[paulsim]

Hi all,

I would like some advice about Splinter Slam Bidding


I.- Responder's strength:

a.- 11-14 HCP
b.- 12-14 HCP
c.- 9-11 HCP

(I saw this 9-11 range a lot, but seems awkward to move to Slam if Opener is not strong enough

II.- When opener doesn't sign off:

1S...…..4C
4D

Opener says: no wasted values in clubs, or maybe better said: still an opening hand, not counting wasted points
Ace maybe worth half strengh and KQJ worth almost 0

III.- ¿Slam needs more or less 28HCP if singleton and 26HCP if void?

So far, so good?


IV.- When opener shows life: do we still need for find 28HCP or just fly around and see what happens?

1S...…...4C
4D

Opener shows diamond control and a living hand

Responder should:

a.- Control bid in hearts?
b.- Or 4H is last train? So, it would show a minimun hand: 11-12HCP?
c.- 4S would mean lack of control or a minum hand?

So, in other words:

Opener's first duty is to show if his hand is not damaged by the short suit
That is made with a control bid.
Going back to the trump suit is sign off

Responder's firs duty when opener accepts the Splinter is:
¿Show a minimun-maximun hand or show controls up the line?

How controls up the line affect "Last Train"?


Thanks all
Kind Regards,
Paul_S

[DavidKok]

A lot of good questions. Here are my thoughts on the topic, in no particular order:

• Strength requirements: I think the biggest difference here is whether people count HCP or playing strength. Confusingly, both are commonly called 'points'. It is not surprising that having 4-card support for partner's long suit and a side singleton or void is worth several upgrades. Make sure you clarify what you are thinking of when agreeing on a range for a splinter. The most common treatment, and implied in all discussions of splinters but I'll be the fool who writes it out explicitly, is split range. Both a weak range, intending to leave all future decisions to partner, and a strong range, so strong that you will bid on over a signoff and feel confident going to the 5-level. The second hand type is very rare, and people often do not discuss this second range at all, but technically most partnerships do play split range (even if they haven't discussed it). Now personally I play Maas 2NT, which has a good 9-bad 14 range. I prefer for my splinters to have a similar range, that way my Maas is no-splinter (or, more precisely, all hands contained in it with shortage have 3-card support). This gives me a lower splinter range of 9-12 (most splinters are 'good' hands, so the 13-14 doesn't come up). The strong variant, willing to commit to the 5-level, is approximately 17+. This is in HCP - with adjustments these hands would be markedly stronger.
  
• Shape requirements: I have three things to say about this. 1) It is somewhat popular to distinguish voids from singletons. The idea is that the void 'is magical', i.e. has a really strong trick-taking potential. Even on a trump lead, a side void and four card support in dummy means you can expect to ruff three cards or so in that suit if you have entries to your hand. Contrast this with a singleton, where not only can you not ruff the first round, having to possibly lose the lead also means the opponents can draw trumps twice. Personally I do not play this style, but if you do, some people use e.g. 1♥-3♠ and 1♠-3NT as a singleton splinter (with the next step asking the strain) and direct splinter bids (and 1♥-3NT as a substitute spade splinter) as voids. Alternatively, some lower route, e.g. Jacoby 2NT, can include one type of splinters while the direct route shows the other. 2) There is some debate on whether all hands within the range with 4(+)-card support and a suitable singleton or void should bid a splinter. Personally I think there is merit in requiring that the hand is three-suited, i.e. has length and values in all three non-short suits. This covers most of the splinter hands and allows for better hand evaluation. Plus, the hands that fail to qualify for this have a 2/1 auction, as these hands are two-suiters (one long side suit, 4(+)-card support, two shorter suits). This idea is seen e.g. on competitive auctions as the distinction between a splinter and a fitbid, but also on e.g. 1m-1M; 4m showing 6(+)m4M - and therefore, by necessity, an outside singleton or void. 3) Keep in mind that splintering with a singleton ace, king or even queen can be a poor idea. The whole purpose of a splinter is that we simplify partner's hand evaluation for slam purposes, and if partner's holding in our short suit turns out to be a valuable source of tricks (especially if partner has the king, which is normally low value opposite a splinter) that can backfire. Depending on how strict you are with this, you may have to misdescribe your shape on hands where you have a singleton top honour.
  
• Point requirements for slam after a splinter: the purpose of a splinter is to allow for nuanced hand evaluation, so points aren't that great a tool on these auctions. Nevertheless, as a rule of thumb we have the theories of the '30 point deck' and the '34 point deck' - these apply when partner has a void and singleton respectively. The idea is that only the ace, or no high cards at all, have trick-taking potential in partner's short suit. To find rules of thumb for how many working points (i.e. points in other suits) you then need for slam, we take the usual 32/40 = 80% for slam and scale it down to these decks. So 80%*30 = 24 HCP opposite a void, and 80%*34 ~= 27 HCP opposite a singleton. Keep in mind these are coarse estimates, and instead you should develop personal hand evaluation to assist on these auctions. You can apply the same trick to find grand slam coarse estimates.
  
• Control bidding and Last Train: I do not treat these splinter auctions different from other control auctions. Our main priority is to show controls, with Last Train taking up the highest bid under 4M to ask if partner has extras (though it may be an advance cue). Keep in mind that, more than on most other auctions, opener has captaincy, i.e. is much better informed of responder's hand than vice versa. This means that e.g opener should almost always be the one to make asking bids such as Blackwood (if any), and often opener will have to make the decision on whether or not to go on to slam. We will still have a dialogue auction after a splinter through control bids and Last Train, but the information shared is very skewed and the rest of the auction will relfect that. In particular, on your example
• 1♠-4♣; 4♦: Slam interest opposite a club splinter (otherwise sign off), diamond control. Responder can now bid 4♠ (signoff, no heart control) or 4♥ (Last Train, either a heart control or a maximum for the bid). The 'maximum for the bid' option is rare but typically also implies a heart control, by the three-suited nature of my splinters. Respnder will rarely bid past 4♠ here, as opener is in a better position to bid Blackwood (though with a nice maximum responder may bid 4♥-and-bid-on-over-a-signoff).

[mw64ahw]

The simplistic/systematic approach in my bidding tool.
Splinter criteria?

a) assume your 9-14 hcp
b) 7/6.5 mod. losers
c) sufficient controls/keycards, you're probably looking at something like A,K,Qtrumps for a 9hcp hand in addition to the splinter.

Now opener can
a) sign-off with worse than 6.5 mod. losers or an hcp count (adjusted) that is too low.
b) sign-off with poor controls
c) bid a control
d) initiate a key card sequence (showing for me to allow for further control checking)

Now responder can
a) bid a control (defer to opener)
b) sign-off min. or missing a skipped control
c) initiate a key card sequence max.

Whether you use 'last train' or not is up to agreement. (I dont)

One other consideration is whether you need to use a singleton splinter via this route rather than another and reserve the jumps for void splinters. This is the approach I've picked up from others.

Alternatively, as David I think points out, just bid 1st/2nd round controls rather than limiting to splinters.

And then when you bid the slam you have to get the play correct if it's makeable or an advantageous lead if not.

[mike777]

paulsim, on 2025-February-13, 03:53, said:

Hi all,

I would like some advice about Splinter Slam Bidding


I.- Responder's strength:

a.- 11-14 HCP
b.- 12-14 HCP
c.- 9-11 HCP

(I saw this 9-11 range a lot, but seems awkward to move to Slam if Opener is not strong enough

II.- When opener doesn't sign off:

1S...…..4C
4D

Opener says: no wasted values in clubs, or maybe better said: still an opening hand, not counting wasted points
Ace maybe worth half strengh and KQJ worth almost 0

III.- ¿Slam needs more or less 28HCP if singleton and 26HCP if void?

So far, so good?


IV.- When opener shows life: do we still need for find 28HCP or just fly around and see what happens?

1S...…...4C
4D

Opener shows diamond control and a living hand

Responder should:

a.- Control bid in hearts?
b.- Or 4H is last train? So, it would show a minimun hand: 11-12HCP?
c.- 4S would mean lack of control or a minum hand?

So, in other words:

Opener's first duty is to show if his hand is not damaged by the short suit
That is made with a control bid.
Going back to the trump suit is sign off

Responder's firs duty when opener accepts the Splinter is:
¿Show a minimun-maximun hand or show controls up the line?

How controls up the line affect "Last Train"?


Thanks all
Kind Regards,
Paul_S

all good posts above.

I guess first agree to play limited or unlimited strength splinters, surprisingly this is still hotly debated in 2025. I strongly encourage limited splinters.
Really simply style, Bergen limits it to 12 HCP.
Bit more advanced, different bids for stiffs and voids.

after you decide the above, you can get more fancy, but that is a solid start...good luck

[akwoo]

The first thing you need to understand about slam bidding is the following:

Experts do NOT evaluate their hands for slam based on points (of any sort)

I'm not exactly an expert, but when I hear partner splinter, I look at my hand, and I see whether there is a possible, well-fitting, but "normal-looking for a splinter" minimum-in-hcp hand for partner for which slam makes.

Let's say it goes 1H-4C, and we're playing that minimum for a splinter is about 11 hcp.

If I have x AQxxx KQx xxxx, I think "Partner could have Axxx Kxxx Axxx x which would be about the best 11 count I could hope for" - hmm even that's not quite good enough if opps lead a trump - so I sign off.

But if I have x AQxxx KQJx xxxx or x AQxxxx KQx xxx, both of which make slam opposite the example above, I would bid RKCB - what do I need to know other than whether or not partner has 3 key cards (or 2 and a void)?

If I have something like Ax Kxxxxx AQx xx - now I bid 4♦ because I make slam opposite Kxx Axxx Kxxxx x but I need to know information other than keycards.

(I think the usual agreements are that 4M says that there is almost no hope of slam, and any other suit bid shows a control in that suit with slam hopes.)

[mycroft]

David's response is as usual thorough and complete, so I'm just going to go around the edges and explicitly point out what he might have just implied.

Likely the reason for the different ranges is that some are counting straight HCP and some are "HCP + shape", or "playing values". They're similar hands, though.

The reason we want a tight range (or ranges, if you're thinking about David's "min GF or near slam force" split range) is that the splinter bid takes up so much room that could be used for control cuebidding or investigating double fits or any of the other tools that will help partner feel out their interest in slam. That is also why the shape requirements are so stringent.

You are painting a picture of your hand in one call.

A common aphorism is "when you have described your strength and shape tightly, partner is captain. After doing so, your job is to answer questions and pass when partner tells you to." As a result, a lot of your questions about "what does opener show, how do I understand..." go away into "if partner asks questions, you answer."

If partner bids game, you pass (if you have the normal range). If partner cuebids, he's interested in slam, continue if you can. If partner keycards, answer.

AKWoo talks well about how opener thinks about their hand. Again, place cards. Partner has made a very specific call, you should be able to put the right cards in place (or the "hoped for" cards in place), and know what else you need to know about (if anything).