[shugart24]

I'm working with my son on opening leads. When leading partner's suit, regardless of whether defending NT or a suit, our algorithm is as follows:

a) If you have the Ace, and not the King lead it
b) Lead the King when you also hold the Ace or Queen and look for Partner's signal
c) If you have touching Honors, lead the top one
d) If you have a doubleton, lead the top one
e) If you have 3 or more headed by the 9 or better, lead the 3rd and partner will apply rule of 12
f) If you have xxx(x), lead the top one if you have supported partner's suit
g) If you have xxx(x), give count if you have not supported partners suit
h) if you have a really bad hand with little chance of getting in again, and hold Hxx(x), lead H

It's clear that when you make an opening lead of partner's suit, there will be times when the partner cannot 'read' what your lead means initially. For example, a lead of a Queen might be from Q or Qx or QJ(x).

If you hold Q 8 5, the algorithm says lead the 5 regardless of whether or not you supported partner's suit and partner will have to figure out what the '5' meant.

Does this algorithm make sense, or should it be tweaked?

[bluenikki]

shugart24, on 2025-February-18, 05:40, said:

I'm working with my son on opening leads. When leading partner's suit, regardless of whether defending NT or a suit, our algorithm is as follows:

a) If you have the Ace, and not the King lead it
b) Lead the King when you also hold the Ace or Queen and look for Partner's signal
c) If you have touching Honors, lead the top one
d) If you have a doubleton, lead the top one
e) If you have 3 or more headed by the 9 or better, lead the 3rd and partner will apply rule of 12
f) If you have xxx(x), lead the top one if you have supported partner's suit
g) If you have xxx(x), give count if you have not supported partners suit
h) if you have a really bad hand with little chance of getting in again, and hold Hxx(x), lead H

It's clear that when you make an opening lead of partner's suit, there will be times when the partner cannot 'read' what your lead means initially. For example, a lead of a Queen might be from Q or Qx or QJ(x).

If you hold Q 8 5, the algorithm says lead the 5 regardless of whether or not you supported partner's suit and partner will have to figure out what the '5' meant.

Does this algorithm make sense, or should it be tweaked?

There isn't really anything you can do about partner's ability to read your spot card.

[shugart24]

bluenikki, on 2025-February-18, 06:42, said:

There isn't really anything you can do about partner's ability to read your spot card.

yes, until dummy comes down and he can see his and dummy's cards and narrow it down. Does the algorithm make sense? I am mostly wondering if I have, say Q 6 3 should I lead the 3 regardless of whether or not I supported his suit?

An alternative might be " If I supported his suit, make an attitude lead and if I didn't support his suit give a count lead "

[Stephen Tu]

shugart24, on 2025-February-18, 05:40, said:

a) If you have the Ace, and not the King lead it
b) Lead the King when you also hold the Ace or Queen and look for Partner's signal
c) If you have touching Honors, lead the top one
d) If you have a doubleton, lead the top one
e) If you have 3 or more headed by the 9 or better, lead the 3rd and partner will apply rule of 12
f) If you have xxx(x), lead the top one if you have supported partner's suit
g) If you have xxx(x), give count if you have not supported partners suit
h) if you have a really bad hand with little chance of getting in again, and hold Hxx(x), lead H

Does this algorithm make sense, or should it be tweaked?

This is mostly reasonably sensible standard practice, except for the following tweaks:

a) vs NT, holding the ace without the K, and 3+ cds, you should usually underlead. This accomplishes a few things, like sometimes preventing declarer from being able to hold up, or sometimes preventing declarer's say QJxx from becoming a double stopper. Vs suit, of course one should often consider whether leading the suit at all looks best or not.

e) I think most use "T or better" rather than 9 or better, still leading high from 9xx if you have supported as in f). Lead count as per agreements otherwise. Also holding an honor you don't always lead 3rd best, you lead your agreed count card. For example if you are playing "3rd from even, low from odd" leads, you'd lead 3rd from Hxx or Hxxx, because in the first it's the lowest from odd, and in the second it's 3rd from even, but from Hxxxx you'd lead the lowest, not 3rd. But you'd lead 4th from the latter two if you agreement is 4th best.

h) is usually reserved for defending high level contracts when you have a very big trump fit where you suspect opponent often only has singleton + you want to find switch at trick 2. Otherwise you lead normally because you still want to keep honor over declarer's potential honor(s) in the suit, as this sometimes decreases their trick taking potential. Also you don't do this too often at low levels when haven't supported if partner might mistakenly play you for Hx doubleton and try to give a non-existent ruff or give up a ruff-sluff not realizing you had more cards than usual.

[DavidKok]

I don't play this algorithm, but think it's sensible enough. However, there is an experience I'd like to share.

I've recently played some very sophisticated leads and carding methods with a few partners. We played different leads against NT and trump contracts, different rules at the 5-level and up compared to 4-level and below, different agreements when partner had bid the suit, different agreements whether we had raised or not, different signals trick one compared to tricks two through twelve, different agreements for switching to an unbid suit through declarer and through dummy, different signals depending on whether dummy had one, two or three cards in the suit led, different discards versus NT and versus trumps, and more.
It was horrible. My partner and I made around 8-10 mistakes each according to the system per evening. Almost all of them did not impact our score - this was so much sophistry, and the crucial information on each deal was usually public after the first signal. Each of these agreements was in theory better than the converse, but the situations where they would benefit never came up¹, while the confusion cost around once or twice an evening. What's more, even when it did not cost the headache of trying to keep it all straight made us perform worse on other deals.

Around the same time I started seriously looking into the theoretical merits of different signalling systems. Why play one set of agreements over another? Historically, which agreements have been winners? What holdings are relevant where system A fails but system B prevails? My search was inconclusive but what little I found was depressing - pretty much all of the evidence for some carding agreements over others was anecdotal, and not supported by either theoretical analysis or play results. For this reason I have moved towards simpler carding agreements - my current goal is to give partner two or three bits of information (in the sense of computer science) per deal reliably and accurately, and then hope that it is enough to solve all questions they might have. I prioritise clarity of the communication channels over theoretical optimality - in no small part because I couldn't find sufficient evidence to support one agreement over another anyway.

You cite an 8-step algorithm to decide on a lead, having committed to a particular suit. I'd consider simplifying that.

[Stephen Tu]

Quote

You cite an 8-step algorithm to decide on a lead, having committed to a particular suit. I'd consider simplifying that

I disagree. I don't think his algorithm is all that complicated, it just explicitly lists out things that experienced players just have absorbed and do automatically with little thought, but beginners/ints need all of this spelled out in excruciating detail until ingrained.

It's basically, if you follow my recs:
- lead high from honor sequences (except K from AK), and interior sequences
- lead count card from honor non sequence, or from no honors if haven't supported
- lead top of nothing if supported with no honor
- don't underlead ace at suit contracts, but do underlead vs NT, vs suits consider leading different suit entirely if holding ace.
- consider leading honor with big trump fit at high level to retain lead

These are all reasonable things to do and experienced players easily follow all of these rules. OP just was kind of more verbose and detailed and broke down some of these things into separate rules, neither is really more simple under examination.

[DavidKok]

That phrasing helps quite a lot, but is still more complicated than what I was thinking of. I don't follow all of those rules and prefer other lead agreements, which I think makes me inexperienced.

[shugart24]

Stephen Tu, on 2025-February-18, 12:20, said:

I disagree. I don't think his algorithm is all that complicated, it just explicitly lists out things that experienced players just have absorbed and do automatically with little thought, but beginners/ints need all of this spelled out in excruciating detail until ingrained.

It's basically, if you follow my recs:
- lead high from honor sequences (except K from AK), and interior sequences
- lead count card from honor non sequence, or from no honors if haven't supported
- lead top of nothing if supported with no honor
- don't underlead ace at suit contracts, but do underlead vs NT, vs suits consider leading different suit entirely if holding ace.
- consider leading honor with big trump fit at high level to retain lead

These are all reasonable things to do and experienced players easily follow all of these rules. OP just was kind of more verbose and detailed and broke down some of these things into separate rules, neither is really more simple under examination.

Thanks for the comment/suggestions and thanks David. I hear what David is saying. This algorithm is strictly when leading partner's suit on opening lead if that's the suit you have chosen to lead. We have different algorthims for blind NT and blind suit defense (Journalist leads)