be pulled (where rule is used, such as in my game), but says nothing about whether hand can do it or not. Average player says I need 20, takes a look at hand and estimates whether hand is good enough to do that, and then bids meld + 20.

The ones who can estimate how many points team is likely to pull are the ones who can keep bidding higher without getting set. I've seen on here a system based on estimating tricks pulled at 2.5 points per trick. I do not know that system, it may be better than mine but I don't know the criteria for identifying tricks that may be won. (Nor have looked into it, I have my own system.) Using that system the hand with 5 Aces and 7 trump and distribiution of cards, short suits, etc., an estimate would arrive at higher than 20, say 24 to 26, and thus now 75 is highest bid you can make.

I've mentioned before my system I developed for my bot and then decided to publish with my game rules to help beginners by taking all guess work out, it is (Aces + Trump)*2 where Ace of Trump count twice. For that hand that is 24 and that is what I or my bot adds to determine highest bid. Quite frankly it seems a little too simple but it has been road tested since 1982 when I was writing my game on a TRS-80 and we tested the heck out of it to see how close it was to points pulled. Given that points are pulled by two players one of whose hand you don't see you would think there is no way it can be accurate but it wins bids and doesn't get set unless opponents get a nasty cross cut going or something, or have a 19 pounder sitting behind you, and other wonderful scenarios. Smile
Reply
#9
(03-22-2016, 09:50 PM)mickmackusa Wrote:  Basically you just need to gauge risk of loss.  You do yourself a disservice when you lead an Ace with questionable probabilities of winning a trick.
...
The LHO not getting in the lead means that the Declarer's Partner ... will be able to cash ... all of his/her Aces while they are still "good" (low risk).
 
Okay, but rdwrites said "E led a good A Spades. There's no reason whatsoever to lead a good ace. The longer you hold on to it, the more chance that someone who might be cutting it runs out of trump. It's good in end game. Keep good aces."

It seems like you are saying "Play a 'good Ace' when it is very safe" whereas rdwrites is saying "Keep your 'good Aces' until the end of the game."

Why should Declarer's Partner lead their good Aces, but defense should not?  I'm confused.
Play Pinochle at World of Card Games!
Reply
#10
I'll let mick respond on his part but it is unlikely, I said rarely in above post, that you are able to keep good aces till end of game. Any time that suit is led you will be taking it and in lead. You use that to control the game by leading suits your partner is cutting, or bleeding trump by leading jacks and queens for the opponents to cut and sometimes have to beat each other, forcing out high trump, etc.

You don't lead good aces, but you take the trick if that suit is led.
That applies to both teams, no difference.
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)