Test your Suit Combination Play

[eyhung]

Fred has stated in the past that thinking about suit combinations is a good way to improve one's game. Here is a good suit combination I hadn't seen before that came up in a club game Friday evening.

 

[hv=d=s&v=n&n=sat8hj2dakqj5cj54&s=sk9hak986d73cak86]133|200|Scoring: MP[/hv]

 

The bidding went, uncontested:

1♥ - 2♦

3♣ - 3♠

3NT - 4NT

6NT

 

Lead is the ♦2. How would you play the heart suit in isolation for 4 tricks? Does that change how you would play the heart suit on this hand?

[qwery_hi]

Best I can come up with is low to the Jack, and finesse the 9 the next time for the suit in isolation.

[rogerclee]

Good combination, I learned something new from this.

[qwery_hi]

Best I can come up with is low to the Jack, and finesse the 9 the next time for the suit in isolation.

Scratch that - I forgot how to count.

 

I'd go with finessing the Jack and if it loses, playing to drop the 10 the next time.

If the Jack wins, we should repeat the finesse

[realnumpty]

You may be interested in some software that I recently discovered. It is called Scansuit and the link is:

http://www.scanbridge.net/Fichiers_HTM/tex...sentationAn.htm

 

Due to the limitations imposed in the demonstration version I was unable to input your problem suit as given and had to change the ♥6 to the ♥3 which I don't think changes things too much.

 

The solution for 4 tricks in the suit appears to be low to the ace followed by running the jack if not covered which is apparently a 72.6% chance.

 

I hope that this is of interest.

[Feegle]

'Suitplay' agrees with 'Scansuit'

[Finch]

This is a more complex hand than just the heart suit in isolation.

If diamonds play for 5 tricks (which is worth checking fairly early), then you only need one more trick out of both hearts and clubs.

 

I don't know what the right line in the heart suit in isolation is, but if it involves a losing the first trick, and then taking a finesse, it's wrong on this hand(!)

 

I'll trust you that ace of hearts, run the jack is correct (which wasn't immediately obvious to me at all, I probably would have led towards the jack instead).

 

So diamond lead, heart to the ace, diamond to table (everyone following, assume) jack of hearts run losing to the queen on our left, and we are likely to get a spade back now. We win this in hand and cash a top heart, but don't know what to discard from the table....

[gnasher]

So diamond lead, heart to the ace, diamond to table (everyone following, assume) jack of hearts run losing to the queen on our left, and we are likely to get a spade back now. We win this in hand and cash a top heart, but don't know what to discard from the table....

An improvement is to cash all the diamonds before running ♥J, throwing two clubs and a heart. Then run ♥J, win the spade return, cash ♣AK, and play the remaing top heart, throwing ♣J unless it's good. That caters for RHO guarding hearts and LHO guarding clubs, though not the other way around.

[eyhung]

I can confirm than Ace, then Jack is the best way to play the suit in isolation. Definitely not obvious but Justin was able to work it out when I gave it to him. His reasoning was: cashing both AK is strong (loses only to QTxx(xx) in the same hand), and A then J is clearly superior -- you gain vs. xx offside and lose vs. Qx offside.

 

And yes, all will follow to the second round of diamonds, so you have some squeeze options. gnasher's line is pretty good -- I'm not sure if there's a better one...

[Lobowolf]

Computers always duck the Qx when you lead the 8.

[mtvesuvius]

I thought it would be best to lead the 8 toward dummy. Most players won't duck from the Queen. Assuming it's ducked, you duck in dummy, and then finesse the Jack if the 8 lost to the ten. This may not be right, but it seems to work most of the time.

[Lobowolf]

I thought it would be best to lead the 8 toward dummy. Most players won't duck from the Queen. Assuming it's ducked, you duck in dummy, and then finesse the Jack if the 8 lost to the ten. This may not be right, but it seems to work most of the time.

Against all but fairly good opponents, the chance of their either grabbing the queen or tanking long enough to give the position of the queen away (so you can hop up and succeed against Qxx on the left and Txx on the right) will usually give you some percentage points that aren't measured by simulation.

[mtvesuvius]

I thought it would be best to lead the 8 toward dummy. Most players won't duck from the Queen. Assuming it's ducked, you duck in dummy, and then finesse the Jack if the 8 lost to the ten. This may not be right, but it seems to work most of the time.

Against all but fairly good opponents, the chance of their either grabbing the queen or tanking long enough to give the position of the queen away (so you can hop up and succeed against Qxx on the left and Txx on the right) will usually give you some percentage points that aren't measured by simulation.

Yes, the psychological aspect also adds some merit to this line...

[qwery_hi]

I can confirm than Ace, then Jack is the best way to play the suit in isolation. Definitely not obvious but Justin was able to work it out when I gave it to him. His reasoning was: cashing both AK is strong (loses only to QTxx(xx) in the same hand), and A then J is clearly superior -- you gain vs. xx offside and lose vs. Qx offside.

 

And yes, all will follow to the second round of diamonds, so you have some squeeze options. gnasher's line is pretty good -- I'm not sure if there's a better one...

Can someone explain why Ace, then finesse the Jack is better than finesse the Jack, then AK, for winning 4 tricks? AFAICS they will win 4 tricks equal number of times.

[JLOL]

low to the jack fails to pick up stiff Q in either hand which is the difference.

[qwery_hi]

low to the jack fails to pick up stiff Q in either hand which is the difference.

So Ace first is better in two cases. But there are two cases when finessing Jack first is better, these are (void QT7543) and (7 QT543). So the two lines are equal.

[JLOL]

low to the jack fails to pick up stiff Q in either hand which is the difference.

So Ace first is better in two cases. But there are two cases when finessing Jack first is better, these are (void QT7543) and (7 QT543). So the two lines are equal.

My answer was assuming you meant low to the jack first and then AK (which picks up neither of the combos you described).

 

When you said finesse the jack did you mean run the jack and then play AK?

[JLOL]

Assuming you meant J first then AK, you are losing to Qx offside (2 combos) and stiff Q in both spots to pick up the 51 and 60 split you described.

 

This is why Jack first is a winner with AK987 opp Jx because you pick up all QTxxx's(4) onside, as well as a stiff Q but without the 7 you only pick up 1 QTxxx.

[qwery_hi]

Assuming you meant J first then AK, you are losing to Qx offside (2 combos) and stiff Q in both spots to pick up the 51 and 60 split you described.

 

This is why Jack first is a winner with AK987 opp Jx because you pick up all QTxxx's(4) onside, as well as a stiff Q but without the 7 you only pick up 1 QTxxx.

Sorry I wasn;t clear - I meant float J first and then AK.

 

With Qx offside, both Ace first and float Jack first get only 3 tricks, so its the same for both lines.

 

I think the only difference between --

 

1. Float J, then AK

2. Ace, then float J are

 

Stiff Q either hand where 2 is better, and

(void QT7543), (7 QT543) where 1 is better.

[JLOL]

Assuming you meant J first then AK, you are losing to Qx offside (2 combos) and stiff Q in both spots to pick up the 51 and 60 split you described.

 

This is why Jack first is a winner with AK987 opp Jx because you pick up all QTxxx's(4) onside, as well as a stiff Q but without the 7 you only pick up 1 QTxxx.

Sorry I wasn;t clear - I meant float J first and then AK.

 

With Qx offside, both Ace first and float Jack first get only 3 tricks, so its the same for both lines.

 

I think the only difference between --

 

1. Float J, then AK

2. Ace, then float J are

 

Stiff Q either hand where 2 is better, and

(void QT7543), (7 QT543) where 1 is better.

Yes you are right...soon Frances will get mad at me for staying up all night again I think :) This makes cashing the ace first slightly better.

[qwery_hi]

Assuming you meant J first then AK, you are losing to Qx offside (2 combos) and stiff Q in both spots to pick up the 51 and 60 split you described.

 

This is why Jack first is a winner with AK987 opp Jx because you pick up all QTxxx's(4) onside, as well as a stiff Q but without the 7 you only pick up 1 QTxxx.

Sorry I wasn;t clear - I meant float J first and then AK.

 

With Qx offside, both Ace first and float Jack first get only 3 tricks, so its the same for both lines.

 

I think the only difference between --

 

1. Float J, then AK

2. Ace, then float J are

 

Stiff Q either hand where 2 is better, and

(void QT7543), (7 QT543) where 1 is better.

Yes you are right...soon Frances will get mad at me for staying up all night again I think :) This makes cashing the ace first slightly better.

:) As long as you don't have to work today...

 

But if there are 2 cases where 1 is better and 2 where 2 is better, why is cashing the ace slightly better and not equal?

[JLOL]

An instance of a 60 split is less likely than an instance of a 51 split. It is always true that the more even the split, the more likely it is. A combo of a 22 break is more likely than a combo of a 31 break for instance. This is a rule with no exceptions so it's pretty handy.

 

This is the opposite of the rule that if there are an even number of cards missing in a suit then they rate to NOT be split (for instance a 31 break is more likely than a 22 break, or a 42 break is more likely than a 33 break), but this is only because in those cases either player can have the longer holding (so it is really 42 + 24 is ore likely than 33). In this case we are always talking about 1 side having the longer one since it's only one combo we're thinking of rather than 2.

 

I'm sure that the percentage difference between a specific 51 and a 60 is very small though since they're both unlikely to begin with.

[qwery_hi]

An instance of a 60 split is less likely than an instance of a 51 split. It is always true that the more even the split, the more likely it is. A combo of a 22 break is more likely than a combo of a 31 break for instance. This is a rule with no exceptions so it's pretty handy.

 

This is the opposite of the rule that if there are an even number of cards missing in a suit then they rate to NOT be split (for instance a 31 break is more likely than a 22 break, or a 42 break is more likely than a 33 break), but this is only because in those cases either player can have the longer holding (so it is really 42 + 24 is ore likely than 33). In this case we are always talking about 1 side having the longer one since it's only one combo we're thinking of rather than 2.

 

I'm sure that the percentage difference between a specific 51 and a 60 is very small though since they're both unlikely to begin with.

Great explanation - this had been bothering me for a long time. So just counting the number of instances isn't enough when deciding which line is better - this makes it more tough to find the right line.